Cluster Data Ontap 8.3 Data Protection Student Guide 4s6m69

-count [-prefix ] The Comment Use the volume snapshot modify command to change the text comment that is associated with a Snapshot copy. Example: Cluster > volume snapshot modify -vserver vs0 -volume vol3 -snapshot vol3_snapshot -comment "Pre-upgrade snapshot"

2-14

Data ONTAP Data Protection istration: Data Protection with NetApp Snapshot Copies

© 2015 NetApp, Inc. This material is intended only for training. Reproduction is not authorized.

NetApp University - Do Not Distribute

The Label The SnapMirror label is used by the vaulting subsystem when you back up Snapshot copies to the vault destination. If an empty label ("") is specified, the existing label is deleted.

2-15




The Snapshot Policy Snapshot policies automatically manage Snapshot copy schedules and retention on FlexVol volumes. Snapshot Policy Job Schedule

SVM

Cluster 15

FlexVol Volume

© 2015 NetApp, Inc. All rights reserved. NetApp Proprietary – Limited Use Only

THE SNAPSHOT POLICY A Snapshot policy allows you to configure the frequency and maximum number of Snapshot copies that are created automatically.   

You can create new Snapshot polices as necessary. You can apply one or more schedules to the Snapshot policy. The Snapshot policy can also have zero schedules.

When you create a storage virtual machine (or SVM, formerly called Vserver), you can specify a Snapshot policy that becomes the default for all FlexVol volumes that are created for that SVM. When you create a FlexVol volume, you can specify which Snapshot policy you want to use, or you can allow the FlexVol to inherit the SVM Snapshot policy. The default Snapshot policy might meet your needs. The default Snapshot copy policy is useful if s rarely lose files. The default Snapshot policy is as follows:   

Weekly schedule to keep two weekly Snapshot copies Daily schedule to keep two daily Snapshot copies Hourly schedule to keep six hourly Snapshot copies

However, if s commonly lose files, then you should adjust the default policy to keep Snapshot copies longer. The recommended Snapshot copy policy for this situation:   

2-16

Weekly schedule to keep two weekly Snapshot copies Daily schedule to keep six daily Snapshot copies Hourly schedule to keep eight hourly Snapshot copies




For typical systems, only five to ten percent of the data changes each week: Six daily and two weekly Snapshot copies consume 10 to 20 percent of disk space. Adjust the Snapshot copy reserve for this amount of disk space for Snapshot copies. Each volume on an SVM can use a different Snapshot copy policy. For very active volumes, create a Snapshot schedule that creates Snapshot copies every hour and keeps them for just a few hours, or turn off the Snapshot copy feature.

2-17




Lesson 3

Implementing a Snapshot Copy Solution

16


LESSON 3: IMPLEMENTING A SNAPSHOT COPY SOLUTION

2-18




Typical Workflow  Create a job schedule, or use the default.  Create a Snapshot policy and specify the job schedule.  Assign the Snapshot policy to an SVM.  Create a FlexVol volume and specify the Snapshot policy, or inherit a Snapshot policy from the SVM.

17


TYPICAL WORKFLOW To implement a Snapshot copy solution, first, decide on a job schedule.  

Use a pre-existing schedule or create a new schedule, as required. You can create multiple schedules that create and delete Snapshot copies.

A Snapshot policy is created that automatically manages Snapshot copy schedules and retention on FlexVol volumes or infinite volumes. You can apply the Snapshot policy to the cluster or to an SVM. Unless you specify a Snapshot policy when you create a FlexVol volume or an infinite volume, a volume inherits the Snapshot policy that is associated with its containing SVM. The default Snapshot policy for an SVM with a FlexVol volume is named default, and the default Snapshot policy for an SVM with infinite volume is named default-1weekly.

2-19




Create a New Job Schedule  Check to see if an existing job schedule fits your needs: cluster1::> job schedule show

 Create a new job schedule: cluster1::> job schedule cron create -name 2HourSnapshot -month all -dayofweek all -day all -hour 0,2,4,6,8,10,12,14,16,18,20,22 -minute 5

 that the job was created: cluster1::> job schedule cron show 18


CREATE A NEW JOB SCHEDULE Find out which job schedules exist. If a job schedule meets your needs, then use it. If not, then create a new job schedule. cluster1::> job schedule show Name

Type

Description

-----------

---------

-----------------------------------------

5min

cron @:00,:05,:10,:15,:20,:25,:30,:35,:40,:45,:50,:55

8hour

cron

@2:15,10:15,18:15

Auto Balance Aggregate Scheduler interval

Every 1h

RepositoryBalanceMonitorJobSchedule interval

Every 10m

daily

cron

@0:10

hourly

cron

@:05

weekly

cron

Sun@0:15

7 entries were displayed.

2-20




If you need to create a new job schedule, run the job schedule cron create command: cluster1::> job schedule cron create Usage: [-name]

Name

[ -month , ... ]

Month

[ -dayofweek , ... ]

Day of Week

[ -day , ... ]

Day

[ -hour , ... ]

Hour

[-minute] , ...

Minute

Example: Create a new job that runs every month, every day of the week, every two hours, at five minutes past the hour: cluster1::> job schedule cron create -name 2HourSnapshot -month all dayofweek all -day all -hour 0,2,4,6,8,10,12,14,16,18,20,22 -minute 5 cluster1::> job schedule cron show Name

Description

----------------

---------------------------------------------------

2HourSnapshot @0:05,2:05,4:05,6:05,8:05,10:05,12:05,14:05,16:05,18:05,20:05,22:05 5min

@:00,:05,:10,:15,:20,:25,:30,:35,:40,:45,:50,:55

8hour

@2:15,10:15,18:15

daily

@0:10

hourly

@:05

weekly

Sun@0:15

6 entries were displayed. You will use the 2HourSnapshot job when you create the Snapshot policy.

2-21




Create a New Snapshot Policy  Check the current volume Snapshot policies: cluster1::> volume snapshot policy show

 Create a new volume Snapshot policy: cluster1::> volume snapshot policy create -vserver svm1 -policy 2Hour_Policy -enabled true -comment "Takes a Snapshot Copy Every 2 Hours" -schedule 2HourSnapshot -count 84 Keep 84 Snapshot copies.

 that the policy was created:

cluster1::> volume snapshot policy show -policy 2Hour_Policy

19


CREATE A NEW SNAPSHOT POLICY cluster1::> volume snapshot policy show Vserver: cluster1 Number of

Is

Policy Name

Schedules

Enabled

Comment

------------------------

---------

-------

------------------------

true

Default policy with

default 3 hourly, daily & weekly schedules. Schedule

Count

Prefix

SnapMirror Label

---------------------- -----

--------------------

--------

hourly

6

hourly

-

daily

2

daily

daily

weekly

2

weekly

weekly

default-1weekly

3

true

Default policy with 6 hourly, 2 daily & 1 weekly schedule.

2-22




Schedule

Count

Prefix

SnapMirror Label

---------------------- -----

--------------------

--------

hourly

6

hourly

-

daily

2

daily

-

weekly

1

weekly

-

none

0

false

Policy for no automatic snapshots.

Schedule

Count

Prefix

SnapMirror Label

----------------

-----

-------------

----------------

-

-

-

-

3 entries were displayed. Example: Create a volume Snapshot policy for vserver svm1 that uses the 2HourSnapshot job that you created in the previous step. You keep one week of Snapshot copies, for a total of 84 copies: cluster1::> volume snapshot policy create -vserver svm1 -policy 2Hour_Policy -enabled true -comment "Takes a Snapshot Copy Every 2 Hours" -schedule 2HourSnapshot -count 84 cluster1::> volume snapshot policy show Vserver: cluster1 Number of

Is

Policy Name

Schedules

Enabled

Comment

------------------------

---------

-------

------------------------

true

Default policy with

default 3 hourly, daily & weekly schedules. Schedule

Count

Prefix

SnapMirror Label

---------------------- -----

---------------------- --------

hourly

6

hourly

-

daily

2

daily

daily

weekly

2

weekly

weekly default-1weekly 3 6 hourly, 2 daily & 1 weekly schedule.

2-23

true

Default policy with

Schedule

Count

Prefix

SnapMirror Label

--------------

-----

----------------------

--------

hourly

6

hourly

-

daily

2

daily

-

weekly

1

weekly

-




none

0

false

Policy for no automatic snapshots.

Schedule

Count

Prefix

SnapMirror Label

--------------

-----

----------------------

--------

-

-

-

-

Vserver: svm1 Number of

Is

Policy Name

Schedules

Enabled

Comment

---------------------------------

---------

-------

------------------------

true

Takes a Snapshot Copy

2Hour_Policy Every 2 Hours

1

Schedule

Count

Prefix

--------------

-----

----------------------

SnapMirror Label --------

84

2HourSnapshot

-

2HourSnapshot

4 entries were displayed. Another method to check the volume Snapshot policy: cluster1::> volume snapshot policy show -policy 2Hour_Policy Vserver: svm1 Number of

Is

Policy Name

Schedules

Enabled

Comment

------------------------

---------

-------

------------------------

true

Takes a Snapshot Copy

2Hour_Policy Every 2 Hours

2-24

1

Schedule

Count

Prefix

SnapMirror Label

-------------2HourSnapshot

----84

---------------------2HourSnapshot

--------




Assign the Snapshot Policy to an SVM  For an existing SVM, check to see which policy is already assigned: cluster1::> vserver show -vserver svm1

 Modify an SVM to use an existing Snapshot policy: cluster1::> vserver modify -vserver svm1 -snapshot-policy 2Hour_Policy

 that the SVM configuration is using a new Snapshot policy: cluster1::> vserver show -vserver svm1

20


ASSIGN THE SNAPSHOT POLICY TO AN SVM First, check the current Snapshot policy for SVM svm1: cluster1::> vserver show -vserver svm1 Vserver: svm1 Vserver Type: data Vserver Subtype: default Vserver UUID: cd645ac3-425d-11e4-9d2d0050560140b9 Root Volume: svm1_root Aggregate: n1_n1_aggr1 NIS Domain: Root Volume Security Style: ntfs LDAP Client: Default Volume Language Code: C.UTF-8 Snapshot Policy: default  The current Snapshot policy Comment: Quota Policy: default List of Aggregates Assigned: 2-25




Limit on Maximum Number of Volumes allowed: unlimited Vserver State: running Vserver Operational State: running Vserver Operational State Stopped Reason: Allowed Protocols: nfs, cifs, f, iscsi, ndmp Disallowed Protocols: Is Vserver with Infinite Volume: false QoS Policy Group: Config Lock: false IPspace Name: Default Change the current Snapshot policy so that the SVM uses the Snapshot policy that you created in the previous step: cluster1::> vserver modify -vserver svm1 -snapshot-policy 2Hour_Policy Check your work: cluster1::> vserver show -vserver svm1 Vserver: svm1 Vserver Type: data Vserver Subtype: default Vserver UUID: cd645ac3-425d-11e4-9d2d0050560140b9 Root Volume: svm1_root Aggregate: n1_n1_aggr1 NIS Domain: Root Volume Security Style: ntfs LDAP Client: Default Volume Language Code: C.UTF-8 Snapshot Policy: 2Hour_Policy  New Snapshot policy Comment: Quota Policy: default List of Aggregates Assigned: Limit on Maximum Number of Volumes allowed: unlimited Vserver State: running Vserver Operational State: running Vserver Operational State Stopped Reason: Allowed Protocols: nfs, cifs, f, iscsi, ndmp

2-26




Disallowed Protocols: Is Vserver with Infinite Volume: false QoS Policy Group: Config Lock: false IPspace Name: Default

2-27




Create a New FlexVol Volume  Create a new FlexVol volume that inherits the Snapshot policy from the SVM: cluster1::> volume create -vserver svm1 -volume svm1_vol2 -aggregate n1_aggr1 -size 200M [Job 79] Job succeeded: Successful

 Confirm the Snapshot policy: cluster1::> volume show -vserver svm1 -volume svm1_vol2 -fields Snapshot-Policy vserver volume

snapshot-policy

------- --------- ---------------

svm1 21

The policy is inherited from the SVM.

svm1_vol2 2Hour_Policy


CREATE A NEW FLEXVOL VOLUME Create a new FlexVol volume without specifying the Snapshot policy. The new volume is expected to inherit the Snapshot policy from the current setting on the SVM. Create a new FlexVol volume: cluster1::> volume create -vserver svm1 -volume svm1_vol2 -aggregate n1_aggr1 -size 20M Observe the results of the volume create command: cluster1::> volume show Vserver

Volume

-------- ---------

Aggregate

State

Type

Size Available

Used%

---------- --------- ---- -------- ---------- -----

cluster1-01 vol0

aggr0_n1

online

RW

5.16GB

4.29GB

16%

vol0

aggr0_n2

online

RW

5.16GB

4.32GB

16%

svm1

svm1_root

n1_aggr1

online

RW

20MB

18.85MB

5%

svm1

svm1_vol1

n1_aggr1

online

RW

50MB

47.36MB

5%

svm1

svm1_vol2

n1_aggr1

online

RW

20MB

18.88MB

5%

cluster1-02


2-28




Confirm the Snapshot policy on the new volume: cluster1::> volume show -vserver svm1 -volume svm1_vol2 -fields SnapshotPolicy vserver

volume

snapshot-policy

-------

--------------

---------------

svm1

svm1_vol2

2Hour_Policy

2-29




Check the Volume Snapshot Activity 1 of 2

After some time elapses, check the Snapshot activity in the volume: cluster1::> volume snapshot show -vserver svm1 –volume svm1_vol2

NOTE: The job schedule is set to run every two hours at five minutes past the hour.

22


CHECK THE VOLUME SNAPSHOT ACTIVITY: 1 OF 2 After some time elapses, check the volume Snapshot activity: cluster1::>volume snapshot show -vserver svm1 -volume svm1_vol2 ---Blocks--Vserver

Volume

-------- -------svm1

Snapshot

------------------------------- ------ ------ -----

svm1_vol2 2HourSnapshot.2014-07-28_2005

60KB

0%

31%

2HourSnapshot.2014-07-28_2205

40KB

0%

23%


2-30

Size Total% Used%




Check the Volume Snapshot Activity 2 of 2

cluster1::>volume snapshot show -vserver svm1 -volume svm1_vol2 ---Blocks--Vserver svm1

Volume

Snapshot

Size Total% Used%

svm1_vol2 2HourSnapshot.2014-07-28_2005 60KB 0% 31%

The Snapshot copy takes the schedule name because the prefix option was not used.

23


CHECK THE VOLUME SNAPSHOT ACTIVITY: 2 OF 2

2-31




Answer this Question

What can a Snapshot policy be applied to? a) a storage virtual machine b) a FlexVol volume c) the cluster

d) all of the above

24


ANSWER THIS QUESTION

2-32




Lesson 4

Monitoring and Managing Snapshot Copies

25


LESSON 4: MONITORING AND MANAGING SNAPSHOT COPIES

2-33




Managing Snapshot Copy Policies and Schedules Snapshot policies automatically manage Snapshot copy schedules and retention on FlexVol volumes or infinite volumes

26


MANAGING SNAPSHOT COPY POLICIES AND SCHEDULES Snapshot policies automatically manage Snapshot copy schedules and retention on FlexVol volumes or infinite volumes. A new volume inherits the Snapshot policy that is associated with its containing SVM, unless you specify a Snapshot policy when you create the FlexVol volume. In some circumstances, you might need to manually manage the Snapshot policies and schedules. The volume snapshot policy command directory contains commands that you use to manage Snapshot copy policies. cluster1::> volume snapshot policy ? add-schedule

Add a schedule to snapshot policy

create

Create a new snapshot policy

delete

Delete a snapshot policy

modify

Modify a snapshot policy

modify-schedule

Modify a schedule within snapshot policy

remove-schedule

Remove a schedule from snapshot policy

show

Show snapshot policies

2-34




Managing Snapshot Copy Policies and Schedules (Cont.) Three preconfigured policies exist. These policies are d with the cluster.

27

Policy Name

Schedule

Count

Default

Hourly Daily Weekly

6 2 2

Default-1 weekly

Hourly Daily Weekly

6 2 1

None

N/A

N/A


MANAGING SNAPSHOT COPY POLICIES AND SCHEDULES (CONT.)

2-35




Managing Snapshot Copy Policies and Schedules (Cont.)  Cron job schedules are schedules that run at a specific time: 5min 8hour daily hourly weekly

@:00,:05,:10,:15,:20,:25,:30,:35,:40,:45,:50,:55 @2:15,10:15,18:15 @0:10 @:05 Sun@0:15

 Each Snapshot policy contains at least one schedule.

28


MANAGING SNAPSHOT COPY POLICIES AND SCHEDULES (CONT.) The job schedule cron command directory contains commands that you use to manage schedules that can be used for Snapshot policies. cluster1::> job schedule cron show Name

Description

----------------

---------------------------------------------------

5min

@:00,:05,:10,:15,:20,:25,:30,:35,:40,:45,:50,:55

8hour

@2:15,10:15,18:15

daily

@0:10

hourly

@:05

weekly

Sun@0:15


2-36




Manage Snapshot Copies Snapshot copy management commands: cluster1::> volume snapshot ? autodelete> Manage snapshot autodelete settings modify Modify autodelete settings show Display autodelete settings create Create a snapshot delete Delete a snapshot modify Modify snapshot attributes partial-restore-file Restore part of a file from a snapshot policy> Manage snapshot policies rename Rename a snapshot restore-file Restore a file from a snapshot show Display a list of snapshot copies

29


MANAGE SNAPSHOT COPIES The volume snapshot command directory contains commands for managing Snapshot copies.

2-37




Obtain Information About Snapshot Copies  Obtain general information about all Snapshot copies: cluster1::> volume snapshot show

 Obtain granular information about a particular Snapshot copy: cluster1::> volume snapshot show -volume svm1_vol1 -snapshot hourly.2014-10-01_0605 -instance Displays detailed information about all fields

30


OBTAIN INFORMATION ABOUT SNAPSHOT COPIES The volume snapshot show command displays information about Snapshot copies. The command output depends on the parameters that are specified with the command. If no parameters are specified, the command displays a table with the following information about all the available Snapshot copies:      

Vserver name Volume name Snapshot copy name Creation time Size Blocks: – –

2-38

Percentage of total blocks in the parent volume Number of used blocks in the parent volume




Example: cluster1::> volume snapshot show ---Blocks--Vserver

Volume Size

-------- -----svm1 svm1_vol2

Snapshot Total% Used% ------

State

----------------

--------

0%

2HourSnapshot.2014-07-10_1005 36%

valid

64KB

0%

2HourSnapshot.2014-07-10_1205 36%

valid

64KB

0%

2HourSnapshot.2014-07-10_1405 36%

valid

64KB

2HourSnapshot.2014-07-10_1605 34%

valid

60KB

0%

svm1_root 0%

hourly.2014-07-10_1105 33%

valid

64KB

0%

hourly.2014-07-10_1205 33%

valid

64KB

---Blocks--Vserver

Volume Size

-------svm1

-----svm1_root

Snapshot Total% Used% ------

64KB

0%

68KB

0%

---------------hourly.2014-07-10_1404 33% hourly.2014-07-10_1405 35%

-------valid valid

0%

hourly.2014-07-10_1604 35%

valid

68KB

0%

hourly.2014-07-10_1605 32%

valid

60KB


2-39

State




Other options for the volume snapshot show command: cluster1::> volume snapshot show ? [ -instance | -time | -fields , ... ] [ -vserver ]

Vserver

[[-volume] ]

Volume

[[-snapshot] <snapshot name>]

Snapshot

[ -create-time ]

Creation Time

[ -busy {true|false} ]

Snapshot Busy

[ -owners , ... ]

List of Owners

[ -size { [KB|MB|GB|TB|PB]} ]

Snapshot Size

[ -blocks ]

Percentage of Total Blocks

[ -usedblocks ]

Percentage of Used Blocks

[ -comment ]

Comment

[ -is-7-mode {true|false} ]

7-Mode Snapshot

[ -snapmirror-label ]

Label for SnapMirror Operations

[ -state {valid|invalid|partial} ]

Snapshot State

[ -is-constituent {true|false} ]

Constituent Snapshot

Example: cluster1::> volume snapshot show -volume svm1_vol1 -snapshot hourly.201410-01_0605 -instance Vserver: svm1 Volume: svm1_vol1 Snapshot: hourly.2014-10-01_0605 Creation Time: Wed Oct 01 06:05:00 2014 Snapshot Busy: false List of Owners: Snapshot Size: 80KB Percentage of Total Blocks: 0% Percentage of Used Blocks: 33% Comment: 7-Mode Snapshot: false Label for SnapMirror Operations: Snapshot State: Constituent Snapshot: false

2-40




Manually Create a Snapshot Copy  Manually create a Snapshot copy: cluster1::> volume snapshot create -vserver svm1 -volume svm1_vol1 -snapshot MySnapshot1 -comment "Manually created Snapshot Copy“

 Obtain information about the Snapshot copy: cluster1::> volume snapshot show -volume svm1_vol1 -snapshot MySnapshot1 ---Blocks--Vserver Volume Snapshot Size Total% Used% svm1 svm1_vol1 MySnapshot1 80KB 0% 32% cluster1::> volume snapshot show -volume svm1_vol1 -snapshot MySnapshot1 -instance 31


MANUALLY CREATE A SNAPSHOT COPY The volume snapshot create command creates a Snapshot copy of a specified volume. cluster1::> volume snapshot create Usage: [ -vserver ]

Vserver (default: svm1)

[-volume]

Volume

[-snapshot] <snapshot name>

Snapshot

[[-comment] ]

Comment

[ -foreground {true|false} ]

Foreground Process (default: true)



Example: cluster1::> volume snapshot create -vserver svm1 -volume svm1_vol1 snapshot MySnapshot1 -comment "Manually created Snapshot Copy" cluster1::> volume snapshot show -volume svm1_vol1

2-41




---Blocks--Vserver -------svm1

Volume Size

Snapshot Total%

Used%

-------svm1_vol1

---------

-----------------------

100KB

0%

daily.2014-09-30_1710 37%

76KB

0%

daily.2014-10-01_0010 31%

96KB

0%

hourly.2014-10-01_0105 36%

96KB

0%

hourly.2014-10-01_0205 36%

80KB

0%

MySnapshot1 32%

5 entries were displayed. cluster1::> volume snapshot show -volume svm1_vol1 -snapshot MySnapshot1 ---Blocks--Vserver

Volume Size

--------------svm1 svm1_vol1 80KB

Snapshot Total%

Used%

------------------------------------- --------

0%

--

MySnapshot1 32%

cluster1::> volume snapshot show -volume svm1_vol1 -snapshot MySnapshot1 -instance Vserver: svm1 Volume: svm1_vol1 Snapshot: MySnapshot1 Creation Time: Wed Oct 01 08:38:35 2014 Snapshot Busy: false List of Owners: Snapshot Size: 80KB Percentage of Total Blocks: 0% Percentage of Used Blocks: 32% Comment: Manually created Snapshot Copy

2-42




7-Mode Snapshot: false Label for SnapMirror Operations: Snapshot State: Constituent Snapshot: false

2-43




Modify a Snapshot Copy  Modify an existing Snapshot copy comment or SnapMirror label: cluster1::> volume snapshot modify –vserver svm1 –volume svm1_vol1 -snapshot MySnapshot1 –comment “This is a new comment”

 Rename the Snapshot copy: cluster1::> volume snapshot rename -vserver svm1 -volume svm1_vol1 -snapshot MySnapshot1 -new-name MySnapshot_Renamed

32


MODIFY A SNAPSHOT COPY The volume snapshot modify command enables you to change the text comment that is associated with a Snapshot copy: cluster1::> volume snapshot modify Usage: [ -vserver ]

Vserver (default: svm1)

[-volume]

Volume

[-snapshot] <snapshot name>

Snapshot

[ -comment ]

Comment



Example: cluster1::> volume snapshot modify –vserver svm1 -volume svm1_vol1 – comment “This is a new comment” cluster1::> volume snapshot rename -vserver svm1 -volume svm1_vol1 snapshot MySnapshot1 -new-name MySnapshot_Renamed cluster1::> volume snapshot show -vserver svm1 -volume svm1_vol1

2-44




---Blocks--Vserver

Volume Size

Snapshot Total%

Used%

--------

--------

---------------

--------

svm1

svm1_vol1 96KB

MySnapshot_Renamed 0%

------

36%

cluster1::> volume snapshot show -vserver svm1 -volume svm1_vol1 snapshot MySnapshot_Renamed -fields comment vserver volume

snapshot

comment

------- --------- ------------------ --------------------svm1

2-45

svm1_vol1 MySnapshot_Renamed This is a new comment




--

Snapshot Copy Space Management Find out the amount of disk space that is reclaimed by deleting a specific Snapshot copy: cluster1::> volume snapshot compute-reclaimable -vserver svm1 -volume svm1_vol2 -snapshots hourly.2014-10-13_1405 A total of 77824 bytes can be reclaimed.

33


SNAPSHOT COPY SPACE MANAGEMENT

2-46




Delete Snapshot Copies  Manually delete a Snapshot copy: cluster1::> volume snapshot delete -vserver svm1 -volume svm1_vol1 -snapshot MySnapshot_Renamed Warning: Deleting a Snapshot copy permanently removes any data that is stored only in that Snapshot copy. Are you sure you want to delete Snapshot copy "MySnapshot_Renamed" for volume "svm1_vol1" in Vserver "svm1" ? {y|n}: y

 Delete all Snapshot copies on a volume: cluster1::> volume snapshot delete -vserver svm1 -volume svm1_vol1 -snapshot *

34


DELETE SNAPSHOT COPIES

2-47




Try This  In your lab kit, to cluster 1.  Enter: volume snapshot partial-restore-file What do you need to know before you can restore a partial file from a Snapshot copy?

 Enter: volume snapshot policy show  Which policies have built-in SnapMirror labels?

 Enter: job schedule cron create test  What is a required field?

 Enter: job schedule cron create test –minute 8 Did this command successfully create the job schedule? When will this job run?

35


TRY THIS

2-48




Lesson 5

Restoring Data from Snapshot Copies

36


LESSON 5: RESTORING DATA FROM SNAPSHOT COPIES

2-49




Ken Asks

What should I consider before I restore data from a Snapshot copy?

37


KEN ASKS

2-50




Considerations Before Restoring Data

SnapRestore license requirement

Privilege requirements

Volume I/O activity

SnapMirror relationships

Use of an entire volume, entire file, or partial file

Use of the root volume

38


CONSIDERATIONS BEFORE RESTORING DATA Customers should consider the following factors before they decide to restore data from a Snapshot copy:      

2-51

SnapRestore license requirement: A SnapRestore license is required to be able to perform Snapshot copy restoration. Privilege requirements: You must have the advanced privilege level or higher to restore the contents of a volume from a Snapshot copy. Snapshot files carry the same permissions as the original files, which keeps the integrity of the security system intact. Volume I/O activity: To restore the contents of a volume from a Snapshot copy, there must be no I/O activity on the volume. Use of an entire volume, entire file, or partial file: You can restore the entire volume from a Snapshot copy, restore an entire file, or restore part of a file. Use of the root volume: If you are required to restore the root volume from a Snapshot copy, the node is rebooted. Nonroot volumes do not require a reboot. SnapMirror relationships: You should manually update all the SnapMirror relationships of a volume immediately after you restore its Snapshot copy. If you do not do so, the result might be unusable SnapMirror relationships that must be deleted and re-created.




Use Data ONTAP Commands to Restore Data Use these Data ONTAP commands to restore data from Snapshot copies:  volume snapshot restore  volume snapshot restore-file

39


USE DATA ONTAP COMMANDS TO RESTORE DATA

2-52




Whole Volume Snapshot Restore 1 of 3

General guidelines for performing a whole volume Snapshot copy restore:  Review the existing Snapshot copies in the volume, and carefully choose the Snapshot copy to restore.  Execute the volume snapshot restore command and answer the prompts.  Review your results.

40


WHOLE VOLUME SNAPSHOT RESTORE: 1 OF 3

2-53





 Review the existing Snapshot copies in the volume: cluster1::> volume snapshot show -vserver svm1 -volume svm1_vol1

 Perform the snapshot restore command for the desired Snapshot copy: cluster1::> volume snapshot restore -vserver svm1 -volume svm1_vol1 -snapshot hourly.2014-10-07_0205

41


WHOLE VOLUME SNAPSHOT RESTORE: 2 OF 3 This volume has eight Snapshot copies: cluster1::> volume snapshot show -vserver svm1 -volume svm1_vol1 ---Blocks--Vserver

Volume

Snapshot

Size

Total%

Used%

--------

--------

-------

--------------------

svm1

svm1_vol1 96KB

0%

weekly.2014-10-04_1715 37%

96KB

0%

weekly.2014-10-05_0015 37%

100KB

0%

daily.2014-10-06_0010 38%

84KB

0%

daily.2014-10-07_0010 34%

0%

hourly.2014-10-07_0105 38%

100KB

2-54




100KB

0%

hourly.2014-10-07_0205 38%

96KB

0%

hourly.2014-10-07_0305 37%

96KB

0%

hourly.2014-10-07_0405 37%

8 entries were displayed. Perform the snapshot restore command: cluster1::> volume snapshot restore -vserver svm1 -volume svm1_vol1 snapshot hourly.2014-10-07_0205 Warning: Snapshot copy "hourly.2014-10-07_0205" is not the most recent copy. Promoting this Snapshot copy will delete all copies made after it. Do you want to continue? {y|n}: y Warning: Quota rules currently enforced on volume svm1_vol1 may change during this operation. If the currently enforced quota rules are different from those in Snapshot copy hourly.2014-10-07_0205, you may have to resize or reinitialize quotas on this volume after this operation. Do you want to continue? {y|n}: y Warning: Export policies currently enforced on the qtrees of volume "svm1_vol1" will not change during this operation. If the currently enforced export policies are different from those in Snapshot copy "hourly.2014-10-07_0205", reassign the export policies of the qtrees on this volume after this operation. Do you want to continue? {y|n}: y

2-55





 Check the Snapshot copies on the volume: cluster1::> volume snapshot show -vserver svm1 -volume svm1_vol1

 Notice that Snapshot copies that are newer than the restored Snapshot copy are deleted.

42


WHOLE VOLUME SNAPSHOT RESTORE: 3 OF 3 Check the Snapshot copies on the volume: cluster1::> volume snapshot show -vserver svm1 -volume svm1_vol1 ---Blocks--Vserver

Volume Size

Snapshot Total%

Used%

--------

--------

----------

------------------------

svm1

svm1_vol1 96KB

0%

weekly.2014-10-04_1715 37%

96KB

0%

weekly.2014-10-05_0015 37%

100KB

0%

daily.2014-10-06_0010 38%

84KB

0%

daily.2014-10-07_0010 34%

100KB

0%

hourly.2014-10-07_0105 38%

96KB

0%

hourly.2014-10-07_0205 37%

6 entries were displayed. 2-56




Notice that Snapshot copies that are newer than hourly.2014-10-07_0205 are deleted.

2-57




Single File Snapshot Restore  Check the volume Snapshot copies: cluster1::> volume snapshot show

 Restore the file (MyFile.txt) from a Snapshot copy: cluster1::> volume snapshot restore-file -vserver svm1 -volume svm1_vol2 -snapshot 2HourSnapshot.2014-09-22_1405 -path /vol/svm1_vol2/MyFile.txt

43


SINGLE FILE SNAPSHOT RESTORE This volume contains three Snapshot copies: cluster1::> volume snapshot show ---Blocks--Vserver

Volume Size

Snapshot Total%

Used%

State

--------

-------

------------

----------------

svm1

svm1_vol2

------

0%

ManualSnapshot1 39%

valid

92KB

0%

ManualSnapshot2 39%

valid

92KB

valid

0%

ManualSnapshot3 36%

80KB

3 entries were displayed. From a Windows host, there is a mapped drive letter to the volume, and you have created a text file named MyFile.txt.

2-58




Check the syntax: cluster1::> volume snapshot restore-file ? [ -vserver ]

Vserver Name (default: svm1)

[ -volume ]

Volume Name

[-snapshot|-s] <snapshot name>

Snapshot Name

[-path]

Filepath

[ -restore-path|-r ]

Restore Filepath

[ -split-disabled [true] ]

Disable Space Efficient LUN Splitting

Perform the volume snapshot restore-file command: cluster1::> volume snapshot restore-file -vserver svm1 -volume svm1_vol2 -snapshot ManualSnapshot2 -path /vol/svm1_vol2/MyFile.txt Since you restored a single file, all the Snapshot copies still exist on the volume: cluster1::> volume snapshot show ---Blocks--Vserver

Volume Size

--------

-------

svm1

svm1_vol2

Snapshot Total% ------------

State Used% ----------------

92KB

0%

ManualSnapshot1 35%

0%

ManualSnapshot2 34%

valid

88KB

0%

ManualSnapshot3 32%

valid

80KB


2-59

------




valid

Use Client Tools to Restore Data  Clients can be enabled to access the Snapshot directories on each volume or on any volume.  Make the Snapshot copy directories visible to NFS and CIFS clients: cluster1::> volume modify -volume svm1_vol1 -snapdir-access true Volume modify successful on volume svm1_vol1 of Vserver svm1.

44


USE CLIENT TOOLS TO RESTORE DATA Customers can use native client tools to access the Snapshot directories on each volume or on any volume. The CIFS or NFS client must have access to the root directory of the volume. Ensure that the client can see the Snapshot directories on the volume.

2-60




Restoring Previous Versions (Windows)

45


RESTORING PREVIOUS VERSIONS (WINDOWS) In Windows, right-click the file, and from the list, choose Restore previous versions.

2-61




Previous Versions Tab (Windows)

46


PREVIOUS VERSIONS TAB (WINDOWS) In the Windows Previous Versions tab, you see choices to restore, open, or copy any previous version of the file. If no previous versions are listed, the file has not changed.

2-62




Previous Versions for Folders (Windows) 1 of 2

47


PREVIOUS VERSIONS FOR FOLDERS (WINDOWS): 1 OF 2 The Previous Versions feature is also available for folders.

2-63




Previous Versions for Folders (Windows) 2 of 2

48


PREVIOUS VERSIONS FOR FOLDERS (WINDOWS): 2 OF 2

2-64




Topic for Discussion

If a tries to recover a file in a Snapshot copy that the has no permissions for, what happens?

49


TOPIC FOR DISCUSSION

2-65




Lesson 6

Snapshot Copy Advanced Topics

50


LESSON 6: SNAPSHOT COPY ADVANCED TOPICS

2-66




Snapshot Copy Advanced Topics  Managing Snapshot copy disk space

 Deleting Snapshot copies automatically  Managing busy Snapshot copies  Managing Snapshot copies on infinite volumes

51


SNAPSHOT COPY ADVANCED TOPICS

2-67




Ken Asks

I really need to have a design where the Snapshot copies are automatically managed. Is that possible?

52


KEN ASKS

2-68




Managing Snapshot Copy Disk Space  Know how much disk space is consumed by all Snapshot copies and by each Snapshot copy.  Predict Snapshot copy disk space consumption.  Delete Snapshot copies automatically.

 Find out how much disk space is recovered if a Snapshot copy is deleted.

53


MANAGING SNAPSHOT COPY DISK SPACE

2-69




Managing Snapshot Copy Disk Space Snapshot Copy Disk Space Consumption

 Monitor the Snapshot copy disk space consumption: cluster1::> volume snapshot show

 Find out how much Snapshot copy space is being used: cluster1::> volume show -vserver svm1 -volume svm1_vol2 -fields snapshot-space-used

54


MANAGING SNAPSHOT COPY DISK SPACE: SNAPSHOT COPY DISK SPACE CONSUMPTION The design of the Snapshot copy feature ensures that the data that is referenced by a Snapshot copy cannot be accidentally deleted. You can monitor Snapshot copy disk consumption by using the volume show command, which displays the amount of free space on a disk. cluster1::> volume snapshot show ---Blocks--Vserver -------svm1

2-70

Volume Size

Snapshot Total%

Used%

-------svm1_root

-------------

-----------------------------

0B

0%

hourly.2014-09-22_1404 0%

0B

0%

hourly.2014-09-22_1405 0%

0B

0%

hourly.2014-09-22_1604 0%

0B

0%

hourly.2014-09-22_1605 0%




svm1_vol1 0B

0%

hourly.2014-09-22_1404 0%

0B

0%

hourly.2014-09-22_1405 0%

0B

0%

hourly.2014-09-22_1604 0%

0B

0%

hourly.2014-09-22_1605 0%

64KB

0%

2HourSnapshot.2014-09-22_1405 29%

68KB

0%

2HourSnapshot.2014-09-22_1605 30%

svm1_vol2

10 entries were displayed. You can also find out how much Snapshot copy space is being used: cluster1::> volume show -vserver svm1 -volume svm1_vol2 -fields snapshotspace-used vserver volume

snapshot-space-used

------- --------- ------------------svm1

2-71

svm1_vol2 13%




Delete Snapshot Copies Automatically Snapshot Copy Autodelete

 View the volume autodelete policy settings that are in effect currently: cluster1::> volume snapshot autodelete show -vserver svm1 -volume svm1_vol2

 Enable the automatic deletion of Snapshot copies, and set the trigger to volume for the svm1_vol2 volume, which is d with the svm1 storage virtual machine: cluster1::> volume snapshot autodelete modify –vserver svm1 -volume svm1_vol2 -enabled true -trigger volume

*Important! For SAN solutions, refer to the Clustered Data ONTAP 8.3 SAN istration Guide. 55


DELETE SNAPSHOT COPIES AUTOMATICALLY: SNAPSHOT COPY AUTODELETE   

You can define and enable a policy to automatically delete Snapshot copies. You can better manage space usage if you automatically delete Snapshot copies. You can automatically delete Snapshot copies from read/write volumes. You cannot set up automatic deletion of Snapshot copies from infinite volumes or from read-only volumes, for example, SnapMirror destination volumes. You can define and enable a policy to automatically delete Snapshot copies by using the volume snapshot autodelete modify command.

Viewing settings for the automatic deletion of Snapshot copies View the volume autodelete policy settings that are currently in effect: cluster1::> volume snapshot autodelete show -vserver svm1 -volume svm1_vol2 Vserver Name: svm1 Volume Name: svm1_vol2 Enabled: false  The autodelete policy is not turned on. Commitment: try Defer Delete: _created Delete Order: oldest_first Defer Delete Prefix: (not specified) Target Free Space: 20% Trigger: volume 2-72




Destroy List: none Is Constituent Volume: false Snapshot policy autodelete modify command The volume snapshot autodelete modify command enables you to modify Snapshot automatic deletion policy settings.  

Based on the defined policy, automatic deletion of Snapshot copies is triggered. Automatic deletion of Snapshot copies is useful when you want to automatically reclaim space that is consumed by the Snapshot copies from the volume, when the volume is low in available space. This command works only on a read/write parent volume. You cannot set up automatic Snapshot copy deletion for striped volumes and read-only volumes.



cluster1::> volume snapshot autodelete modify ? -vserver

Vserver Name

[-volume]

Volume Name

[[-enabled] {true|false}]

Enabled

[ -commitment {try|disrupt|destroy} ]

Commitment

[ -defer-delete {scheduled|_created|prefix|none} ]

Defer Delete

[ -delete-order {newest_first|oldest_first} ]

Delete Order

[ -defer-delete-prefix ]

Defer Delete Prefix

[ -target-free-space ]

Target Free Space

[ -trigger {volume|snap_reserve|space_reserve}

Trigger

[ -destroy-list ]

Destroy List

Enable the automatic deletion of Snapshot copies, and set the trigger to volume for the svm1_vol2 volume, which is part of the svm1 SVM: cluster1::> volume snapshot autodelete modify -vserver svm1 -volume svm1_vol2 -enabled true -trigger volume Volume modify successful on volume: svm1_vol2 Check your work: cluster1::> volume snapshot autodelete show -vserver svm1 -volume svm1_vol2 Vserver Name: svm1 Volume Name: svm1_vol2 Enabled: true  The autodelete feature is enabled. Commitment: try Defer Delete: _created Delete Order: oldest_first Defer Delete Prefix: (not specified) Target Free Space: 20%

2-73




Trigger: volume  The trigger is volume free space. Destroy List: none Is Constituent Volume: false This policy triggers when the free space in the volume falls below 20%.

2-74




Manage Busy Snapshot Copies 1 of 2

A Snapshot copy is in a busy state if there are any applications that are using it:  Create a FlexClone of a volume: cluster1::> volume clone create -vserver svm1 -flexclone svm1_vol1_Clone -type RW -parent-volume svm1_vol1

 Check to see if there are any busy Snapshot copies appearing in the volume: cluster1::> volume snapshot show -vserver svm1 -volume svm1_vol1 -busy true

56


MANAGE BUSY SNAPSHOT COPIES: 1 OF 2 Create a FlexClone of a volume: cluster1::volume clone create -vserver svm1 -flexclone svm1_vol1_Clone type RW -parent-volume svm1_vol1 Check to see if there are any busy Snapshot copies in the volume: cluster1::volume snapshot show -vserver svm1 -volume svm1_vol1 -busy true ---Blocks--Vserver -------svm1

Volume Size

Snapshot Total%

State Used%

------svm1_vol1

------------

-------

--------

0%

clone_svm1_vol1_C.0 0%

valid

0B

NOTE: The –busy true switch shows only the busy Snapshot copies in the output.

2-75




Manage Busy Snapshot Copies 2 of 2

Check the owner of a busy Snapshot copy: cluster1::> volume snapshot show -vserver svm1 -volume svm1_vol1 -fields owners vserver volume

snapshot

------- --------- ---------------------svm1 clone"

57

owners ------

svm1_vol1 clone_svm1_vol1_clone.0 "volume


MANAGE BUSY SNAPSHOT COPIES: 2 OF 2 cluster1::> volume snapshot show -vserver svm1 -volume svm1_vol1 -fields owners vserver

volume

snapshot

owners

-------

---------

----------------------

------

svm1

svm1_vol1

clone_svm1_vol1_clone.0

"volume clone"

After you know the owner of the Snapshot copy, then you can determine if you can stop the application that is holding the Snapshot copy busy. When the Snapshot copy is not busy, then you can safely delete it.

2-76




Snapshot Copies on Infinite Volumes Guidelines for Creating Snapshot Copies of Infinite Volumes

 The volume must be online.

 The Snapshot copy schedule should not be less frequent than hourly.  Time should be synchronized across all nodes.  The Snapshot copy creation job can run in the background.  You cannot rename the Snapshot copy.

 You cannot modify the comment or SnapMirror label.

58


SNAPSHOT COPIES ON INFINITE VOLUMES: GUIDELINES FOR CREATING SNAPSHOT COPIES OF INFINITE VOLUMES Guidelines for creating Snapshot copies of infinite volumes You can create, manage, and restore Snapshot copies of infinite volumes. However, be aware of the factors that affect the Snapshot creation process and the requirements for managing and restoring the copies.   

The volume must be online. You cannot create a Snapshot copy of an infinite volume if the infinite volume is in a mixed state because a constituent is offline. The Snapshot copy schedule should not be less than hourly.

It takes longer to create a Snapshot copy of an infinite volume than of a FlexVol volume. If you schedule Snapshot copies of infinite volumes for less than hourly, Data ONTAP attempts the schedule but might not meet it. Scheduled Snapshot copies are missed when the previous Snapshot copy is still being created.   

Time should be synchronized across all the nodes that the infinite volume spans. Synchronized time helps schedules for Snapshot copies run smoothly and helps the restoration of Snapshot copies function properly. The Snapshot copy creation job can run in the background.

Creating a Snapshot copy of an infinite volume is a volume-level job (unlike the same operation on a FlexVol volume), and the operation spans multiple nodes in the cluster. You can force the job to run in the background by setting the -foreground parameter of the volume snapshot create command to false. After you create Snapshot copies of an infinite volume, you cannot rename the copy or modify the comment or SnapMirror label for the copy. For more information about infinite volumes, refer to the Clustered Data ONTAP 8.3 Infinite Volumes Management Guide.

2-77




Snapshot Copies on Infinite Volumes Guidelines for Managing Snapshot Copy Disk Consumption

 You cannot calculate the amount of disk space that can be reclaimed.  The size of a Snapshot copy for an infinite volume excludes the size of namespace mirror constituents.

 To reclaim disk space that is used by Snapshot copies of infinite volumes, you must manually delete the copies.

59


SNAPSHOT COPIES ON INFINITE VOLUMES: GUIDELINES FOR MANAGING SNAPSHOT COPY DISK CONSUMPTION Guidelines for managing Snapshot copy disk consumption   

You cannot calculate the amount of disk space that can be reclaimed if Snapshot copies of an infinite volume are deleted. The size of a Snapshot copy for an infinite volume excludes the size of namespace mirror constituents. To reclaim disk space that is used by Snapshot copies of infinite volumes, you must manually delete the copies.

You cannot use the automatic Snapshot copy deletion feature to automatically delete Snapshot copies of infinite volumes. However, you can manually delete Snapshot copies of infinite volumes, and you can run the delete operation in the background. For more information about infinite volumes, refer to the Clustered Data ONTAP 8.3 Infinite Volumes Management Guide.

2-78




Snapshot Copies on Infinite Volumes Guidelines for Restoring Snapshot Copies of Infinite Volumes

 You must restore the entire Snapshot copy of the infinite volume.  The Snapshot copy must be in a valid state.  Restored Snapshot copies inherit the current efficiency settings of the infinite volume.

60


SNAPSHOT COPIES ON INFINITE VOLUMES: GUIDELINES FOR RESTORING SNAPSHOT COPIES OF INFINITE VOLUMES Guidelines for restoring Snapshot copies of infinite volumes    

You must restore the entire Snapshot copy of the infinite volume. You cannot restore single files or parts of files. You also cannot restore a Snapshot copy of a single constituent. The Snapshot copy must be in a valid state.

You cannot use the privilege to restore a Snapshot copy of an infinite volume if the copy is in a partial or invalid state, because the commands require diagnostic privilege. However, you can technical to run the commands for you. Restored Snapshot copies inherit the current efficiency settings of the infinite volume. If the State field contains a dash ("-") when you run the volume snapshot show command after restoring a Snapshot copy of the infinite volume, one or more constituents are offline or have a different state than the other constituents. To align the State field for constituents in an infinite volume, set the efficiency settings again on the infinite volume. For more information about infinite volumes, refer to the Clustered Data ONTAP 8.3 Infinite Volumes Management Guide.

2-79




References  NetApp University courses:  Clustered Data ONTAP istration

 NetApp Documentation    

Clustered Data ONTAP 8.3 Data Protection Guide Clustered Data ONTAP 8.3 Infinite Volumes Management Guide Clustered Data ONTAP 8.3 Data Protection Tape Guide NetApp Technical Report TR-4210:  Operational How-To Guide: NetApp Snapshot Management

61


REFERENCES

2-80




What Ken Learned in this Module

I learned how to design and deploy an effective Snapshot copy solution. I can now manage my Snapshot copies effectively. I feel confident restoring data from Snapshot copies, and I understand how to give my s access to recover their own data from Snapshot copies.

62


WHAT KEN LEARNED IN THIS MODULE

2-81




Exercise

Module 2: Data Protection with NetApp Snapshot Copies

Time Estimate: 30 Minutes

63


EXERCISE Please refer to your exercise guide.

2-82




Module 3

NetApp Replication Technologies

1


MODULE 3: NETAPP REPLICATION TECHNOLOGIES

3-1

Data ONTAP Data Protection istration: NetApp Replication Technologies



Module Objectives After this module, you should be able to:

 Compare deployment use cases for NetApp replication solutions  Recognize important features of each replication solution

 Understand design criteria for both SnapMirror and SnapVault relationships  Recognize SnapMirror policy and Data ONTAP volume types  Understand cluster and storage virtual machine (SVM) peering relationships 2


MODULE OBJECTIVES

3-2




Lesson 1

Data ONTAP Replication Technology

3


LESSON 1: DATA ONTAP REPLICATION TECHNOLOGY

3-3




Integrated Data Protection Source Volume

Mirrored Volume

Snapshot Copies

SnapMirror

WAN SnapMirror

FlexClone Volume

4

Mirrored Volume

Tape


INTEGRATED DATA PROTECTION NetApp Snapshot technology is the foundation for SnapRestore, SnapManager, SnapDrive, SnapMirror, FlexClone, and SnapVault operations. Data protection capabilities are integrated within the Data ONTAP operating system. SnapMirror technology is integrated with Snapshot technology. Because these technologies are integrated, you can quickly and efficiently create on-disk replicas or point-in-time copies of data that do not require an actual copy operation. You can use NetApp integrated data protection to create an on-disk, quickly accessible history of applicationconsistent Snapshot copies, which eliminates the need for traditional backup windows. SnapMirror software then replicates the history of Snapshot copies to destination volumes that you can use for backup, disaster recovery, or testing and development. SnapMirror replication is efficient, because it replicates only the 4-kilobyte (KB) blocks that changed or were added since the previous update. You gain efficiency when SnapMirror software is combined with NetApp storage efficiency technologies. When FAS deduplication is used on the primary storage, only unique data is replicated to the disaster recovery site. If compression is enabled on the source, then compression is maintained on the destination. Data is not uncompressed, because it is replicated. These technologies can result in telecommunications savings and significant storage capacity savings.

3-4




SnapMirror for Disaster Recovery

SnapMirror

Production Site

5

Disaster Recovery Site


SNAPMIRROR FOR DISASTER RECOVERY SnapMirror technology is integral to disaster recovery planning. If critical data is replicated to a different physical location, a serious disaster does not have to result in extended periods of unavailable data. Clients can access replicated data across the network until the damage that is caused by the disaster is repaired. Application servers at the recovery site can access replicated data to restore operations for business-critical applications for as long as necessary to recover the production site. Recovery might include recovery from corruption, natural disaster at the production site, accidental deletion, and so on. If a disaster that requires a failover occurs, and the primary storage is not completely lost, SnapMirror technology provides an efficient means of resynchronizing the primary and disaster recovery sites. When the primary site is back online, SnapMirror technology resynchronizes the two sites, transferring only changed or new data back to the primary site from the disaster recovery site. To do this, SnapMirror technology simply reverses the SnapMirror relationships. After the primary production site resumes normal application operations, the SnapMirror transfers to the disaster recovery facility resume without requiring another complete data transfer.

3-5




SnapVault for Long-Term Retention of Backups

Snapshot1 Snapshot2 Snapshot3 Snapshot4 Snapshot5

Snapshot1 Snapshot2 Snapshot3

6


SNAPVAULT FOR LONG-TERM RETENTION OF BACKUPS SnapVault technology enables you to configure a policy that can retain more than two Snapshot copies on the source and destination volumes. You can use SnapVault technology to restore data to the primary, similar to a traditional backup application.

3-6




FlexClone Volumes for Disaster Recovery Testing Development Testing Disaster Recovery Testing

SnapMirror

FlexCone Volume Snapshot Copies

7


FLEXCLONE VOLUMES FOR DISASTER RECOVERY TESTING You can use NetApp FlexClone technology to quickly create a read/write copy of a SnapMirror destination FlexVol volume, which eliminates the need for additional copies of the data.  

3-7

For example, a 10-gigabyte (GB) FlexClone volume does not require another 10-GB FlexClone volume; it requires only the metadata that is needed to define the FlexClone volume. FlexClone volumes store only data that is written or changed after a clone is created.




Data Distribution and Remote Data Access

SnapMirror

8


DATA DISTRIBUTION AND REMOTE DATA ACCESS You can use SnapMirror technology to distribute large amounts of data throughout the enterprise, which enables read-only access to data at remote locations.  

3-8

Remote data access provides faster access to data by clients in the remote locations. Remote data access enables more efficient and more predictable use of expensive network and server resources, because WAN usage occurs at a predetermined replication time. Storage s can replicate production data at a specific time, to minimize overall network utilization.




Backup Offloading and Remote Tape Archiving

SnapMirror

9


BACKUP OFFLOADING AND REMOTE TAPE ARCHIVING You can use SnapMirror technology for backup consolidation and for offloading tape backup overhead from production servers.  

3-9

The SnapMirror process facilitates centralized backup operations and reduces backup istrative requirements at remote locations. Because Snapshot technology eliminates the traditional backup window on the primary storage system, you can greatly reduce the overhead of backup operations on production storage systems by offloading tape backups to a SnapMirror destination.




Unified Architecture Flexibility

SnapMirror

10


UNIFIED ARCHITECTURE FLEXIBILITY You can use SnapMirror technology between FAS and FlexArray Storage Virtualization Software storage systems. You can deploy systems with different performance characteristics and different costs at the primary and at disaster recovery sites. For example, depending on the capabilities that are required, the disaster recovery site might contain these:   

3-10

A lower-model storage system SATA disk versus FC disk The iSCSI or FCoE protocol versus FC




NetApp Replication Technologies Key Terminology

     

11

Data protection mirror copies Primary volume Secondary volume Baseline transfer Incremental transfer SnapMirror label


NETAPP REPLICATION TECHNOLOGIES: KEY TERMINOLOGY Data protection mirror copies provide asynchronous disaster recovery. Data protection mirror relationships enable you to perform these actions:   

Periodically create Snapshot copies of data on one volume. Copy those Snapshot copies to a partner volume (the destination volume), usually on another cluster. Retain those Snapshot copies.

The mirror copy on the destination volume ensures quick availability and restoration of data from the time of the latest Snapshot copy, if the data on the source volume is corrupted or lost. A primary volume is a volume that contains data that is to be backed up.  

In cascade or fan-out backup deployments, the primary volume is the volume that is backed up to a SnapVault backup, regardless of where the SnapVault source is in the chain. In a cascade or chain configuration in which A has a mirror relationship to B, and B has a SnapVault relationship to C, B serves as the source for the SnapVault backup. B does this even though it is a secondary destination in the chain.

A secondary volume is a volume to which data is backed up from a primary volume. Such a volume can be a secondary or tertiary (and onward) destination in a cascade or fan-out backup configuration. The SnapVault secondary system maintains Snapshot copies for long-term storage and possible restore operations. A baseline transfer is an initial complete backup of a primary storage volume to a corresponding volume on the secondary system. An incremental transfer is a follow-up backup to the secondary system that contains only the changes to the primary data since the last transfer action.

3-11




A SnapMirror label is an attribute that identifies Snapshot copies for the purpose of selection and retention in SnapVault backups. Each SnapVault policy configures the rules for selecting Snapshot copies on the primary volume and transferring the Snapshot copies that match a given SnapMirror label.

3-12




Answer this Question

Which NetApp replication technology allows you to retain as many Snapshot copies as you need? a) SnapMirror b) SnapVault c) All of the above d) None of the above

12


ANSWER THIS QUESTION

3-13




Lesson 2

Design Criteria for SnapVault and SnapMirror Relationships

13


LESSON 2: DESIGN CRITERIA FOR SNAPVAULT AND SNAPMIRROR RELATIONSHIPS

3-14




Ken Asks

What are the rules for deploying a SnapVault solution?

14


KEN ASKS

3-15




General Guidelines for SnapVault Relationships 1 of 2

 A volume can be in multiple relationships.  A volume can be the secondary volume for only one relationship.

 Multiple primary volumes cannot replicate to a single secondary volume.  SnapVault can integrate with data protection mirror relationships.  Primary or secondary volumes cannot be 32-bit volumes.  The primary volume of a SnapVault backup should not be a FlexClone volume. 15


GENERAL GUIDELINES FOR SNAPVAULT RELATIONSHIPS: 1 OF 2 The following guidelines apply to all SnapVault relationships:   

      

3-16

A volume can be in multiple relationships, either as the secondary or the primary. A volume can be the primary for multiple secondary volumes and also the secondary volume for another primary volume. A volume can be the secondary volume for only one SnapVault relationship. You cannot configure SnapVault relationships from multiple primary volumes to a single SnapVault secondary volume. For example, if you want to back up an entire SVM to a SnapVault backup, then you must create a separate secondary volume for each volume in the SVM and create a separate SnapVault relationship for each primary volume. You can configure SnapVault relationships to be used simultaneously with data protection mirror relationships. Primary or secondary volumes cannot be 32-bit volumes. The primary of a SnapVault backup should not be a FlexClone volume. The relationship will work, but the efficiency that is provided by FlexClone volumes is not preserved. Primary and secondary volumes must have the same volume language settings. After you establish a SnapVault relationship, you cannot change the language that is assigned to the secondary volume. A SnapVault relationship can be only one leg of a cascade chain. After you establish a SnapVault relationship, you can rename primary or secondary volumes. If you rename a primary volume, it can take a few minutes for the relationship to recover from the name change.




General Guidelines for SnapVault Relationships 2 of 2

 Primary and secondary volumes must have the same volume language settings.  After you establish a SnapVault relationship, you cannot change the language that is assigned to the secondary volume.  A SnapVault relationship can be only one leg of a cascade chain.  After you establish a SnapVault relationship, you can rename primary or secondary volumes.

16


GENERAL GUIDELINES FOR SNAPVAULT RELATIONSHIPS: 2 OF 2 The following guidelines apply to all SnapVault relationships:     

3-17

A SnapVault secondary volume cannot be the primary volume of FlexCache volumes. Primary and secondary volumes must have the same volume language settings. After you establish a SnapVault relationship, you cannot change the language that is assigned to the secondary volume. A SnapVault relationship can be only one leg of a cascade chain. After you establish a SnapVault relationship, you can rename primary or secondary volumes. If you rename a primary volume, it can take a few minutes for the relationship to recover from the name change.




Planning the Snapshot Copy Schedule  SnapVault technology allows flexibility in planning Snapshot scheduling and retention.  Choose an appropriate Snapshot label.  The SnapVault policy contains important information about the relationship.

17


PLANNING THE SNAPSHOT COPY SCHEDULE It is important to plan the Snapshot copy transfer schedule and retention for your SnapVault backups. When planning SnapVault relationships, consider these guidelines: 

Before you create a SnapVault policy, you should plan which Snapshot copies you want to replicate to the SnapVault secondary volume and how many of each copy you want to keep. For example: – – –

 

The primary volume should have an assigned Snapshot policy that creates Snapshot copies at the intervals you need and that labels each Snapshot copy with the appropriate snapmirror-label attribute name. The SnapVault policy that is assigned to the SnapVault relationship: – –

3-18

Hourly (periodically throughout the day): Does the data change often enough throughout the day to make it worthwhile to replicate a Snapshot copy every hour, every two hours, or every four hours? Nightly: Do you want to replicate a Snapshot copy every night, or just on workday nights? Weekly: How many weekly Snapshot copies is it useful to keep in the SnapVault secondary volume?

The policy should select the Snapshot copies that you want from the primary volume, which are identified by the snapmirror-label attribute name. The policy should specify how many Snapshot copies of each name you want to keep on the SnapVault secondary volume.




SnapVault and Storage Efficiencies A new feature for the clustered Data ONTAP version of SnapVault software is the ability to preserve the storage efficiencies of the primary volume throughout a SnapVault transfer.

18


SNAPVAULT AND STORAGE EFFICIENCIES Preserving storage efficiencies 

If deduplication and compression are enabled on the primary volume, these efficiencies are preserved during SnapVault transfers. The result: – – –

 



Less data that is transmitted over the network Shorter backup windows Bandwidth savings

In addition, data is already deduplicated and compressed on the secondary volume after the SnapVault transfer completes, without running the deduplication and compression processes on the secondary volume. A successful SnapVault transfer requires at least the same amount of available free space on the secondary volume as is required to copy the data from the primary volume in a nondeduplicated and noncompressed format. During the SnapVault transfer, it appears that the expanded size of the primary data is consumed on the secondary. However, as soon as the transfer completes, the storage efficient dataset size is reflected in the amount of space that is consumed. Deduplication and compression can run on the secondary volume after the SnapVault transfer has completed, independent of the storage efficiencies that are present on the primary volume. However, enabling the compression process to run on the SnapVault secondary volume causes the storage efficiencies that are present on the primary volume to not be preserved on the SnapVault secondary volume.

For more information about this caveat and other best practices that are associated with using storage efficiencies with SnapVault software, refer to section 14.1 of TR-3966: NetApp Data Compression and Deduplication Deployment and Implementation Guide. TR-3966 also helps you understand how much disk space you can save by using storage efficiencies with SnapVault software.

3-19




SnapVault Sizing Considerations  Disk space

 Data throughput  Client I/O impact

19


SNAPVAULT SIZING CONSIDERATIONS Disk space   

The SnapVault secondary, or target, volume must be sized so that adequate disk space is available to retain all of the planned backups. You can estimate this space fairly accurately by using known information about the primary data and the required recovery point objectives (RPOs). This calculation is independent of the systems that are used. The data that you need to calculate the required disk space on the SnapVault target is calculated with this information: – – – –

The size of the primary data The size of the daily, weekly, and monthly data change rates The number of daily, weekly, and monthly backup copies to be kept The space savings that can be anticipated by using NetApp deduplication and compression

Data throughput It is also important to size a SnapVault solution so that data can be transferred fast enough to complete backups in the amount of time that is available.   

3-20

First, determine how much data, on average, is transferred in a SnapVault incremental update and how much time must be allotted to complete the backup. Then, use these numbers to determine the required data throughput in units such as megabytes per second. If you have multiple SnapVault relationships, to consider the size of all the backups that must be completed during a given backup window. After you determine the required data throughput speed, select an appropriate NetApp FAS system that can handle the data speed that is required.




Client I/O impact When sizing a SnapVault solution, consider the impact of the SnapVault processes on the other workloads that are running on a system. Clients generally experience more latency with a greater number of concurrent SnapVault streams. If a large number of transfers need to take place during a given backup window, it may be best to stagger the updates so that they don’t all run at the same time and cause unacceptable latency to clients. For example, if 80 relationships must update within a 10-hour window, and it is determined that, on average, eight concurrent updates can finish in one hour, then you can create a transfer schedule that starts eight new updates every hour during the given 10-hour window. When sizing a SnapVault solution, first establish an acceptable client I/O latency threshold. Then, determine what the client I/O latency impact will be for a specific configuration, based on these factors:   

3-21

The system that is used The type of dataset that is backed up The number of concurrent transfers that are needed to complete all backups within a given window




Basic Backup Deployment

The SnapMirror source volume is online and writable.

20

The SnapMirror destination volume is online and readonly.


BASIC BACKUP DEPLOYMENT A basic data protection deployment consists of two volumes, either FlexVol volumes or infinite volumes, in a one-to-one, source-to-destination relationship. This deployment backs up data to one location, which provides a minimal level of data protection.  

Source volumes are the data objects that need to be replicated. Typically, s can access and write to source volumes. Destination volumes are data objects to which the source volumes are replicated. Destination volumes are read-only. – – – –

3-22

Destination FlexVol volumes are usually placed on a different SVM from the source SVM. Destination infinite volumes must be placed on a different SVM from the source SVM. Destination volumes can be accessed by s, in case the source becomes unavailable. s can use SnapMirror commands to make the replicated data at the destination accessible and writable.




Fan-Out and Cascade Relationships  Fan-out (one-to-many relationship):

 Mirror-SnapVault fan-out (FlexVol volumes only)  Multiple-mirrors fan-out (FlexVol volumes and infinite volumes)

 Cascade (one-to-one-to-one relationship):

 Mirror-mirror cascade (FlexVol volumes only)  Mirror-SnapVault cascade (FlexVol volumes only)  SnapVault-mirror cascade (FlexVol volumes only)

21


FAN-OUT AND CASCADE RELATIONSHIPS

3-23




Fan-Out Topologies Mirror-SnapVault Fan-Out SnapMirror

SnapVault

Multiple Mirrors Fan-Out SnapMirror

SnapMirror

22


FAN-OUT TOPOLOGIES Fan-out topologies In a fan-out topology, a single primary volume is replicated to multiple destinations.  

A primary volume can be protected and backed up. A read-only copy is available at a secondary site.

Only one SnapVault relationship is allowed in a fan-out topology.

3-24




Cascading SnapVault and SnapMirror Relationships Mirror-SnapVault Cascade SnapMirror

SnapVault

SnapVault-SnapMirror Cascade SnapVault

SnapMirror

Mirror-Mirror Cascade SnapMirror

23

SnapMirror


CASCADING SNAPVAULT AND SNAPMIRROR RELATIONSHIPS Cascading SnapMirror and SnapVault relationships are ed in clustered Data ONTAP 8.3.  

A SnapMirror secondary can be the source of a SnapVault relationship (backing up a data recovery mirror), or a SnapVault secondary can be the source of a SnapMirror relationship (protecting a backup). In the case of a SnapMirror to SnapVault relationship, it is not possible to specify which Snapshot copies are transferred to the SnapVault secondary; SnapVault always transfers the SnapMirror exported Snapshot copy or the base copy of the SnapMirror relationship. This is similar to the snapmirror base snapshot only option in Data ONTAP operating in 7-Mode.

An can choose the following:  

When the most recent version of the SnapMirror exported Snapshot copy is transferred to the SnapVault secondary How many copies to keep on the SnapVault secondary

Only one SnapVault relationship is ed in a cascade chain configuration, but many SnapVault relationships are ed in a fan-out configuration; multiple mirror relationships are ed. The longer you configure a chain of relationships, or the more you add fan-out destinations, the greater the risk that Snapshot copies will be locked on the source. Depending on the update schedule, the worst case is when one Snapshot copy is locked for each cascade or fan-out destination.

3-25




Fan-In SnapVault Topologies SnapVault Fan-In Storage System B

Storage System C

Storage System A

24


FAN-IN SNAPVAULT TOPOLOGIES Fan-in topologies In a fan-in topology, multiple SnapVault primaries back up to one SnapVault secondary. The primary use case for this topology is multiple remote sites that back up to one central data center. System-level fan-in In clustered Data ONTAP, because the relationships are configured at the volume level, it is not possible to back up multiple volumes to one secondary volume. However, it is possible to back up SnapVault primary volumes from multiple SVMs and clusters to separate volumes in a single destination SVM, which can be in a different cluster.

3-26




Creating a Tiered Backup Policy In a tiered backup strategy, a SnapVault policy can have several rules. Each rule identifies a different set of Snapshot copies.

25


CREATING A TIERED BACKUP POLICY Data ONTAP uses the snapmirror-label attribute to identify Snapshot copies between primary and secondary FlexVol volumes in a SnapVault relationship. When you configure rules in a SnapVault policy, you enter the snapmirror-label name that you want to use to identify the Snapshot copies to which the rule applies. In a tiered backup strategy, a SnapVault policy might have several rules, and each rule identifies a different set of Snapshot copies. In this example, you have a volume to which you have assigned a Snapshot policy that specifies the following schedule:   

3-27

An hourly Snapshot copy: Every two hours, a Snapshot copy is created and is assigned the attribute snapmirror-label hourly. A daily Snapshot copy: Every day at 5:00 p.m., a Snapshot copy is created and is assigned the attribute snapmirrorlabel daily. A weekly Snapshot copy: Every Friday at 6:00 p.m., a Snapshot copy is created and is assigned the attribute weekly.




SnapMirror Design Considerations Naming Guidelines

Before you create a SnapMirror policy or SnapVault policy, create a unique policy name.  Cluster-wide policy names must be unique within the cluster and must not conflict with any SVM-wide policy names.  SVM-wide policy names must be unique within the SVM in which the policy is created. However, an SVM policy name can be the same as a policy name that was created in a different SVM, as long as the name does not conflict with any cluster-wide policy name.

26


SNAPMIRROR DESIGN CONSIDERATIONS: NAMING GUIDELINES Cluster-wide policy names must be unique within the cluster and must not conflict with any SVM-wide policy names. SVM-wide policy names must be unique within the SVM in which the policy is created. However, an SVM policy name can be the same as a policy name that was created in a different SVM, if the name does not conflict with any cluster-wide policy names. Path name pattern matching:    

You can use pattern matching when you use SnapMirror commands, so that the command works on selected mirroring relationships. SnapMirror commands use fully qualified path names in the following format: vserver:volume. You can abbreviate the path name if you do not enter the SVM name. In such a case, the SnapMirror command assumes the local SVM context of the . For example, if the SVM is called “vserver1” and the volume is called “vol1,” the fully qualified path name is vserver1:vol1. You can use the asterisk (*) in paths as a wildcard to select matching, fully qualified path names. The following table provides examples of using the wildcard to select a range of volumes. *

Matches all paths

vs*

Matches all SVMs and volumes with SVM names that begin with vs

*:*src*

Matches all SVMs with volume names that contain the src text

*:vol*

Matches all SVMs with volume names that begin with vol

Language setting requirements: 

3-28

Source and destination FlexVol volumes or infinite volumes of a mirror relationship must use the same language setting; otherwise, NFS or CIFS clients might be unable to access data. Data ONTAP Data Protection istration: NetApp Replication Technologies



 

3-29

For FlexVol volumes, it is not a problem if the source and destination volumes are located on the same SVM, because the language is set on the SVM. For FlexVol volumes and infinite volumes with mirror relationships between volumes on two different SVMs, the language setting on the SVMs must be the same.




Limitations for Data Protection Mirror Relationships 1 of 2

 Snapshot copies cannot be deleted on destination volumes.  An empty junction path on a destination FlexVol volume is not accessible from CIFS clients.  A FlexClone volume should not be the source of a data protection mirror relationship.

27


LIMITATIONS FOR DATA PROTECTION MIRROR RELATIONSHIPS: 1 OF 2 Snapshot copies cannot be deleted automatically on destination volumes:   

You cannot automatically delete old Snapshot copies on destination FlexVol volumes or infinite volumes of mirror relationships, because the destination volume is a read-only version of the source volume. The destination volume should contain the same data as the source volume. This is not true of Snapshot copies on destination FlexVol volumes of SnapVault relationships. You can delete old Snapshot copies on SnapVault secondary volumes. If you use the snap autodelete command to automatically delete Snapshot copies from a destination volume, to remove older Snapshot copies, the execution will fail.

An empty junction path on a destination volume is not accessible from CIFS clients:  

If internally mounted FlexVol volumes form a namespace, and you have a mirror relationship, CIFS clients on a destination volume that attempt to view mirrored volumes that are not at the highest level of the namespace are denied access. This situation occurs when you create a namespace by using more than one volume, in which one volume is the source volume of a mirror relationship and the other volumes are of the namespace. For example, assume that you have two volumes: vol x, which has a junction path /x, and vol y, which has a junction path /x/y: – –

When a SnapMirror transfer occurs, a directory under vol x is created for vol y on the destination volume. From an NFS client, you can see that the directory is empty, but from a CIFS client, you get the following message: access is denied.

The primary of a SnapVault backup should not be a FlexClone volume. The relationship will work, but the efficiency that is provided by FlexClone volumes is not preserved.

3-30




Limitations for Data Protection Mirror Relationships 2 of 2

Fan-out limits depend on the type of mirror relationship that you want to fan out from a single source volume:  For load-sharing mirror relationships, you can fan out a maximum of one destination volume per node for one source volume.  For data protection mirror relationships, you can fan out a maximum of eight destination volumes from one source volume.  A source volume can have both one load-sharing destination volume on a node and eight data protection destination volumes. 28


LIMITATIONS FOR DATA PROTECTION MIRROR RELATIONSHIPS: 2 OF 2 Mirror relationship fan-out limits When you are planning the number and types of mirror relationships for a source volume, that the source volume is limited in the number of destination volumes that it can have. Fan-out limits depend on the type of mirror relationship that you want to fan out from a single source volume:    

3-31

For load-sharing mirror relationships, you can fan out a maximum of one destination volume on a node for a single source volume. The maximum number of nodes within a cluster depends on the platform model and licensed protocols. For detailed information about cluster size limits, see the Hardware Universe (formerly the System Configuration Guide) at .netapp.com/knowledge/docs/hardware/NetApp/syscfg/ index.shtml. For data protection mirror relationships, you can fan out a maximum of four destination volumes from a single source volume. A single source volume can contain both one load-sharing destination volume on a node and four data protection destination volumes.




Topic for Discussion

What is a good reason to deploy load-sharing mirror relationships?

29


TOPIC FOR DISCUSSION

3-32




Load-Sharing Mirror Copies for ReadOnly Workloads The server accesses the read/write path to the source volume.

RW

SnapMirror

LS*

LS*

LS*

LS*

*

Read-Only LS Mirror Copies

NFSv3 and CIFS clients access read-only from multiple copies. 30


LOAD-SHARING MIRROR COPIES FOR READ-ONLY WORKLOADS To improve performance for read-only workloads, you can use load-sharing mirror copies to distribute data volumes to other nodes in the cluster. Load-sharing mirrors for SVM namespace root volumes:   

A namespace root volume is very small. It contains only directories that are used as mount points, which are the paths where data volumes are junctioned (mounted) into the namespace. However, namespace root volumes are extremely important for NAS clients, which are not able to access data if the SVM root volume is unavailable. SAN client connections (that is, FC, FCoE, or iSCSI) do not depend on the SVM root volume.

Best practice: Create a load-sharing mirror copy of a NAS SVM namespace root volume on every node in the cluster, so that the root of the namespace is available, regardless of node outages or node failovers.

3-33




Lesson 3

Policies and Volume Types

31


LESSON 3: POLICIES AND VOLUME TYPES

3-34




Ken Asks

What are the different volume types I keep hearing about?

32


KEN ASKS

3-35




Volume Types  Volume types:  FlexVol volumes  Infinite volumes

 Volume variations:  FlexClone volumes  Data protection mirrors  Load-sharing mirrors

33


VOLUME TYPES

3-36




FlexVol Volumes  Flexible volumes in clustered Data ONTAP are the same as in environments with Data ONTAP 7G or Data ONTAP operating in 7-Mode.  Any single volume can exist within a single data SVM only.

 Volumes are ed together through junctions to create the namespace of an SVM.  Volumes are the unit of data management. Volumes can be moved, copied, mirrored, backed up, or copied by using Snapshot copies.  Data ONTAP 7-Mode volumes cannot be used in clustered Data ONTAP systems, and vice versa. 34


FLEXVOL VOLUMES Clustered Data ONTAP systems use flexible volumes differently than Data ONTAP 7-Mode and Data ONTAP 7G systems do. Because Data ONTAP clusters are inherently flexible, largely due to the volume move capability, volumes are deployed as freely as UNIX directories and Windows folders are deployed to separate logical groups of data. You can perform these actions:    

Create and delete volumes. Mount and unmount volumes. Move volumes around. Back up volumes.

To take advantage of this flexibility, cluster deployments typically use many more volumes than traditional Data ONTAP 7G deployments use. In a high-availability (HA) pair, aggregate and volume limits apply to each node individually, thus, the overall limit for the pair is effectively doubled.

3-37




Infinite Volumes Definition Applications

Applications

OnCommand System Manager or Cluster Shell

NFSv3, NFSv4.1, pNFS, CIFS

Clustered Data ONTAP 8.3 Operating System Single Namespace, Single Storage Pool Up to 2 Billion Files (16-Terabyte [TB] Maximum File Size) NAS Container (More than 20 Petabytes [PB] of Raw Capacity)

Two to Ten Nodes in a Cluster

35


INFINITE VOLUMES: DEFINITION Infinite volumes are boundless, easily istered, scalable containers that exceed the current Data ONTAP limits for FlexVol capacity. Infinite volumes are ideal for customers who need a single storage pool in a single namespace across an entire cluster. Infinite volumes consist of many FlexVol volumes. However, these volumes are presented to the storage as a single manageable volume.   

The constituents of an infinite volume are standard FlexVol volumes. Individual constituents are not managed by the storage . Although the constituents are not visible in the default displays, the can choose options to view the constituents.

To learn more about infinite volumes, refer to the Clustered Data ONTAP 8.3 Infinite Volumes Management Guide.

3-38




FlexClone Volumes

cluster1::> volume clone create -vserver svm1 -flexclone vol1clone -parent-volume svm_vol1

Aggregate

FlexVol Volume

Snapshot Copy of svm_vol1 Parent

vol1clone

36


FLEXCLONE VOLUMES A FlexClone volume is a point-in-time, space-efficient, writable copy of the parent volume.  

A FlexClone volume is a fully functional standalone volume. If changes are made to the parent volume after the FlexClone volume is created, those changes are not reflected in the FlexClone volume, and changes to the FlexClone volume are not reflected in the parent volume.

FlexClone volumes are created in the same SVM (which was previously known as a Vserver) and aggregate as the parent volume, and FlexClone volumes share common blocks with the parent volume.  

While a FlexClone copy of a volume exists, the parent volume cannot be deleted or moved to another aggregate. You can sever the connection between the parent and the FlexClone volume by executing a split operation. A FlexClone split causes the FlexClone volume to use its own disk space, but the FlexClone split enables you to delete the parent volume and to move the parent or the FlexClone volume to another aggregate.

To manage cloning on a cluster, use the volume clone command.

3-39




Load-Sharing and Data Protection Mirror Copies

RW

DP

Load-sharing and data protection mirror copies are similar in concept, but the type determines how the mirror copy is used and maintained.

37

RW

LS

LS

LS


LOAD-SHARING AND DATA PROTECTION MIRROR COPIES Mirror copies are read-only volumes.  

Each mirror copy has an association read/write volume. Each mirror copy is labeled as a load-sharing mirror copy or a data protection mirror copy.

Mirror copies are copies of read/write volumes.  

Mirror copies are synchronized with the read/write source volumes only if an keeps the mirror copies synchronized through manual replication or scheduled (automated) replication. Generally, data protection mirror copies do not need to be as up-to-date as load-sharing mirror copies, because they serve different purposes.

A replication schedule can be associated with each mirror copy. The schedule determines when (cron) or how often (interval) replications are performed on the mirror copy.   

All load-sharing mirror copies of a volume are treated as a unified group and use the same schedule. The schedule is enforced by the UI, so if you choose a different schedule for one load-sharing mirror copy, the other load-sharing mirror copies of that volume are automatically changed to match. Each data protection mirror copy is independent; it does not have to use the same schedule as other data protection mirror copies.

Load-sharing mirror relationships stay within the SVM of the source volume. Data protection mirror copies can be within a SVM, between SVMs within the cluster, or between SVMs of two different clusters.

3-40




SnapMirror and SnapVault Relationships To manage a data protection mirror or SnapVault relationship, you must assign a policy and a schedule to the relationship. SnapMirror Policy SnapMirror Relationship RW

38

DP

Schedule


SNAPMIRROR AND SNAPVAULT RELATIONSHIPS When a SnapMirror and SnapVault relationship is created, an optional update schedule is applied. The cron job schedule is normally created to control the frequency of the SnapMirror or SnapVault update. You use a policy to maximize the efficiency of the transfers to the backup secondaries and to manage the update operations. A SnapMirror and SnapVault policy consists of the following: Vserver name: The SVM for the policy Policy name: Must be unique within the cluster or SVM Policy type: Vault, async-mirror, mirror-vault, sync-mirror Comment: Text comment Tries: Number of times to attempt each transfer Transfer-priority: Normal or low Ignore-atime: Whether incremental transfers ignore files that have had only their access time changed (applies only to vault relationships) Restart: Defines the behavior if an interrupted transfer exists Is-network-compression-enabled: Enables or disables network compression for transfers FlexVol volumes data protection mirror and SnapVault relationships and policies. Infinite volumes only data protection mirror relationships and policies.

3-41




Cluster s use snapmirror policy commands to create and manage all data protection mirror and SnapVault policies. SVM s use the same commands to create and manage all data protection mirror and SnapVault policies within SVMs. All policy-management commands (except for the snapmirror policy show command) must be run on the SVM that contains the destination volume.

3-42




Lesson 4

Cluster and SVM Peering

39


LESSON 4: CLUSTER AND SVM PEERING

3-43




Ken Asks

How can I get the SVMs in my two clusters to replicate data?

40


KEN ASKS

3-44




Cluster and SVM Peering  Before you can set up a SnapMirror or SnapVault relationship between volumes in different clusters or different SVMs, a peering relationship must exist.

 A peer relationship between two clusters or SVMs enables the backup and recovery of data on clusters or SVMs.  Data ONTAP 8.3 s up to 255 cluster peer relationships.

41


CLUSTER AND SVM PEERING If you establish cluster peer relationships, you can create data protection mirroring relationships from one cluster to another, and you can manage the jobs on a remote cluster. Cluster peer authentication Before creating a cluster peer relationship, the s of the two clusters that will be peers agree on a text string, which is called a phrase, that they will use to authenticate one cluster to the other.  

The phrase is used when the relationship from the first cluster to the second cluster is created. The phrase is used again when the relationship from the second cluster to the first cluster is created.

NOTE: The phrase is not exchanged on the network by the Data ONTAP operating system. However, when Data ONTAP creates the cluster peer relationship, each cluster in the relationship recognizes the phrase.   

3-45

When you create the cluster peer relationship from the first cluster to the second cluster, the first cluster waits for the of the second cluster to create the cluster peer relationship. The of the second cluster must create the cluster peer relationship before the waiting period expires, within one hour by default. If the cluster peer relationship is not created from the second cluster to the first cluster before the waiting period expires, the cluster peer relationship is not created, and the s must restart the process.




Cluster Peering and SVM Peering Compare Cluster and SVM Peering Cluster Peering

SVM Peering

Cluster peering enables data to be backed up across cluster volumes or SVM volumes.

SVM names must be unique.

The relationship can be authenticated or unauthenticated.

SVM peering can be intercluster or intracluster.

Cluster peering requires an intercluster Intercluster peering requires cluster network. peering to be functional first.

Peering network topologies should be designed for fault tolerance. 42


CLUSTER PEERING AND SVM PEERING: COMPARE CLUSTER AND SVM PEERING

3-46




Cluster Peering Requirements Requirements of a Cluster Peer (1 of 2)

 Cluster time must be synchronized within 300 seconds. (Clusters can be in different time zones.)  At least one intercluster LIF must be created on every node in the cluster.  Every intercluster LIF requires an IP address that is dedicated for intercluster replication.

43


CLUSTER PEERING REQUIREMENTS: REQUIREMENTS OF A CLUSTER PEER (1 OF 2) Requirements for a cluster peer relationship:   

3-47

Cluster time must be synchronized within 300 seconds (clusters can be in different time zones). At least one intercluster LIF must be created on every node in the cluster. Every intercluster LIF requires an IP address that is dedicated for intercluster replication.




Cluster Peering Requirements Requirements of a Cluster Peer (2 of 2)

 All paths on a node that is used for intercluster networking should have equal performance characteristics.  The intercluster network must provide connectivity among all intercluster LIFs on all nodes in the cluster peers.  Every intercluster LIF on every node in a cluster must be able to connect to every intercluster LIF on every node in the peer cluster.

44


CLUSTER PEERING REQUIREMENTS: REQUIREMENTS OF A CLUSTER PEER (2 OF 2) Requirements for a cluster peer relationship:    

3-48

The correct maximum transmission unit (MTU) value must be used. All paths on a node that is used for intercluster networking should have equal performance characteristics. The intercluster network must provide connectivity among all intercluster LIFs on all nodes in the cluster peers. Every intercluster LIF on every node in a cluster must be able to connect to every intercluster LIF on every node in the peer cluster.




Intercluster Peering Network Connections Multiple T Connections, One LIF

Intercluster LIF

To send data, Data ONTAP establishes a minimum of 12 intercluster T connections.

12 T connections that share one LIF

Intercluster LIF

45


INTERCLUSTER PEERING NETWORK CONNECTIONS: MULTIPLE T CONNECTIONS, ONE LIF In clustered Data ONTAP, the number of intercluster LIFs determines the number of T connections that are established between the source and destination node for SnapMirror technology. T connections are not created per volume or per relationship. 



 

that, starting with the clustered Data ONTAP 8.2 operating system, Data ONTAP establishes at least 12 intercluster T connections for sending data. This is true even if the source and destination nodes have only one intercluster LIF. Enough connections are created so that all intercluster LIFs on both the source and destination nodes are used. If the source node, the destination node, or both nodes are configured with two intercluster LIFs, then Data ONTAP establishes 12 T connections for sending data. However, both connections do not use the same LIFs. Instead, one connection uses one LIF pair, and the other connection uses the other LIF pair. It is not possible to select a specific LIF pair to use for a certain T connection; Data ONTAP manages the LIF pairs automatically. After scaling past 12 intercluster LIFs on a node, Data ONTAP creates additional intercluster T connections. Enough connections are created so that all intercluster LIFs are used. The creation of additional intercluster T connections continues, as more intercluster LIFs are added to either the source node or the destination node. Currently, a maximum of 24 intercluster connections are ed for SnapMirror technology on a single node in Data ONTAP.

Best practice To ensure operational consistency, the same number of intercluster LIFs can be configured on both the source node and destination node. However, this configuration is not required. Multiple intercluster LIFs can be created to enable active-active multipathing across multiple physical paths. For example, if a node is configured with four 1-gigabit Ethernet (GbE) ports for intercluster replication, then four intercluster LIFs are required. One LIF is assigned to each port, to ensure that all paths are used to provide bandwidth beyond just one 1-GbE link. 3-49




Cluster Peering Topologies Cluster 1

Cluster 2 Intercluster LIF

Intercluster LIF

 An intercluster network must be created.  Intercluster replication occurs between two clusters on the intercluster network only.

46


CLUSTER PEERING TOPOLOGIES To ensure constant reliable communication, the best cluster peering topology is full-mesh connectivity. Full mesh means that all intercluster LIFs in a cluster can communicate with all intercluster LIFs on all remote clusters. To enable intercluster communication, an intercluster network must be created. Intercluster LIFs must be created for this explicit purpose. Intercluster replication occurs between two clusters on the intercluster network only. The IP addresses that are used for the intercluster LIFs can reside on the same subnet as data LIFs, or they can reside in a different subnet. When you create an intercluster LIF, you automatically create an intercluster routing group on that node. You can assign intercluster LIFs to ports that have the data role, or you can dedicate ports to the intercluster role.

3-50




SVM Peering and Cluster Peering

SVM1

SVM Peer Relationship SVM3

SVM2

SVM4

Cluster 1

Cluster 2

Cluster Peer Relationship 47


SVM PEERING AND CLUSTER PEERING SVM peering is the act of connecting two SVMs, to allow replication to occur between them. Before you create SnapMirror relationships between a pair of SVMs, an SVM peer relationship must exist between the pair of SVMs.   

These SVMs can be local (intracluster) or remote (intercluster). SVM peering is a permission-based mechanism. SVM peering is a one-time operation that must be performed by the cluster s.

The SVM peer infrastructure enables you to set up a backup and recovery mechanism between SVMs. You can set up a mirroring relationship at the volume level between peered SVMs. If a volume in the SVM becomes unavailable, the cluster or SVM can configure the respective mirrored volume of the peered SVM to serve data. One SVM can be peered with multiple SVMs within a cluster or across clusters. You can set up only SnapMirror data protection and SnapVault (XDP) relationships by using the SVM peer infrastructure.

3-51




Firewall Requirements for Intercluster SnapMirror Technology Open the following ports on the intercluster network between all source and destination nodes:  Port 11104  Port 11105 Clustered Data ONTAP uses port 11104 to manage intercluster communication sessions; it uses port 11105 to transfer data.

48


FIREWALL REQUIREMENTS FOR INTERCLUSTER SNAPMIRROR TECHNOLOGY

3-52




Lesson 5

Implementing Cluster and SVM Peering

49


LESSON 5: IMPLEMENTING CLUSTER AND SVM PEERING

3-53




Configure Cluster Peer Relationships Are there existing peer relationships between SVMs or clusters? cluster1::> cluster peer show This table is currently empty.

cluster1::> vserver peer show There are no SVM peer relationships.

50


CONFIGURE CLUSTER PEER RELATIONSHIPS

3-54




Configure Cluster Peer Relationships Create Failover Groups for Intercluster Networking

This command creates the failover group and assigns port e0c to it.

cluster1::> network interface failover-groups create -vserver cluster1 -failover-group failover_cluster1 -targets cluster1-01:e0c cluster1::> network interface failover-groups addtargets -vserver cluster1 -failover-group failover_cluster1 -targets cluster1-01:e0d This command adds port e0d to the failover group. Note that you used the vserver (cluster1) in these commands.

51


CONFIGURE CLUSTER PEER RELATIONSHIPS: CREATE FAILOVER GROUPS FOR INTERCLUSTER NETWORKING

3-55




Configure Cluster Peer Relationships Perform the Same Steps on Node2 of the Cluster

You must create a failover group with a unique name: cluster1::> network interface failover-groups create -vserver cluster1 -failover-group failover_cluster1-2 -targets cluster1-02:e0c cluster1::> network interface failover-groups addtargets -vserver cluster1 -failover-group failover_cluster1-2 -targets cluster1-02:e0d

52


CONFIGURE CLUSTER PEER RELATIONSHIPS: PERFORM THE SAME STEPS ON NODE2 OF THE CLUSTER

3-56




Configure Cluster Peer Relationships Create an Intercluster LIF on Each Node in the Cluster (1 of 2)

cluster1::> network interface create -vserver cluster1 -lif failover_cluster1 -role intercluster -home-node cluster1-01 -home-port e0c -address 192.168.1.50 -netmask 255.255.255.0 -failovergroup failover_cluster1 -failover-policy localonly -firewall-policy intercluster

This command creates the intercluster LIF on node cluster1-01.

53


CONFIGURE CLUSTER PEER RELATIONSHIPS: CREATE AN INTERCLUSTER LIF ON EACH NODE IN THE CLUSTER (1 OF 2)

3-57




Configure Cluster Peer Relationships Create an Intercluster LIF on Each Node in the Cluster (2 of 2)

This command creates the intercluster LIF on node cluster1-02: cluster1::> network interface create -vserver cluster1 -lif failover_cluster1-2 -role intercluster -home-node cluster1-02 -home-port e0c -address 192.168.1.60 -netmask 255.255.255.0 -failover-group failover_cluster1 -failover-policy local-only -firewall-policy intercluster

54


CONFIGURE CLUSTER PEER RELATIONSHIPS: CREATE AN INTERCLUSTER LIF ON EACH NODE IN THE CLUSTER (2 OF 2)

3-58




Configure Cluster Peer Relationships Check the Network Interface Configuration

 that the intercluster LIFs were created properly: cluster1::> network interface show -role intercluster

 that the intercluster LIFs are configured to be redundant: cluster1::> network interface show -role intercluster -failover

55


CONFIGURE CLUSTER PEER RELATIONSHIPS: CHECK THE NETWORK INTERFACE CONFIGURATION Check your work. that the intercluster LIFs were created properly: cluster1::> network interface show -role intercluster

SVM

Logical

Status

Current

Is

Interface

/Oper

------ ----------

Network Address/Mask

Node

Current Port

Home

------ ----

---------- -------------

--------

up/up

cluster1-0

e0c

true

cluster1-02

e0c

true

cluster1 failover_cluster1

192.168.1.50/24

failover_cluster1-2 up/up 192.168.1.60/24 2 entries were displayed. that the intercluster LIFs are configured to be redundant:

3-59




cluster1::> network interface show -role intercluster –failover Logical

Home

Failover

Failover Vserver

Interface

Node:Port

Policy

Group --------

----------

--------------

----------

failover_cluster1 cluster1-01:e0c

local-only

failover_cluster1

Failover Targets: cluster1-01:e0c,

cluster1-01:e0d

cluster1

failover_cluster1-2 cluster1-02:e0c local-only Failover Targets: cluster1-02:e0c 2 entries were displayed.

3-60




failover_cluster1

Configure Cluster Peer Relationships Check for Network Routes

 Display the routes in the cluster: cluster1::> network route show

 Create an intercluster route: cluster1::> network route create -vserver cluster1 -destination 0.0.0.0/0 -gateway 192.168.1.1 -metric 40 cluster1::> network route show

Important: The destination cluster needs to be configured with intercluster LIFs. 56


CONFIGURE CLUSTER PEER RELATIONSHIPS: CHECK FOR NETWORK ROUTES Display the routes in the cluster: cluster1::> network route show Vserver

Destination

Gateway

Metric

------------------- --------------- --------------- -----Cluster1

0.0.0.0/0

192.168.0.1

20

Create an intercluster route: cluster1::> network route create -vserver svm1 -destination 0.0.0.0/0 gateway 192.168.1.1 -metric 40 cluster1::> network route show Vserver

Destination

Gateway

Metric

------------------- --------------- --------------- -----cluster1 0.0.0.0/0

192.168.0.1

20

0.0.0.0/0

192.168.1.1

40

svm1 2 entries were displayed. Important: These steps need to be performed on the peer cluster.

3-61




Configure Cluster Peer Relationships Check the Network Interfaces on the Peer Cluster

In this example, the second cluster is a single-node cluster.

 Check the cluster2 network ports: cluster2::> network port show

 Create a failover group on cluster2: cluster2::> network interface failover-groups create -vserver cluster2 -failover-group failover_cluster2 -targets cluster2-01:e0c

cluster2::> network interface failover-groups addtargets -vserver cluster2 -failover-group failover_cluster2 -targets cluster2-01:e0d

57


CONFIGURE CLUSTER PEER RELATIONSHIPS: CHECK THE NETWORK INTERFACES ON THE PEER CLUSTER In this example, cluster2 is a single-node cluster. Steps for cluster2 (with command output): cluster2::> network port show Speed (Mbps) Node

Port

IPspace

Broadcast Domain Link

MTU

/Oper

------ --------- ------------ ---------------- ----- ------- -----------cluster2-01 e0a

Default

Default

up

1500

auto/1000

e0b

Default

Default

up

1500

auto/1000

e0c

Default

Default

up

1500

auto/1000

e0d

Default

Default

up

1500

auto/1000

e0e

Default

Default

up

1500

auto/1000

e0f

Default

Default

down

1500

auto/10


3-62




Create a failover group on cluster2: cluster2::> network interface failover-groups create -vserver cluster2 failover-group failover_cluster2 -targets cluster2-01:e0c cluster2::> network interface failover-groups add-targets -vserver cluster2 -failover-group failover_cluster2 -targets cluster2-01:e0d

3-63




Configure Cluster Peer Relationships Create the Intercluster LIF on Cluster2

cluster2::> network interface create -vserver cluster2 -lif failover_cluster2 -role intercluster -home-node cluster2-01 -home-port e0c -address 192.168.1.51 -netmask 255.255.255.0 -failovergroup failover_cluster2 -failover-policy localonly -firewall-policy intercluster cluster2::> network interface show -role intercluster

cluster2::> network interface show -role intercluster -failover

58


CONFIGURE CLUSTER PEER RELATIONSHIPS: CREATE THE INTERCLUSTER LIF ON CLUSTER2 Steps for cluster2 (with command output): cluster2::> network interface create -vserver cluster2 -lif failover_cluster2 -role intercluster -home-node cluster2-01 -home-port e0c -address 192.168.1.51 -netmask 255.255.255.0 -failover-group failover_cluster2 -failover-policy local-only -firewall-policy intercluster

Check your work: cluster2::> network interface show -role intercluster Logical

Status

Network

Current

Current

Is

/Oper

Address/Mask

Node

Port

----------

--------------

--------

----

Interface Vserver

Home

-------

---------

cluster2 failover_cluster2 up/up

192.168.1.51/24

true 3-64




cluster2-01 e0c

cluster2::> network interface show -role intercluster –failover Logical Vserver

Home Interface

-------- ----------

----------

Failover

Failover

Node:Port

Policy

-----------

----------- --------

Group

cluster2 failover_cluster2

cluster2-01:e0c local-only

Failover Targets: cluster2-01:e0c, cluster2-01:e0d

3-65




failover _cluster2

Configure Cluster Peer Relationships Create the Network Route on the Peer Cluster

cluster2::> network route create -vserver cluster2 -destination 0.0.0.0/0 -gateway 192.168.1.1 -metric 40

that the network route was created correctly: cluster2::> network route show

59


CONFIGURE CLUSTER PEER RELATIONSHIPS: CREATE THE NETWORK ROUTE ON THE PEER CLUSTER Create the network route on cluster2 for the intercluster subnet: cluster2::> network route create -vserver cluster2 -destination 0.0.0.0/0 -gateway 192.168.1.1 -metric 40

Check your work: cluster2::> network route show Vserver

Destination

Gateway

Metric

------------------- --------------- --------------- -----cluster2 0.0.0.0/0

192.168.0.1

20

0.0.0.0/0

192.168.1.1

40


3-66




Configure Cluster Peer Relationships Create the Cluster Peer Relationship

Create the peer relationship between the two clusters: cluster1::> cluster peer create -peer-addrs 192.168.1.51 The IP address points to the IP address that is configured on the destination cluster.

You will be prompted for a phrase.

60


CONFIGURE CLUSTER PEER RELATIONSHIPS: CREATE THE CLUSTER PEER RELATIONSHIP Create the peer relationship between the two clusters. cluster1::> cluster peer create -peer-addrs 192.168.1.51 Notice: Choose a phrase of 8 or more characters. To ensure the authenticity of the peering relationship, use a phrase or sequence of characters that would be hard to guess. Enter the phrase: Confirm the phrase: Notice that the IP address in the previous command points to the IP address that was created on the destination cluster.

3-67




Configure Cluster Peer Relationships Configure the Peer Cluster

 Repeat the same step on the peer cluster: cluster2::> cluster peer create -peer-addrs 192.168.1.50,192.168.1.60 The two IP addresses point to the two IP addresses that are configured on the peer cluster.

 that cluster peering was created correctly: cluster1::> cluster peer show cluster2::> cluster peer show -instance

61


CONFIGURE CLUSTER PEER RELATIONSHIPS: CONFIGURE THE PEER CLUSTER Repeat the same step on the peer cluster: cluster2::> cluster peer create -peer-addrs 192.168.1.50,192.168.1.60 Notice: Choose a phrase of 8 or more characters. To ensure the authenticity of the peering relationship, use a phrase or sequence of characters that would be hard to guess. Enter the phrase: Confirm the phrase: Check your work: cluster2::> cluster peer show Peer Cluster Name

Cluster Serial Number Availability

Authentication

------------------- --------------------- -------------- ---------cluster1

1-80-000099

Available

ok

cluster1::> cluster peer show Peer Cluster Name

Cluster Serial Number Availability

Authentication

---------------------- --------------------- ------------- ---------cluster2

1-80-000099

Available

cluster1::> cluster peer show -instance

3-68




ok

Peer Cluster Name: cluster2 Remote Intercluster Addresses: 192.168.1.51 Availability of the Remote Cluster: Available Remote Cluster Name: cluster2 Active IP Addresses: 192.168.1.51 Cluster Serial Number: 1-80-000099 Remote Cluster Nodes: cluster2-01 Remote Cluster Health: true Unreachable Local Nodes: Address Family of Relationship: ipv4 Authentication Status istrative: use-authentication Authentication Status Operational: ok Last Update Time: 9/23/2014 12:58:53 cluster2::> cluster peer show -instance Peer Cluster Name: cluster1 Remote Intercluster Addresses: 192.168.1.50, 192.168.1.60 Availability of the Remote Cluster: Available Remote Cluster Name: cluster1 Active IP Addresses: 192.168.1.60, 192.168.1.50 Cluster Serial Number: 1-80-000099 Remote Cluster Nodes: cluster1-01, cluster1-02 Remote Cluster Health: true Unreachable Local Nodes: Address Family of Relationship: ipv4 Authentication Status istrative: use-authentication Authentication Status Operational: ok Last Update Time: 9/23/2014 15:06:48

3-69




SVM Peering SnapVault Peer Relationships

After you successfully configure the cluster peering, you can configure the peering of the SVMs.

SVM1

SVM2

62


SVM PEERING: SNAPVAULT PEER RELATIONSHIPS

3-70




SVM Peering Peer Relationships Between SVMs

 Set up a peer relationship between SVMs: cluster1::> vserver peer create -vserver svm1 -peer-vserver svm1_2 -peer-cluster cluster2 -applications snapmirror

 that the peer state is “peered”: cluster1::> vserver peer show-all

 The peer SVM must accept the peer request: cluster2::> vserver peer accept -vserver svm1_2 -peer-vserver svm1

63


SVM PEERING: PEER RELATIONSHIPS BETWEEN SVMS Set up a peer relationship between the SVMs: cluster1::> vserver peer create -vserver svm1 -peer-vserver svm1_2 -peercluster cluster2 -applications snapmirror cluster1::> vserver peer show-all Peer

Peer

Vserver

Vserver

State

---------

---------

svm1

svm1_2

Peering Peer Cluster

Applications

---------

---------------

---------------

initiated

cluster2

snapmirror

Check the state of the peering request from cluster2: cluster2::> vserver peer show-all Peer

Peer

Vserver

Vserver

State

Peer Cluster

Applications

---------

---------

---------

---------------

---------------

svm1_2

svm1

pending

3-71

Peering

cluster1




snapmirror

The peer SVM must accept the peer request: cluster2::> vserver peer accept -vserver svm1_2 -peer-vserver svm1 that the peer state is initiated: cluster1::> vserver peer show-all Peer

Peer

Peering

Vserver

Vserver

State

Peer Cluster

Applications

---------

---------

---------

---------------

---------------

svm1

svm1_2

peered

cluster2

snapmirror

cluster2::> vserver peer show-all Peer

Peer

Vserver

Vserver

State

Peer Cluster

Applications

---------

---------

---------

---------------

---------------

svm1_2

svm1

peered

cluster1

snapmirror

3-72

Peering




Lesson 6

Initial Transfer and Updates

64


LESSON 6: INITIAL TRANSFER AND UPDATES

3-73




The Initial Transfer To initialize the data protection mirror copy, use the snapmirror initialize command.

65


THE INITIAL TRANSFER The initial transfer is a complete backup of a primary storage volume to a volume on the secondary system. After the initial transfer successfully completes, subsequent transfers contain only the changes that were made to the primary data since the previous transfer.

3-74




Replication Updates  You can perform replication updates manually.  In the next module you learn how to implement a SnapMirror solution. SnapMirror Policy SnapMirror Relationship RW

66

DP


REPLICATION UPDATES

3-75




Schedule

References  NetApp University courses:  Clustered Data ONTAP 8.3 istration

 NetApp Documentation:    

Clustered Data ONTAP 8.3 Data Protection Guide Clustered Data ONTAP 8.3 Commands: Manual Page Reference Clustered Data ONTAP 8.3 Infinite Volumes Management Guide NetApp Technical Report TR-4037 Introduction to NetApp Infinite Volume

 NetApp Technical Report TR-4210 Operational How-To Guide: NetApp Snapshot Management

67


REFERENCES

3-76





I learned the terminology that is used for NetApp replication technologies. I now understand use cases for NetApp replication solutions. Now I know how to design and configure peering relationships between clusters and SVMs.

68



3-77




Exercise

Module 3: NetApp Replication Technologies


69



3-78




Module 4

Using SnapMirror for Data Protection

1


MODULE 4: USING SNAPMIRROR FOR DATA PROTECTION

4-1

Data ONTAP Data Protection istration: Using SnapMirror for Data Protection




 Describe SnapMirror design criteria  Configure and implement SnapMirror relationships  Use SnapMirror technology to recover data  Use the Data ONTAP CLI to manage SnapMirror relationships

2


MODULE OBJECTIVES

4-2




Lesson 1

SnapMirror Technology

3


LESSON 1: SNAPMIRROR TECHNOLOGY

4-3




Volume-Level Replication Source Volume

Snapshot Copies

SnapMirror

Mirrored Volume

WAN SnapMirror

FlexClone Volumes

4

Mirrored Volume

Tape


VOLUME-LEVEL REPLICATION SnapMirror Uses and Benefits SnapMirror technology in clustered Data ONTAP provides asynchronous volume-level replication based on a configured replication update interval. SnapMirror technology uses NetApp Snapshot technology as part of the replication process. Clustered Data ONTAP 8.3 provides the following replication capabilities:  

4-4

Data protection mirror copies: These copies provide replication to create a backup copy within the same cluster (intracluster) or to create a disaster recovery copy in a different cluster (intercluster). Load-sharing (LS) mirror copies: These copies provide replication from one volume to multiple volumes in the same cluster to distribute a read-only workload across a cluster.




SnapMirror Replication Basics When the scheduler triggers a replication update, the following operations are performed: 1. A new Snapshot copy is created on the source volume. 2. The block-level difference between the new Snapshot copy and the last replication Snapshot copy is determined and then transferred to the destination volume. 3. When the transfer is complete, the new Snapshot copy exists on the destination volume.

5


SNAPMIRROR REPLICATION BASICS Basics of SnapMirror Replication When the scheduler triggers a replication update, the following operations are performed: 1. A new Snapshot copy is created on the source volume. 2. The block-level difference between the new Snapshot copy and the last replication Snapshot copy is determined and then transferred to the destination volume. This transfer includes other Snapshot copies that were created between the last replication Snapshot copy and the new copy. 3. When the transfer is complete, the new Snapshot copy exists on the destination volume. A SnapMirror destination volume is available for read-only access, if it is shared using the CIFS protocol and is exported using the NFS protocol. A logical unit number (LUN) in the replicated volume can be made available to a client that s connection to read-only LUNs. Replication occurs at the volume level. Qtrees can be created in clustered Data ONTAP and replicated with the replicated volume; however, individual qtrees cannot be separately replicated.

4-5




SnapMirror and FlexClone Technology

1. Create an unscheduled Snapshot copy at the source. FlexClone Volume Snapshot Copies

2.

6

Perform a SnapMirror update to replicate the unscheduled Snapshot copy to the destination.

3.

Use the unscheduled Snapshot copy as the base for the FlexClone volume.


SNAPMIRROR AND FLEXCLONE TECHNOLOGY A NetApp FlexClone volume is a writable point-in-time clone of a FlexVol volume. A FlexClone volume shares data blocks with the parent volume and stores only new data or changes that are made to the clone. FlexClone technology also enables you to create a writable volume from a read-only SnapMirror destination without interrupting the SnapMirror replication process or production operations.

4-6




SnapMirror and NetApp Storage Efficiency  If a SnapMirror source volume is in a deduplicated state, then the destination volume is in a deduplicated state. SnapMirror software does not inflate the deduplicated data during the transfer.  Likewise, if a SnapMirror source volume is in a compressed state, then the destination volume is in a compressed state. SnapMirror does not uncompress the source data before or during the transfer to the destination volume. Data is replicated in a compressed state.

7


SNAPMIRROR AND NETAPP STORAGE EFFICIENCY SnapMirror technology s NetApp deduplication and compression storage efficiency technologies. If a SnapMirror source volume is in a deduplicated state, then the destination volume is in a deduplicated state. SnapMirror does not inflate the deduplicated data during the transfer. If a SnapMirror source volume is in a compressed state, then the destination volume is in a compressed state. SnapMirror does not uncompress the source data before or during the transfer to the destination volume. Data is replicated in a compressed state.

4-7




SnapMirror and Volume Move

A data protection source or destination volume can be moved nondisruptively to another node in the cluster. The SnapMirror relationship does not have to be reconfigured on either volume when the volume move is performed.

8


SNAPMIRROR AND VOLUME MOVE You can move a data protection source or destination volume nondisruptively to another node in the cluster. The SnapMirror relationship does not have to be reconfigured on either volume when the volume move is performed.

4-8




SnapMirror and Volume Autosize

When volume autosize increases the size of the source volume of a SnapMirror relationship, the destination volume automatically increases in size. This feature is available only with FlexVol volumes, not infinite volumes. 9


SNAPMIRROR AND VOLUME AUTOSIZE

4-9




Version-Flexible SnapMirror Relationships  Version-flexible SnapMirror technology allows relationships to exist where the source and destination clusters are running different major versions of Data ONTAP.  Version-flexible SnapMirror relationships allow disaster recovery and additional retention on the same destination volume, enabling you to save on the cost of additional storage, network bandwidth, and management.

10


VERSION-FLEXIBLE SNAPMIRROR RELATIONSHIPS Before you create version-independent SnapMirror relationships, consider these guidelines:  

4-10

The schedule frequency for replicating volumes must be greater than 60 minutes. Large to medium-sized file workloads are best suited for replication using version-independent SnapMirror relationships.




Lesson 2

Implementing SnapMirror Relationships

11


LESSON 2: IMPLEMENTING SNAPMIRROR RELATIONSHIPS

4-11




SnapMirror Intercluster Configuration Typical Workflow to Create an Intercluster Configuration

 Check for a SnapMirror license.

 Create a destination volume.  Select an update schedule.  Select the mirror policy.  Initialize the SnapMirror relationship.  Monitor the SnapMirror relationship.

(This lesson demonstrates this workflow.)

12


SNAPMIRROR INTERCLUSTER CONFIGURATION: TYPICAL WORKFLOW TO CREATE AN INTERCLUSTER CONFIGURATION

4-12




SnapMirror Intercluster Configuration Check Licenses and Create a New Destination Volume

 Check to see if the SnapMirror licenses are installed: cluster1::> license show -package SnapMirror

 Create a new destination volume, which must be type DP (data protection): cluster2::> volume create -vserver svm1_2 -volume svm1_2_vol2 -aggregate aggr1 -size 200m -securitystyle ntfs -type dp

13


SNAPMIRROR INTERCLUSTER CONFIGURATION: CHECK LICENSES AND CREATE A NEW DESTINATION VOLUME You can create a SnapMirror relationship between volumes on different clusters for disaster recovery. You must create a new destination volume for creating a SnapMirror relationship. Check to see if you have the SnapMirror licenses installed: cluster1::> license show -package SnapMirror Serial Number: 1-80-000099 Owner: cluster1 Package

Type

Description

Expiration

-----------------

-------

---------------------

-------------

SnapMirror

site

SnapMirror License

-

You can use a slightly different command to view the SnapMirror license on cluster2: cluster2::> license show -instance -package SnapMirror (system license show) Serial Number: 1-80-000099 Package: SnapMirror Owner: cluster2 Expiration Date: Description: SnapMirror License

4-13




Type: site Legacy: no Customer ID: 1-82-0000000000000000000000000 The licenses are in place. Create a new destination volume. The destination volume must be type DP: cluster2::> volume create -vserver svm1_2 -volume svm1_2_vol2 -aggregate aggr1 -size 200m -security-style ntfs -type dp cluster2::> volume show -volume svm1_2_vol2 Vserver

Volume

Aggregate

State

Type Size

Available Used%

--------- ------------ ----------- -------- ---- ----- --------

-----

svm1_2

0%

4-14

svm1_2_vol2

aggr1

online

DP




200MB 199.88MB

SnapMirror Intercluster Configuration Create a Data Protection Mirror Relationship

 Create a data protection mirror relationship by using the 5min job schedule: cluster2::> snapmirror create -destination-path svm1_2:svm1_2_vol2 -source-path svm1:svm1_vol3 -type DP -schedule 5min

 that the relationship was created: cluster2::> snapmirror show

14


SNAPMIRROR INTERCLUSTER CONFIGURATION: CREATE A DATA PROTECTION MIRROR RELATIONSHIP Create a data protection mirror relationship by using the snapmirror create command. Use the 5min job schedule. that cluster peering and storage virtual machine (SVM) peering were configured in the previous module: cluster2::> snapmirror create -destination-path svm1_2:svm1_2_vol2 source-path svm1:svm1_vol3 -type DP -schedule 5min Operation succeeded: snapmirror create for the relationship with destination "svm1_2:svm1_2_vol2". that the relationship was created: cluster2::> snapmirror show

Source Path

Progress Destination Type Updated

Mirror Path

------

--------

------- ----------

Relationship State

Total Status

Last Progress

Healthy

-------

--------

-----------

svm1:svm1_vol1 XDP

svm1_2:svm1_2_vol1 Snapmirrored Idle

4-15

-




true

-

svm1:svm1_vol3 DP

svm1_2:svm1_2_vol2 Uninitialized Idle

-


4-16




true

-

SnapMirror Intercluster Configuration Initialize the SnapMirror Relationship

 Initialize the data protection mirror: cluster2::> snapmirror initialize –destination -path svm1_2:svm1_2_vol2

 that the initialization occurred correctly: cluster2::> snapmirror show cluster2::> snapmirror show -instance

15


SNAPMIRROR INTERCLUSTER CONFIGURATION: INITIALIZE THE SNAPMIRROR RELATIONSHIP Initialize the data protection mirror: cluster2::> snapmirror initialize -destination-path svm1_2:svm1_2_vol2 Operation is queued: snapmirror initialize of destination "svm1_2:svm1_2_vol2". that the initialization occurred correctly: cluster2::> snapmirror show Progress Source

Destination

Mirror

Relationship

Total

Path

State

Status

Progress Healthy

Path

Type Updated

------

-------- -----------

------- --------------

Last

-------

--------

true

-

svm1:svm1_vol1 XDP


4-17

-




svm1:svm1_vol3 DP


-

true

-

2 entries were displayed. cluster2::> snapmirror show -instance Source Path: svm1:svm1_vol3 Destination Path: svm1_2:svm1_2_vol2 Relationship Type: DP Relationship Group Type: none SnapMirror Schedule: 5min SnapMirror Policy Type: async-mirror SnapMirror Policy: DPDefault Tries Limit: Throttle (KB/sec): unlimited Mirror State: Snapmirrored Relationship Status: Idle File Restore File Count: File Restore File List: Transfer Snapshot: Snapshot Progress: Total Progress: Network Compression Ratio: Snapshot Checkpoint: Newest Snapshot: snapmirror.1383c700-4345-11e497f5-0050560140c1_2147484679. 2014-09-24_131000 Newest Snapshot Timestamp: 09/24 13:11:15 Exported Snapshot: snapmirror.1383c700-4345-11e497f5-0050560140c1_2147484679. 2014-09-24_131000 Exported Snapshot Timestamp: 09/24 13:11:15 Healthy: true Unhealthy Reason: Constituent Relationship: false Destination Volume Node: cluster2-01 Relationship ID: ff573bda-43ea-11e4-97f50050560140c1 Current Operation ID: 4-18




Transfer Type: Transfer Error: Current Throttle: Current Transfer Priority: Last Transfer Type: update Last Transfer Error: Last Transfer Size: 72KB Last Transfer Network Compression Ratio: 1:1 Last Transfer Duration: 0:0:2 Last Transfer From: svm1:svm1_vol3 Last Transfer End Timestamp: 09/24 13:10:04 Progress Last Updated: Relationship Capability: 8.2 and above Lag Time: 0:0:55 Number of Successful Updates: 1 Number of Failed Updates: 0 Number of Successful Resyncs: 0 Number of Failed Resyncs: 0 Number of Successful Breaks: 0 Number of Failed Breaks: 0 Total Transfer Bytes: 212992 Total Transfer Time in Seconds: 5 2 entries were displayed.

4-19




SnapMirror and Snapshot Copies Check the Snapshot copies on the source and destination volumes: cluster1::> volume snapshot show -volume svm1_vol3

cluster2::> volume snapshot show –volume svm1_2_vol2 SnapMirror maintains a history of one SnapMirror Snapshot copy on the source volume and two copies on the destination volume.

16


SNAPMIRROR AND SNAPSHOT COPIES Best Practice: that SnapMirror updates are not scheduled to occur on the source volume at the same time as other Snapshot copies. Look at the Snapshot copies on the source and destination volumes: cluster1::> volume snapshot show -volume svm1_vol3 ---Blocks--Vserver

Volume Used%

Snapshot

--------

--------

-------------------------

svm1

svm1_vol3

Size ----------

Total% --------

snapmirror.1383c700-4345-11e4-97f5-2014-090050560140c1_2147484679.24_131500 32%

68KB

cluster2::> volume snapshot show -volume svm1_2_vol2

4-20




0%

---Blocks--Vserver

Volume Used%

Snapshot

Size

--------

--------

----------------------- --------

svm1_2

svm1_2_vol2

Total% --------

snapmirror.1383c700-4345-11e4-97f50050560140c1_2147484679.2014-09-24_131000 76KB

0%

35%

snapmirror.1383c700-4345-11e4-97f50050560140c1_2147484679.2014-09-24_131500 0B

0%


4-21




0%

Scheduling SnapMirror Updates Create a New Cron Job Schedule

 Create a new cron job schedule on the destination cluster: cluster2::> job schedule cron create Hourly_SnapMirror –hour all minute 0

This job runs at the top of every hour.

 that the job was created correctly: cluster2::> job schedule cron show

17


SCHEDULING SNAPMIRROR UPDATES: CREATE A NEW CRON JOB SCHEDULE SnapMirror updates are usually configured to run automatically. This example creates a schedule that is called Hourly_SnapMirror, which runs at the top of every hour (on the zero minute of every hour): cluster2::> job schedule cron create Hourly_SnapMirror -hour all -minute 0 that the job was created correctly: cluster2::> job schedule cron show Name

Description

----------------

---------------------------------------------------

5min

@:00,:05,:10,:15,:20,:25,:30,:35,:40,:45,:50,:55

8hour

@2:15,10:15,18:15

Hourly_SnapMirror

@:00

SnapVaultJob

@:10

daily

@0:10

hourly

@:05

weekly

Sun@0:15


4-22




Scheduling SnapMirror Updates Apply the New Schedule to the SnapMirror Relationship

 Apply the Hourly_SnapMirror schedule to an existing relationship: cluster2::> snapmirror modify -destination-path cluster2://svm1_2/svm1_2_vol2 -schedule Hourly_SnapMirror The SnapMirror jobs are now performed hourly.

 that the schedule is applied to the relationship: cluster2::> snapmirror show -destination-path cluster2://svm1_2/svm1_2_vol2 -fields schedule

18


SCHEDULING SNAPMIRROR UPDATES: APPLY THE NEW SCHEDULE TO THE SNAPMIRROR RELATIONSHIP The schedule can be applied to a SnapMirror relationship at the time of creation by using the –schedule option. The schedule can be applied to an existing relationship by using the snapmirror modify command and the –schedule option. Apply the Hourly_SnapMirror schedule to an existing relationship: cluster2::> snapmirror modify -destination-path cluster2://svm1_2/svm1_2_vol2 -schedule Hourly_SnapMirror Operation succeeded: snapmirror modify for the relationship with destination "svm1_2:svm1_2_vol2". cluster2::> snapmirror show -destination-path cluster2://svm1_2/svm1_2_vol2 -fields schedule source-path

destination-path

schedule

-------------- ------------------ ----------------svm1:svm1_vol3 svm1_2:svm1_2_vol2 Hourly_SnapMirror SnapMirror updates are now performed hourly.

4-23




Try This Using cluster1on your lab kit:

 Enter snapmirror set-options.  Are any of the options set?

 Enter man snapmirror set-options.  What can you say about snapmirror options?

19


TRY THIS

4-24




Ken Asks

How should I deploy load-sharing mirror relationships?

20


KEN ASKS

4-25




SnapMirror Load-Sharing Mirror Relationships Introduction

 SnapMirror load-sharing mirror copies can NAS only (CIFS and NFSv3).  Load-sharing mirror copies do not NFSv4 clients or SAN client protocol connections (FC, FCoE, or iSCSI).

21


SNAPMIRROR LOAD-SHARING MIRROR RELATIONSHIPS: INTRODUCTION SnapMirror load-sharing mirror copies increase performance and availability for NAS clients by distributing an SVM namespace root volume to other nodes in the same cluster and by distributing data volumes to other nodes in the cluster, to improve performance for large read-only workloads. NOTE: SnapMirror load-sharing mirror copies are capable of ing NAS only (CIFS/NFSv3). Loadsharing mirror copies do not NFSv4 clients or SAN client protocol connections (FC, FCoE, or iSCSI). Data ONTAP routes NFSv4 clients to the source of the load-sharing mirror for direct read and write access.

4-26




Implementing Load-Sharing SnapMirror Relationships Create Load-Sharing Mirror Destination Volumes

Two volumes, which are named svm1_root_lsm1 and svm1_root_lsm2, are created as load-sharing mirror destination volumes for the SVM root volume named svm1_root: cluster1::> volume create -vserver svm1 -volume svm1_root_lsm1 -aggregate n1_aggr1 -size 20m -type DP

cluster1::> volume create -vserver svm1 -volume svm1_root_lsm2 -aggregate n1_aggr1 -size 20m -type DP

22


IMPLEMENTING LOAD-SHARING SNAPMIRROR RELATIONSHIPS: CREATE LOADSHARING MIRROR DESTINATION VOLUMES   

Load-sharing mirror relationships can be managed only by the Data ONTAP CLI. Currently, load-sharing mirror relationships cannot be managed by using OnCommand System Manager. A group of load-sharing mirror destination volumes that replicate from the same source volume is called a load-sharing mirror set. When a load-sharing mirror set is created, each destination volume must be created in the appropriate aggregate, which creates the destination volumes with a type of data protection.

In this example, two volumes named svm1_root_lsm1 and svm1_root_lsm2 are created as load-sharing mirror destination volumes for the SVM root volume named svm1_root: cluster1::> volume create -vserver svm1 -volume svm1_root_lsm1 -aggregate n1_aggr1 -size 20m -type DP cluster1::> volume create -vserver svm1 -volume svm1_root_lsm2 -aggregate n1_aggr1 -size 20m -type DP

4-27




Implementing Load-Sharing SnapMirror Relationships Create Load-Sharing SnapMirror Relationships

A load-sharing SnapMirror relationship is created for each of the destination volumes, vs1_ls_a and vs1_ls_b, with an hourly update schedule: cluster1::> snapmirror create -source-path svm1:svm1_root -destination-path svm1:svm1_root_lsm1 -type LS -schedule hourly

cluster1::> snapmirror create -source-path svm1:svm1_root -destination-path svm1:svm1_root_lsm1 -type LS -schedule hourly

23


IMPLEMENTING LOAD-SHARING SNAPMIRROR RELATIONSHIPS: CREATE LOADSHARING SNAPMIRROR RELATIONSHIPS After all load-sharing mirror destination volumes are created, each SnapMirror relationship can be created with a type of LS. In this example, a load-sharing SnapMirror relationship is created for each of the destination volumes, vs1_ls_a and vs1_ls_b, with an hourly update schedule. cluster1::> snapmirror create -source-path svm1:svm1_root -destinationpath svm1:svm1_root_lsm1 -type LS -schedule hourly [Job 117] Job is queued: snapmirror create for the relationship with destination[Job 117] Job succeeded: SnapMirror: done cluster1::> snapmirror create -source-path svm1:svm1_root -destinationpath svm1:svm1_root_lsm2 -type LS -schedule hourly [Job 119] Job is queued: snapmirror create for the relationship with destination[Job 119] Job succeeded: SnapMirror: done

4-28




Implementing Load-Sharing SnapMirror Relationships Initialize a Load-Sharing SnapMirror Relationship

 Initialize a load-sharing SnapMirror relationship: cluster1::> snapmirror initialize-ls-set -sourcepath cluster1://svm1/svm1_root All destination volumes are initialized at once.

 that the initialization was performed: cluster1::> snapmirror show

 Manually update the load-sharing mirror relationship: cluster1::> snapmirror update-ls-set -source-path svm1:svm1_root

24


IMPLEMENTING LOAD-SHARING SNAPMIRROR RELATIONSHIPS: INITIALIZE A LOAD-SHARING SNAPMIRROR RELATIONSHIP 



You can update load-sharing mirror relationships manually or by setting the desired schedule in the – schedule option. For load-sharing mirror relationships, you do this by setting the desired schedule on any one of the destinations in the load-sharing mirror set. Data ONTAP automatically applies that schedule to all destinations in that load-sharing mirror set. A later change to the update schedule for any of the destination volumes in the load-sharing mirror set applies the new schedule to all volumes in that load-sharing mirror set. Therefore, in the previous example, the –schedule option was used only in the creation of the last relationship, which applied the schedule to both relationships.

Initialize the LS SnapMirror relationship: cluster1::> snapmirror initialize-ls-set -source-path cluster1://svm1/svm1_root [Job 120] Job is queued: snapmirror initializels-set for source "cluster1://svm1/svm1_root". that the initialization was performed: cluster1::> snapmirror show

4-29




Progress Source Path

Type Updated

Destination Path

Mirror Relationship Total State Status Progress

------------------- ------- ---------cluster1://svm1/svm1_root LS

---------

Last Healthy -------

cluster1://svm1/svm1_root_lsm1 Snapmirrored Transferring 696KB

false

cluster1://svm1/svm1_root_lsm2 Snapmirrored Transferring 696KB

false

-

2 entries were displayed. The initialize process is still transferring data. cluster1::> snapmirror show Source Path -----

Type Updated ----

Destination Path


------------ ------- ----------

Last Healthy

---------

-------

Idle

-

cluster1://svm1/svm1_root LS

cluster1://svm1/svm1_root_lsm1 Snapmirrored

true

cluster1://svm1/svm1_root_lsm2 Snapmirrored Idle

true

-

-

2 entries were displayed. You can update load-sharing mirror relationships on demand by using the snapmirror update-ls-set command. Data ONTAP updates all destination volumes for the LS set in one operation. cluster1::> snapmirror update-ls-set -source-path svm1:svm1_root [Job 121] Job is queued: snapmirror update-ls-set for source "cluster1://svm1/svm1_root".

4-30




Lesson 3

SnapMirror Failover and Resync

25


LESSON 3: SNAPMIRROR FAILOVER AND RESYNC

4-31




Ken Asks

Which steps should I take when a failure of the source volume occurs?

26


KEN ASKS

4-32




SnapMirror Initial Configuration

Clients

Cluster 2

Cluster 1

Source Volume

27

Scheduled Copies

Destination Volume


SNAPMIRROR INITIAL CONFIGURATION  

4-33

Client applications perform read/write actions only on the source volume. Data on the source volume is mirrored to the destination volume. The destination volume is configured for data protection, which makes the volume read-only, if clients are given access to it. The process of failing over and resynchronizing a SnapMirror relationship involves three Data ONTAP operations: break, reverse resync*, and update.




SnapMirror Failover and Resync Volume Disaster Recovery Workflow

When a source volume becomes unavailable for read/write access, you can perform a volume-level disaster recovery failover and resynchronization.

Cluster 2

Cluster 1

Source Volume

28

Scheduled Copies

Destination Volume


SNAPMIRROR FAILOVER AND RESYNC: VOLUME DISASTER RECOVERY WORKFLOW

4-34




SnapMirror Disaster Recovery Workflow Typical SnapMirror Disaster Recovery Workflow

 Activate the destination volume:

 the source volume status.  Quiesce and break the SnapMirror relationship.  the destination volume status.

 Configure the destination volume for data access.  Reactivate the original source volume:

 Resynchronize the source volume.  Update the source volume, if necessary.  Activate the source volume.

29


SNAPMIRROR DISASTER RECOVERY WORKFLOW: TYPICAL SNAPMIRROR DISASTER RECOVERY WORKFLOW Most of the configuration that is necessary for disaster recovery is performed prior to a failover. Thus, the actual steps that are required to fail over during a disaster recovery scenario are greatly reduced. Steps for a NAS environment: 1. Perform a SnapMirror break operation to fail over each volume. In clustered Data ONTAP, you can use wildcards to perform a SnapMirror operation on multiple volumes by using one command. The following example performs failover for all volumes in the destination SVM that is called vs5. Failover can be restricted to certain volumes by using part of the volume name in the command: cluster02::> snapmirror break -destination-path cluster02://vs5/* 2. If the volumes were mounted in the namespace, and CIFS shares and NFS export policies were created and applied, then clients have read-write access to the NAS data. 3. Redirect clients to the recovered storage. – – –

4-35

It is common practice for a disaster recovery system to have a different name than the source system. In disaster recovery failover scenarios, it is typical to change DNS name resolution or use DNS aliases to redirect clients to the names of the recovered storage systems. This enables CIFS shares to be accessible using the same UNC path name. NFS clients can also access the expected path. Alternatively, the failed source storage system might be removed from Active Directory. The recovery storage system might be removed and added again to Active Directory by using the same name as the source system. However, it can take time for this change to propagate through a large Active Directory environment.




Postfailover volume configuration Snapshot copy policies and NetApp storage efficiency policies cannot be assigned to volumes in a data protection state, so they must be assigned after failover. 1. If you are using the Data ONTAP Snapshot copy schedule, assign a Snapshot copy policy to the recovered volumes. In SAN environments, Snapshot copies are typically scheduled in the client. 2. If you are using NetApp storage efficiency technology, assign a storage efficiency policy to the recovered volumes.

4-36




SnapMirror Failover Quiesce the SnapMirror Relationship

 the source volume status: cluster1::> volume show -vserver svm1 -volume svm1_vol2 –fields state vserver volume state ------- --------- ------Perform this task from the destination svm1 svm1_vol2 offline cluster.

 Quiesce the SnapMirror relationship: cluster2::> snapmirror quiesce -source-path svm1:svm1_vol2 cluster2://svm1_2/svm1_svm1_vol2_mirror Operation succeeded: snapmirror quiesce for destination "svm1_2:svm1_svm1_vol2_mirror". 30


SNAPMIRROR FAILOVER: QUIESCE THE SNAPMIRROR RELATIONSHIP Steps for performing a SnapMirror volume recovery: 1. that the source volume is offline and unable to serve data. 2. Quiesce and break the SnapMirror relationship. After quiescing, future SnapMirror transfers are disabled. NOTE: The SnapMirror quiesce process allows transfers in progress to complete, but it disables future transfers for a mirror relationship.

4-37




Topics for Discussion

What occurs when you use the snapmirror quiesce command?

31


TOPICS FOR DISCUSSION

4-38




SnapMirror Failover Break the SnapMirror Relationship

 Break the SnapMirror relationship: cluster2::> snapmirror break -source-path svm1:svm1_vol2 cluster2://svm1_2/svm1_svm1_vol2_mirror

 the state of the SnapMirror relationship: cluster2::> snapmirror show -source-path svm1:svm1_vol2 -fields state The mirror state should be broken off.

32


SNAPMIRROR FAILOVER: BREAK THE SNAPMIRROR RELATIONSHIP The snapmirror break command should be performed from the destination cluster.

4-39




SnapMirror Failover Check the Destination Volume

the state of the destination volume: cluster2::> volume show -volume svm1_svm1_vol2_mirror –fields state vserver volume state ------- --------------------- -----svm1_2 svm1_svm1_vol2_mirror online This is almost ready for client access.

33


SNAPMIRROR FAILOVER: CHECK THE DESTINATION VOLUME

4-40




SnapMirror Failover Configure the Destination Volume for Data Access

that the destination volume has read/write access and that the volume settings match the source volume settings:  Thin provisioning  Deduplication

 Compression  Autogrow  Snapshot copy schedule

34


SNAPMIRROR FAILOVER: CONFIGURE THE DESTINATION VOLUME FOR DATA ACCESS

4-41




SnapMirror Failover Configure the Destination Volume for Data Access (1 of 2)

In a NAS environment:

 Mount the NAS volume into the namespace using the same junction path as the source volume.  Apply the correct access control lists (ACLs) to CIFS shares.

 Assign NFS export policies.  Apply quota rules, as necessary.  Redirect clients to the destination volume.  Remount NFS and CIFS shares on the clients.

35


SNAPMIRROR FAILOVER: CONFIGURE THE DESTINATION VOLUME FOR DATA ACCESS (1 OF 2)

4-42




SnapMirror Failover Configure the Destination Volume for Data Access (2 of 2)

In a SAN environment:

 Map the LUNs to the appropriate initiator group.  For iSCSI, create iSCSI sessions from the SAN host initiators to the SAN logical interfaces (LIFs).

 Perform a storage rescan on the host to detect the connected LUNs.

36


SNAPMIRROR FAILOVER: CONFIGURE THE DESTINATION VOLUME FOR DATA ACCESS (2 OF 2)

4-43




SnapMirror Resync Reactivate the Original Source Volume

When the source volume comes back online, you must resynchronize the latest data from the destination volume.

Cluster 2

Cluster 1

DP

RW Resync

Destination Volume

37

Source Volume


SNAPMIRROR RESYNC: REACTIVATE THE ORIGINAL SOURCE VOLUME

4-44




SnapMirror Resync Reactivate the Original Source Volume (1 of 4)

 Initiate the snapmirror resync command from the former source cluster: cluster1::> snapmirror resync -source-path svm1_2:svm1_svm1_vol2_mirror cluster1://svm1/svm1_vol2

 that the mirror state is shown as “Snapmirrored”: cluster1::> snapmirror show -source-path svm1_2:svm1_svm1_vol2_mirror

38


SNAPMIRROR RESYNC: REACTIVATE THE ORIGINAL SOURCE VOLUME (1 OF 4) Perform the snapmirror resync operation from the former source cluster (which is now the destination cluster): cluster1::> snapmirror resync -source-path svm1_2:svm1_svm1_vol2_mirror cluster1://svm1/svm1_vol2 Warning: All data newer than Snapshot copy snapmirror.db779627-503c-11e49bbc-123478563412_2147484681.2014-10-10_141000 on volume svm1:svm1_vol2 will be deleted. Do you want to continue? {y|n}: y Operation is queued: initiate snapmirror resync to destination "svm1:svm1_vol2". that the mirror state is shown as “Snapmirrored”: cluster1::> snapmirror show -source-path svm1_2:svm1_svm1_vol2_mirror

4-45




Progress Source Path -----

Type Updated ----

Destination Path


------------ ------- ----------

Last Healthy

---------

-------

Idle

-

svm1_2:svm1_svm1_vol2_mirror DP

svm1:svm1_vol2 Snapmirrored

true

4-46

-





 Check the current SnapMirror relationship: cluster1::> snapmirror show -source-path svm1_2:svm1_svm1_vol2_mirror -fields schedule,policy schedule policy DPDefault (This is deprecated output.) Notice the policy and schedule settings.

 At this point, the SnapMirror relationship is reversed from the original configuration. 39


SNAPMIRROR RESYNC: REACTIVATE THE ORIGINAL SOURCE VOLUME (2 OF 4) Check the current SnapMirror relationship: cluster1::> snapmirror show -source-path svm1_2:svm1_svm1_vol2_mirror fields schedule,policy source-path

destination-path

schedule

policy

------------------

----------------

--------

---------

svm1_2:svm1_svm1_vol2_mirror

svm1:svm1_vol2

-

DPDefault

4-47





If you plan to run in this (reversed) SnapMirror relationship for any length of time, consider specifying a SnapMirror policy and schedule that match the protection configuration of the original SnapMirror relationship.

40


SNAPMIRROR RESYNC: REACTIVATE THE ORIGINAL SOURCE VOLUME (3 OF 4)

4-48





To return the SnapMirror relationship back to its original state, you must perform the same steps that you performed for the failover. Cluster 1

Cluster 2 Quiesce and Break

RW Destination Volume

Cluster 2

Cluster 1 Resync

RW Source Volume

41

RW Source Volume

DP Destination Volume


SNAPMIRROR RESYNC: REACTIVATE THE ORIGINAL SOURCE VOLUME (4 OF 4) To return the SnapMirror relationship to its original state, perform the same steps that you performed for the failover. Quiesce the SnapMirror transfers: cluster1::> snapmirror quiesce -source-path svm1_2:svm1_svm1_vol2_mirror -destination-path svm1:svm1_vol2 Operation succeeded: snapmirror quiesce for destination "svm1:svm1_vol2". cluster1::> snapmirror show -source-path svm1_2:svm1_svm1_vol2_mirror Progress Source Path -----

Type Updated ----

Destination Path


------------ ------- ----------

---------



svm1:svm1_vol2 Snapmirrored Quiesced

true

-

Break the SnapMirror relationship:

4-49




-

cluster1::> snapmirror break -source-path svm1_2:svm1_svm1_vol2_mirror destination-path svm1:svm1_vol2 Operation succeeded: snapmirror break for destination "svm1:svm1_vol2". cluster1::> snapmirror show -source-path svm1_2:svm1_svm1_vol2_mirror Progress Source Path -----

Type Updated ----

Destination Path


------------ ------- ----------

Last Healthy

---------

-------

Idle

-


svm1:svm1_vol2 Broken-off

true

-

At this point, you can redirect the clients to the source volume. Ensure that the volume is in the namespace, that NAS and SAN configurations are set up, and that the volume has all required settings. Perform a snapmirror resync operation: cluster2::> snapmirror resync -source-path svm1:svm1_vol2 -destinationpath svm1_2:svm1_svm1_vol2_mirror Warning: All data newer than Snapshot copy snapmirror.67851b33-503b-11e4850e-123478563412_2147484675.2014-10-10_150854 on volume svm1_2:svm1_svm1_vol2_mirror will be deleted. Do you want to continue? {y|n}: y Operation is queued: initiate snapmirror resync to destination "svm1_2:svm1_svm1_vol2_mirror". cluster2::> snapmirror show -source-path svm1:svm1_vol2 Progress Source Path -----

Type Updated ----

Destination Path


------------ ------- ----------

---------


svm1:svm1_vol2 DP

svm1_2:svm1_svm1_vol2_mirror Snapmirrored Idle

true

-

Finally, ensure that you have the required SnapMirror policy and update schedule.

4-50




-

Fail Over Load-Sharing SnapMirror Relationships  Check the LS SnapMirror relationship: cluster1::> snapmirror show -type LS

 Promote the LS destination volume: cluster1::> snapmirror promote -destination-path cluster1://svm1/svm1_root_lsm2

 that the volume is online and writeable: cluster1::> volume show -volume svm1_root_lsm2 Notice that the svm1_root_lsm2 volume is now read/write.

42


FAIL OVER LOAD-SHARING SNAPMIRROR RELATIONSHIPS If the SVM root volume becomes unavailable, and you protected it with a set of load-sharing mirror copies, you can promote one of the mirrored volumes, then rename it to take the place of the original source volume. Check the LS SnapMirror relationship: cluster1::> snapmirror show -type ls Progress Source Path -----

Type Updated ----

Destination Path


------------ ------- ----------

Last Healthy

---------

-------

Idle

-


cluster1://svm1/svm1_root_lsm1 Snapmirrored

true


true

-


4-51




-

Promote the LS destination volume: cluster1::> snapmirror promote -destination-path cluster1://svm1/svm1_root_lsm2 Warning: Promote will delete the read-write volume cluster1://svm1/svm1_root and replace it with cluster1://svm1/svm1_root_lsm2. Do you want to continue? {y|n}: y [Job 122] Job is queued: snapmirror promote of destination "cluster1://svm1/svm1[Job 122] Job succeeded: SnapMirror: done

that the volume is online and writeable: cluster1::> volume show -volume svm1_root_lsm2 Vserver

Volume

Aggregate

State

Type Size

Available Used%

--------- --------- ------------ ---------- ---- ------ --------- ----svm1

svm1_root_lsm2 n1_aggr1

online

RW

20MB

18.85MB

5%

The svm1_root_lsm2 volume is now read/write. The destination volume that is the target of the promote command is now the source volume for the loadsharing mirror set, and all NFS file handles and CIFS connections are not interrupted.   

4-52

that load-sharing mirror volumes are read-only and that all client access requests are directed only to destination volumes. Thus, no read-only connections exist to the source volume that was removed by the promote operation, except for connections that might access the read-write share. NFS file handles and CIFS connections to the read/write share are nondisruptively transferred to the new source volume. Read-only connections to the promoted volume are nondisruptively transferred to other destination volumes in the load-sharing mirror set.




Fail Over Load-Sharing SnapMirror Relationships Rename the Source Volume

 The promote operation also deletes the original source volume, and the specific SnapMirror relationship for the promoted destination volume is removed: cluster1::> snapmirror show -type LS

 To retain the name for the source load-sharing mirror set, rename the volume: cluster1::> volume rename -vserver svm1 -volume svm1_root_lsm2 -newname svm1_root

43


FAIL OVER LOAD-SHARING SNAPMIRROR RELATIONSHIPS: RENAME THE SOURCE VOLUME The promote operation also deletes the original source volume, and the specific SnapMirror relationship for the promoted destination volume is removed. cluster1::> snapmirror show -type ls Progress Source Path -----

Type Updated ----

Destination Path


------------ ------- ----------

Last Healthy

---------

-------

-

true

cluster1://svm1/svm1_root_lsm2 LS


Because the promoted volume would have a different volume name than the original source volume, the new source volume can be renamed to retain that name for the source of the load-sharing mirror set, as shown in the following example. Rename the volume to retain the name for the source load-sharing mirror set:

4-53




cluster1::> volume rename -vserver svm1 -volume svm1_root_lsm2 -newname svm1_root [Job 123] Job succeeded: Successful cluster1::> snapmirror show -type ls Progress Source Path -----

Type Updated ----

Destination Path


------------ ------- ----------

---------




true

-

-

The source volume has changed for the svm1_root_lsm1 volume. Best practice The promote operation deletes the original source volume; therefore, you might need to create another loadsharing mirror destination volume on the node where the current source volume was located.

4-54




Lesson 4

SnapMirror Advanced Topics

44


LESSON 4: SNAPMIRROR ADVANCED TOPICS

4-55




Converting a SnapMirror Relationship to a SnapVault Secondary FlexVol (vs1P:vol1_vs1P) Primary

WAN Remote

(vs1R:vol1_vs1R) FlexVol

45


CONVERTING A SNAPMIRROR RELATIONSHIP TO A SNAPVAULT SECONDARY Problem: Consider this scenario: A customer who is using SnapMirror in clustered Data ONTAP 8.1 wants to use SnapVault technology in clustered Data ONTAP 8.3, for longer retention. Thus, you need to convert an existing SnapMirror relationship to a SnapVault relationship. Solution: Upgrade your source and destination clusters to clustered Data ONTAP 8.3.   

Your existing SnapMirror relationships continue to remain cluster scope and behave as they did in clustered Data ONTAP 8.1. These SnapMirror relationships do not benefit from the scalability improvements, unless they are deleted and recreated. However, both clustered Data ONTAP 8.1 and clustered Data ONTAP 8.3 use the block-level engine for mirrors, thus no rebaseline is required; only resync is required.

Refer to the Clustered Data ONTAP 8.3 Data Protection Guide for the exact steps to convert a SnapMirror data protection mirror relationship to a SnapVault secondary.

4-56




Intercluster SnapMirror Throttle To limit the amount of bandwidth that is used by intercluster SnapMirror transfers, apply a throttle to intercluster SnapMirror relationships.

46


INTERCLUSTER SNAPMIRROR THROTTLE To limit the amount of bandwidth that is used by the intercluster SnapMirror transfers, apply a throttle to intercluster SnapMirror relationships.  

When creating a new relationship, you can set a throttle through the CLI by adding the –throttle option and a value in kilobytes, by modifying an existing relationship with the snapmirror modify command. NetApp OnCommand System Manager 3.0 does not currently SnapMirror throttle management. In this example, a 10 megabyte (MB) throttle is applied to an existing relationship by using the snapmirror modify command.

cluster02::> snapmirror modify -destination-path cluster02://vs1/vol1 – throttle 10240 To change the throttle of an active SnapMirror relationship, terminate the existing transfer and restart it, to use the new value. The SnapMirror feature restarts the transfer from the last restart checkpoint by using the new throttle value, rather than restarting from the beginning. NOTE: Intracluster SnapMirror relationships, which use the cluster interconnect, do not allow a throttle to be set. However, starting with clustered Data ONTAP 8.2.1, intracluster throttle is ed, and the process works the same way as intercluster throttle works.

4-57




References  NetApp University courses:  Data ONTAP 8.3 istration

 NetApp Documentation:  Clustered Data ONTAP 8.3 Data Protection Guide  Clustered Data ONTAP 8.3 Infinite Volumes Management Guide  Clustered Data ONTAP File Access and Protocols Management Guide  NetApp Technical Report: TR-4015: SnapMirror Configuration Best Practices Guide for Clustered Data ONTAP 8.2

47


REFERENCES

4-58





Now I understand how to design and implement a SnapMirror solution. More importantly, I understand how to recover my data by using SnapMirror software, in case a disaster happens.

48



4-59




Exercise

Module 4: Using SnapMirror for Data Protection


49



4-60




Module 5

SnapVault Backup and Recovery

1


MODULE 5: SNAPVAULT BACKUP AND RECOVERY

5-1

Data ONTAP Data Protection istration: SnapVault Backup and Recovery




 Describe SnapVault technology  Create, schedule, and monitor SnapVault replication operations

 Manage SnapVault backup and restore operations  Use SnapVault software to recover data  Understand monitoring and management tasks for SnapVault and SnapMirror solutions

2


MODULE OBJECTIVES

5-2




Lesson 1

SnapVault Technology

3


LESSON 1: SNAPVAULT TECHNOLOGY

5-3




SnapVault Technology  The SnapVault backup solution provides long-term, storageefficient retention of backups.  SnapVault relationships enable you to back up selected Snapshot copies of volumes to a destination volume and to retain the backups.

4


SNAPVAULT TECHNOLOGY Snapshot technology is the foundation for SnapRestore, SnapManager, SnapDrive, SnapMirror, FlexClone, and SnapVault operations. SnapVault technology is a NetApp disk-to-disk backup solution that is built into the Data ONTAP operating system. To enable SnapVault software on your NetApp system, you install a license key; no additional hardware or software must be installed. SnapVault software enables you to replicate your data to a secondary volume and to retain the data for longer than you might retain data on your primary volume. A SnapVault backup is a collection of Snapshot copies on a FlexVol volume, from which you can restore data if the primary data is not usable. Snapshot copies are created based on a Snapshot policy. The SnapVault backup backs up Snapshot copies based on its schedule and on SnapVault policy rules. You can create a SnapVault relationship between FlexVol volumes and assign a SnapVault policy to the relationship to create a SnapVault backup. A SnapVault backup contains a set of read-only backup copies, which are located on a secondary volume. NOTE: SnapVault relationships are ed on clustered Data ONTAP 8.2 or later. SnapVault relationships are not ed on infinite volumes. An important architectural change is that SnapVault software in clustered Data ONTAP replicates at the volume Level rather than at the qtree level, as was true in SnapVault for Data ONTAP operating in 7-Mode. This means that the source of a SnapVault relationship must be a volume, and that volume must replicate to its own volume on the SnapVault secondary.

5-4




Clustered Data ONTAP Nondisruptive Operations and SnapVault  SnapVault s can seamlessly rebalance SnapVault primaries and secondaries for performance or capacity needs.  You can move SnapVault primary and secondary volumes to different aggregates or nodes within a cluster without disrupting SnapVault operations.

5


CLUSTERED DATA ONTAP NONDISRUPTIVE OPERATIONS AND SNAPVAULT SnapVault s can now take advantage of nondisruptive operations, which are essential to the architecture of clustered Data ONTAP. SnapVault s can seamlessly rebalance SnapVault primaries and secondaries for performance or capacity needs, because they can now move SnapVault primary and secondary volumes to different aggregates or nodes within a cluster without disrupting SnapVault operations. If a SnapVault transfer is in progress when a volume is moved by using the vol move operation, the transfer may pause for a few minutes during the volume cutover phase. However, the transfer resumes from the most recent transfer checkpoint after the vol move operation completes. s never have to reconfigure a SnapVault relationship just because a volume was moved to another node by using the vol move operation.

5-5




SnapVault Operations

Primary Volume (Read/Write)

Secondary Volume (Data Protection)

Snapshot Policy Label = Svault

SnapVault Policy Label = Svault

6


SNAPVAULT OPERATIONS You create SnapVault relationships to back up and restore volumes. You can select the Snapshot copies that the SnapVault relationship uses to back up and restore volumes. The Snapshot policy that is assigned to the source volume specifies when Snapshot copies are performed and assigns a label to the Snapshot copy. The SnapVault policy that is assigned to the SnapVault relationship specifies which source volume Snapshot copies are replicated to the SnapVault backup, by indicating the Snapshot copy label. Data that is not backed up to a SnapVault backup If you back up an entire storage virtual machine (SVM) to a SnapVault backup by establishing a SnapVault relationship for each volume in the SVM, namespace and root information is not backed up. To protect namespace and root information for an SVM, you must manually create the namespace and root on the SnapVault secondary volume.

5-6





When should I deploy a SnapVault relationship instead of a SnapMirror relationship?

7



5-7




Lesson 2

Implementing SnapVault

8


LESSON 2: IMPLEMENTING SNAPVAULT

5-8




Implementing a SnapVault Solution Typical Workflow

 Create a preconfiguration checklist.  Check the licensing requirements.  Check and create cluster and SVM peer relationships.  Use and create a SnapVault policy.  Use and create a Snapshot policy and Snapshot label.  Choose and create a destination volume.

 Perform a baseline transfer.  Automate updates. (This lesson will demonstrate this workflow.) 9


IMPLEMENTING A SNAPVAULT SOLUTION: TYPICAL WORKFLOW

5-9




SnapVault Preconfiguration Checks  Privileges

 Peer relationships  A SnapVault policy  The snapmirror-label attribute  The amount of time that you need for the baseline transfer

10


SNAPVAULT PRECONFIGURATION CHECKS     

5-10

You must have either cluster privileges to perform this task for a cluster, or SVM privileges to perform this task for an SVM. If the primary and secondary volumes are in different SVMs or clusters, peer relationships must exist. A SnapVault policy must exist. You can either create one or accept the default SnapVault policy (XDPDefault) that is automatically assigned. Only Snapshot copies with labels that are configured in the SnapVault policy rules are replicated by SnapVault. The Snapshot policy that is assigned to the primary volume must include the snapmirror-label attribute. Snapshot copies on the primary volume without the snapmirror-label attribute are ignored by the SnapVault software. Depending on the amount of data on the primary volume and other factors, it might take a long time to complete the baseline transfer.




Implementing a SnapVault Solution Check SnapVault Licenses

Check for the required licenses: cluster1::> system license show Owner: cluster1-01 Package Expiration

Type

Description

----------------- ------- ----------------SnapVault

11

license SnapVault License


IMPLEMENTING A SNAPVAULT SOLUTION: CHECK SNAPVAULT LICENSES Check that you have the required licenses installed: cluster1::> system license show Serial Number: 1-80-000008 Owner: cluster1 Package

Type

Description

Expiration

----------------- ------- --------------------- -------------------Base

license Cluster Base License

Serial Number: 1-81-0000000000000004082368507

5-11




-

Owner: cluster1-01 Package

Type

Description

Expiration

-----------------------

-------

---------------------

-------------

NFS

license

NFS License

-

CIFS

license

CIFS License

-

SnapRestore

license

SnapRestore License

-

SnapMirror

license

SnapMirror License

-

FlexClone

license

FlexClone License

-

SnapVault

license

SnapVault License

-

7 entries were displayed. If necessary, perform this check on both source and destination clusters.

5-12




Implementing a SnapVault Solution Create a Peer SVM on the Destination Cluster

If necessary, create an SVM on cluster2 for SnapVault: cluster2::> vserver create -vserver svm1_2 -rootvolume svm1_2_root -aggregate aggr1 -rootvolume-security-style ntfs

[Job 46] Job succeeded: Vserver creation completed

12


IMPLEMENTING A SNAPVAULT SOLUTION: CREATE A PEER SVM ON THE DESTINATION CLUSTER Create a peer SVM on cluster2: cluster2::> vserver create -vserver svm1 -rootvolume svm1_root -aggregate aggr1 -rootvolume-security-style ntfs [Job 44] Job succeeded: Vserver creation completed

5-13




Implementing a SnapVault Solution Create a New SnapVault Policy

 Use either the default SnapMirror policy that is named XDPDefault, or create a new policy.  Create a new SnapVault policy on the primary (source) cluster: cluster1::> snapmirror policy create -vserver svm1 -policy svm1-vault-policy

13


IMPLEMENTING A SNAPVAULT SOLUTION: CREATE A NEW SNAPVAULT POLICY There is a default policy that you can use, named “XDPdefault,” that protects Snapshot copies with the “daily” and “weekly” SnapMirror labels. The default policy keeps seven daily Snapshot copies and 52 weekly Snapshot copies. Create a new SnapVault policy named svm1-vault-policy: cluster1::> snapmirror policy create -vserver svm1 -policy svm1-vaultpolicy NOTE: Many SnapMirror commands apply to SnapVault software.

5-14




Implementing a SnapVault Solution Add a SnapVault Policy Rule

 Add the snapmirror-label attribute and a Snapshot copy retention: cluster1::> snapmirror policy add-rule -vserver svm1 -policy svm1-vault-policy -snapmirror-label SnapVault -keep 7 -preserve false

The policy keeps seven Snapshot copies on the volume.

 that the rule is using the correct label: cluster1::> snapmirror policy show 14


IMPLEMENTING A SNAPVAULT SOLUTION: ADD A SNAPVAULT POLICY RULE Add the snapmirror-label and a Snapshot copy retention: cluster1::> snapmirror policy add-rule -vserver svm1 -policy svm1-vaultpolicy -snapmirror-label SnapVault -keep 7 -preserve false that the rule is using the correct label: cluster1::> snapmirror policy show SVM

Policy

Policy

Number

Transfer

Name

Name Comment

Type

Of Rules

Tries

Priority

------svm1

-----------------svm1-vault-policy

-----vault

-------1

------8

-------normal -

SnapMirror Label: SnapVault

Keep: 7 Total Keep: 7

The -preserve option allows you to keep any Snapshot copies that are more recent than the common Snapshot copy on the SnapVault secondary volume, and that are not present on the primary volume.

5-15




Implementing a SnapVault Solution The Snapshot Copy Policy

 Check the Snapshot copy policy on the source volume to ensure that it is using the snapmirror-label attribute.  The snapmirror-label attribute must match the snapmirror-label attribute in the SnapVault policy.  You might decide to use a preconfigured Snapshot copy policy or to create a new policy.

15


IMPLEMENTING A SNAPVAULT SOLUTION: THE SNAPSHOT COPY POLICY

5-16




Implementing a SnapVault Solution Modify a Snapshot Copy Policy

 Check the existing Snapshot copy policies: cluster1::> volume snapshot policy show

 Modify the 2Hour_Policy that was created previously: cluster1::> volume snapshot policy modify -vserver svm1 -policy 2Hour_Policy -snapmirror-labels SnapVault The SnapMirror label exactly matches the SnapMirror label in the SnapVault policy.

 that the snapmirror-label attribute is correct: cluster1::> volume snapshot policy show 16


IMPLEMENTING A SNAPVAULT SOLUTION: MODIFY A SNAPSHOT COPY POLICY Modify the Snapshot policy on the primary (source) volume so that it includes the snapmirror-label attribute: cluster1::>volume snapshot policy show SVM: svm1 Number of

Is

Policy Name

Schedules

Enabled

Comment

--------------------

---------

-------

-----------------------------

true

Takes a Snapshot Copy Every 2 Hours

2Hour_Policy

1

Schedule

Count

Prefix

SnapMirror Label

----------------

-----

----------------

-------------

2HourSnapshot

-

2HourSnapshot

84

Modify the 2Hour_Policy that you created earlier: cluster1::> volume snapshot policy modify -vserver svm1 -policy 2Hour_Policy -snapmirror-labels SnapVault

5-17




that the snapmirror-label attribute is correct: cluster1::> volume snapshot policy show SVM: svm1 Number of

Is

Policy Name

Schedules

Enabled

Comment

---------------------

---------

-------

------------------------

true

Takes a Snapshot Copy Every 2 Hours

2Hour_Policy

1

Schedule

Count

Prefix

SnapMirror Label

----------------

-----

----------------

-------------------

2HourSnapshot

84

2HourSnapshot

SnapVault

Notice that the SnapMirror label in the policy exactly matches the snapmirror-label attribute in the SnapVault policy.

5-18




Implementing a SnapVault Solution Create a SnapVault Destination Volume

Create a new volume to be the recipient of the SnapVault transfers: cluster2::> volume create -vserver svm1_2 -volume svm1_2_vol1 -aggregate aggr1 -size 200m -type DP -security-style ntfs

17


IMPLEMENTING A SNAPVAULT SOLUTION: CREATE A SNAPVAULT DESTINATION VOLUME When you create a FlexVol volume using the -type DP option, it is created with settings that reflect best practices for destination volumes. These are different from the default settings that are used for RW volumes. Space setting

RW volume

DP volume

space-guarantee

volume

volume

autosize

false

true

autosize-mode

off

grow_shrink

Percent

85 (percent)

85 (percent)

autosize-shrinkthreshold-Percent

50 (percent)

80 (percent; autosizegrowthreshold-percent -5)

min-autosize

Initial volume size

Initial volume size

max-autosize

120% of volume size

Maximum aggregate size

Snap reserve (percentsnapshot-space)

5 (percent)

0 (percent)

fractional-reserve

100 (percent)

0 (percent)

autosize-growthreshold-

5-19




Create the volume: cluster2::> volume create -vserver svm1_2 -volume svm1_2_vol1 -aggregate aggr1 -size 200m -type DP -security-style ntfs

5-20




Implementing a SnapVault Solution Create a Cron Schedule for the Backup Strategy

 Check the schedules that are currently available on the destination cluster: cluster2::> job schedule show

 Create a new schedule: cluster2::> job schedule cron create -name SnapVaultJob -dayofweek all -hour all -minute 10 This job runs every day, every hour, at 10 minutes past the hour.

 that the new schedule was created correctly: cluster2::> job schedule cron show

18


IMPLEMENTING A SNAPVAULT SOLUTION: CREATE A CRON SCHEDULE FOR THE BACKUP STRATEGY On the destination cluster, review existing schedules to see whether any of them meet your requirements, by using the job schedule show command. If an existing schedule meets your requirements, you do not need to create a new schedule. Check the schedules that are currently available on the destination cluster: cluster2::> job schedule show Name

Type

Description

-----------

---------

---------------------------------------------

5min

cron @:00,:05,:10,:15,:20,:25,:30,:35,:40,:45,:50,:55

8hour

cron

@2:15,10:15,18:15

Auto Balance Aggregate Scheduler interval

Every 1h

RepositoryBalanceMonitorJobSchedule interval

Every 10m

daily

cron

@0:10

hourly

cron

@:05

weekly

cron

Sun@0:15


5-21




Create a new schedule by using the job schedule cron create command: cluster2::> job schedule cron create -name SnapVaultJob -dayofweek all hour all -minute 10 This job will run every day of the week, every hour of the day, at ten minutes past the top of the hour. that the new schedule was created correctly: cluster2::> job schedule cron show Name

Description

----------------

--------------------------------------------------

5min

@:00,:05,:10,:15,:20,:25,:30,:35,:40,:45,:50,:55

8hour

@2:15,10:15,18:15

Hourly_SnapMirror

@:03

SnapVaultJob

@:10

daily

@0:10

hourly

@:05

weekly

Sun@0:15


5-22




Implementing a SnapVault Solution Create a SnapVault Policy on the Destination Cluster

 Create a SnapVault policy: cluster2::> snapmirror policy create -vserver svm1_2 -policy svm1_2_vault-policy

 Add the snapmirror-label attribute: cluster2::> snapmirror policy add-rule -vserver svm1_2 -policy svm1_2_vault-policy -snapmirrorlabel SnapVault -keep 20 The snapmirror label matches the source volume.

The retention is different from the source volume.

 that the snapmirror label is correct: cluster2::> snapmirror policy show -vserver svm1_2 19


IMPLEMENTING A SNAPVAULT SOLUTION: CREATE A SNAPVAULT POLICY ON THE DESTINATION CLUSTER This task is performed from the destination cluster. You can use a default policy that is named “XDPdefault,” which protects Snapshot copies with the “daily” and “weekly” SnapMirror labels. The default policy keeps seven daily Snapshot copies and 52 weekly Snapshot copies. NOTE: Depending on how it is defined, a SnapVault policy can be available to a specific SVM or to the entire cluster. SVM-wide policies can be configured by both SVM s and cluster s, but a cluster-wide policy can be configured only by a cluster . For more information, refer to the Clustered Data ONTAP Data Protection Guide. The example in this procedure creates an SVM-wide policy. Create a SnapVault policy by using the snapmirror policy create command: cluster2::> snapmirror policy create -vserver svm1_2 -policy svm1_2_vault-policy Add the snapmirror-label attribute and Snapshot copy retention: cluster2::> snapmirror policy add-rule -vserver svm1_2 -policy svm1_2_vault-policy -snapmirror-label SnapVault-Backup -keep 20

5-23




that the SnapMirror label is correct. cluster2::> snapmirror policy show -vserver svm1_2 SVM

Policy

Policy Number

Name

Name

Type

Of Rules

Tries

------- ---------

------

--------

------

svm1_2

vault

5-24

svm1_2_vault-policy

Transfer

1

Priority 8

SnapMirror Label:

Keep: 20

SnapVault

Keep: 20




-------normal

Comment ------

Implementing a SnapVault Solution Create a SnapVault Relationship

 Create the SnapVault relationship: cluster2::> snapmirror create -source-path svm1:svm1_vol1 -destination-path svm1_2:svm1_2_vol1 -type XDP -policy svm1_2_vaultpolicy -schedule SnapVaultJob

 Start the baseline transfer: cluster2::> snapmirror initialize -destinationpath cluster2://svm1_2/svm1_2_vol1

 the SnapVault status: cluster2::> snapmirror show

20


IMPLEMENTING A SNAPVAULT SOLUTION: CREATE A SNAPVAULT RELATIONSHIP This task is performed from the destination SVM. On the destination SVM, create a SnapVault relationship, and assign an XDP policy by using the snapmirror create command with the -type XDP parameter and the –policy parameter: cluster2::> snapmirror create -source-path svm1:svm1_vol1 -destinationpath svm1_2:svm1_2_vol1 -type XDP -policy svm1_2_vault-policy -schedule SnapVaultJob Initialize the relationship by using the snapmirror initialize command to start a baseline transfer. The command creates a new Snapshot copy that is transferred to the destination volume and is used as a baseline for subsequent incremental Snapshot copies. The command does not transfer any Snapshot copies that currently exist on the source volume. NOTE: Scheduled updates will not succeed until the SnapVault relationship completes initialization. You do not have to initialize the SnapVault relationship when you create it. You can initialize the relationship from the destination SVM at a later time that can better accommodate the baseline transfer. Start the baseline transfer: cluster2::> snapmirror initialize -destination-path cluster2://svm1_2/svm1_2_vol1

5-25




the SnapVault status: cluster2::> snapmirror show Progress

Source

Destination Relationship

Total Type

Path

Mirror Last

Path

State

Progress Healthy Updated

----------- -------------- ------------

svm1:svm1_vol1

XDP

svm1_2:svm1_2_vol1

true

-

-

-------

-------------------

------------

-------

--

Snapmirrored Idle

Display more detail by using the –instance parameter: cluster2::> snapmirror show -instance Source Path: svm1:svm1_vol1 Destination Path: svm1_2:svm1_2_vol1 Relationship Type: XDP Relationship Group Type: none SnapMirror Schedule: SnapVaultJob SnapMirror Policy Type: vault SnapMirror Policy: svm1_2_vault-policy Tries Limit: Throttle (KB/sec): unlimited Mirror State: Snapmirrored Relationship Status: Idle File Restore File Count: File Restore File List: Transfer Snapshot: Snapshot Progress: Total Progress: Network Compression Ratio: Snapshot Checkpoint: Newest Snapshot: snapmirror.1383c700-4345-11e497f5-0050 560140c1_2147484678.2014-0923_182150 Newest Snapshot Timestamp: 09/23 16:16:29 Exported Snapshot: snapmirror.1383c700-4345-11e497f5-0050 560140c1_2147484678.2014-0923_182150 5-26




Exported Snapshot Timestamp: 09/23 16:16:29 Healthy: true Unhealthy Reason: Constituent Relationship: false Destination Volume Node: cluster2-01 Relationship ID: 466c0484-434e-11e4-97f50050560140c1 Current Operation ID: Transfer Type: Transfer Error: Current Throttle: Current Transfer Priority: Last Transfer Type: initialize Last Transfer Error: Last Transfer Size: 21.31KB Last Transfer Network Compression Ratio: 1:1 Last Transfer Duration: 0:0:19 Last Transfer From: svm1:svm1_vol1 Last Transfer End Timestamp: 09/23 18:22:09 Progress Last Updated: Relationship Capability: 8.2 and above Lag Time: 2:9:8 Number of Successful Updates: 0 Number of Failed Updates: 0 Number of Successful Resyncs: 0 Number of Failed Resyncs: 0 Number of Successful Breaks: 0 Number of Failed Breaks: 0 Total Transfer Bytes: 21820 Total Transfer Time in Seconds: 19

5-27




Lesson 3

Restoring Data with SnapVault

21


LESSON 3: RESTORING DATA WITH SNAPVAULT

5-28




Ken Asks

How do I restore my data with SnapVault software?

22


KEN ASKS

5-29




Restoring Data with SnapVault Modifying and Removing SnapVault Policy Rules

You can use SnapVault software to restore data to these volumes:  Original source volume  New, empty secondary volume

 New secondary volume that already contains data

23


RESTORING DATA WITH SNAPVAULT: MODIFYING AND REMOVING SNAPVAULT POLICY RULES Guidelines for restoring the active file system The restore operation from a SnapVault backup copies a single, specified Snapshot copy from a SnapVault secondary volume to a specified volume. Restoring a volume from a SnapVault secondary volume changes the view of the active file system but preserves all earlier Snapshot copies in the SnapVault backup. Before restoring a volume, you must shut down any application that accesses data in a volume to which a restore is writing data. Therefore, you must dismount the file system, shut down any database, and deactivate and quiesce the local volume manager (LVM), if you are using an LVM. The restore operation is disruptive. When the restore operation finishes, the cluster or SVM must remount the volume and restart all applications that use the volume. The restore destination volume must not be the destination of another mirror or the secondary of another SnapVault relationship. You can restore to the following volumes:   

5-30

Original source volume: You can restore from a SnapVault secondary volume back to the original SnapVault primary volume. New, empty secondary volume: You can restore from a SnapVault secondary volume to a new, empty secondary volume. You must first create the volume as a data protection volume. New secondary volume that already contains data: You can restore from a SnapVault secondary volume to a volume that is prepopulated with data. The volume must have a Snapshot copy in common with the restore primary volume and must not be a data protection volume.




Performing Restore Operations A restore operation from a SnapVault backup consists of a series of actions that are performed on a temporary restore relationship and on the secondary volume:  A new temporary relationship is created from the restore source.  The restore destination might be the original volume or a new volume.  During the restore operation, the destination volume is changed to read-only.  When the restore operation finishes, the temporary relationship is removed. 24


PERFORMING RESTORE OPERATIONS A restore operation from a SnapVault backup consists of a series of actions that are performed on a temporary restore relationship and on the secondary volume. During a restore operation, the following actions occur: 

  

5-31

A new temporary relationship is created from the restore source (which is the original SnapVault relationship secondary volume) to the restore destination. The temporary relationship is a restore type (RST). The snapmirror show command displays the RST type while the restore operation is in progress. The restore destination might be the original SnapVault primary volume or might be a new SnapVault secondary volume. During the restore process, the restore destination volume is changed to read-only. When the restore operation finishes, the temporary relationship is removed, and the restore destination volume is changed to read-write.




Restoring a Volume from a SnapVault Backup  Restore a volume by using the snapmirror restore command: cluster1::> snapmirror restore -destination-path svm1:svm1_vol1 -source-path svm1_2:svm1_2_vol1 -source-snapshot 5min.2014-09-23_2010

Choose exactly and carefully which Snapshot copy to use in the restore operation.

 the restore status: cluster1::> snapmirror show-history

25


RESTORING A VOLUME FROM A SNAPVAULT BACKUP If the data on a volume becomes unavailable, you can restore the volume to a specific time by copying a Snapshot copy in the SnapVault backup. You can restore data to the same primary volume or to a new location. This is a disruptive operation. CIFS traffic must not be running on the SnapVault primary volume when a restore operation is running. This task describes how to restore a whole volume from a SnapVault backup. To restore a single file or LUN, you can restore the whole volume to a different, nonprimary volume, and then select the file or LUN, or you can use the NetApp OnCommand management software online management tools. If the volume to which you are restoring has compression enabled, and the secondary volume from which you are restoring does not have compression enabled, disable compression. You disable compression to retain storage efficiency during the restore. Restore a volume by using the snapmirror restore command: cluster1::> snapmirror restore -destination-path svm1:svm1_vol1 -sourcepath svm1_2:svm1_2_vol1 -source-snapshot 5min.2014-09-23_2010 Warning: All data newer than Snapshot copy 5min.2014-09-23_2010 on volume svm1:svm1_vol1 will be deleted. Export policies currently enforced on the qtrees of volume "svm1:svm1_vol1" will not change during this operation. If the currently enforced export policies are different from those in Snapshot copy "5min.2014-09-23_2010", reassign the export policies of the qtrees on this volume after this operation.

5-32




Do you want to continue? {y|n}: y [Job 104] Job is queued: snapmirror restore from source "svm1_2:svm1_2_vol1" for the snapshot 5min.2014-09-23_2010. the restore status: cluster1::> snapmirror show-history Destination Source Path

Path

----------- -----------

Start

End

Operation

Time

Time

---------

-----------

----------- --------

9/23/2014

9/23/2014

20:14:48

20:14:53

Result

svm1:svm1_vol1 svm1_2:svm1_2_vol1 restore

5-33




success

Restoring a Single File or LUN  You can restore a single file or LUN or a set of files or LUNs from a Snapshot copy in a SnapVault secondary volume to the active file system of a primary volume.  You can restart a failed or aborted single file or LUN restore operation by reissuing the snapmirror restore command. For more information about restoring LUNs, see the Clustered Data ONTAP 8.3 SAN istration Guide.

26


RESTORING A SINGLE FILE OR LUN You can restore a single file or LUN or a set of files or LUNs from a Snapshot copy in a SnapVault secondary volume to the active file system of a primary volume. To restart a failed or aborted single file or LUN restore operation, reissue the snapmirror restore command.

5-34





What should you consider when restoring LUNs?

27



5-35




Lesson 4

SnapVault Advanced Topics

28


LESSON 4: SNAPVAULT ADVANCED TOPICS

5-36




Ken Asks

What can be done to prevent a baseline transfer from using a lot of my network bandwidth?

29


KEN ASKS

5-37




SnapVault Advanced Topics  SnapVault tape seeding

 Managing storage efficiency for SnapVault secondary volumes

30


SNAPVAULT ADVANCED TOPICS

5-38




SnapVault Advanced Topics SnapVault Tape Seeding (1 of 2)

Typically, you create a prepopulated secondary volume when you copy a primary volume to a secondary volume by using tape. This process is called tape seeding.

Destination Volume

Source Volume

31


SNAPVAULT ADVANCED TOPICS: SNAPVAULT TAPE SEEDING (1 OF 2) Tape seeding is an SMTape functionality that helps you initialize a destination FlexVol volume in a data protection mirror relationship. Tape seeding enables you to establish a data protection mirror relationship between a source system and a destination system over a low-bandwidth connection. Incremental mirroring of Snapshot copies from the source to the destination is feasible over a low-bandwidth connection. However, an initial mirroring of the base Snapshot copy takes a long time over a low-bandwidth connection. In such a case, you can perform an SMTape backup of the source volume to a tape and use the tape to transfer the initial base Snapshot copy to the destination. You can then set up incremental SnapMirror updates to the destination system by using the low-bandwidth connection. Before creating a SnapVault relationship to a prepopulated secondary, implement these guidelines: 

The primary and secondary volumes must have a common Snapshot copy.



Snapshot copies on the secondary volume that are newer than the common Snapshot copy are deleted.

When a SnapVault relationship is created, all Snapshot copies on the secondary volume that are more recent than the common Snapshot copy and that are not present on the primary volume are deleted. Newer Snapshot copies on the primary volume that match the configured SnapVault policy are transferred to the secondary volume according to the SnapVault policy. You can use the -preserve option to keep any Snapshot copies that are more recent than the common Snapshot copy on the SnapVault secondary volume and that are not present on the primary volume.

5-39




When you use the -preserve option, data on the secondary volume is logically made the same as the common Snapshot copy. All newer Snapshot copies on the primary volume that match the SnapVault policy are transferred to the secondary volume. This option is useful when the latest common Snapshot copy is deleted from the primary volume but another, older common Snapshot copy between the primary and secondary volumes still exists. For more information, refer to the Clustered Data ONTAP 8.3 Data Protection Tape Backup and Recovery Guide and the Clustered Data ONTAP 8.3 Data Protection Guide.

5-40




SnapVault Advanced Topics SnapVault Tape Seeding (2 of 2)

After a tape seeding operation, convert a data protection destination volume to a SnapVault secondary volume: 1. Break the data protection mirror relationship.

2. Delete the existing data protection mirror relationship. 3. Remove the relationship from the source SVM. 4. Create a SnapVault relationship using the –type XDP parameter. 5. Convert the destination volume from read-write to a SnapVault volume, and establish the SnapVault relationship by using the snapmirror resync command. 32


SNAPVAULT ADVANCED TOPICS: SNAPVAULT TAPE SEEDING (2 OF 2) When you are tape seeding, after you transfer the data from the tape to the volume, the volume is a data protection destination volume. In the case of a SnapVault secondary volume to disaster protection volume cascade, if the SnapVault secondary volume is lost, you can resume SnapVault protection by creating a direct relationship between the SnapVault primary volume and the disaster protection destination volume. To do this, you must make the disaster protection destination volume a SnapVault secondary volume. 1. Break the data protection mirror relationship by using the snapmirror break command. The relationship is broken, and the disaster protection volume becomes a read-write volume. 2. If an existing data protection mirror relationship exists, use the snapmirror delete command to delete the relationship. 3. Remove the relationship information from the source SVM by using the snapmirror release command. This also deletes the Snapshot copies that were created by Data ONTAP from the source volume. 4. Create a SnapVault relationship between the primary volume and the read-write volume by using the snapmirror create command with the -type XDP parameter. 5. Convert the destination volume from a read-write volume to a SnapVault volume, and establish the SnapVault relationship, by using the snapmirror resync command.

5-41




Managing Storage Efficiency for SnapVault Secondary Volumes If the primary volume does not have storage efficiency enabled, you can enable storage efficiency on a SnapVault secondary volume by enabling storage efficiency on the volume:  To enable storage efficiency, use the volume efficiency command with the -on parameter.  If the volume already has data that you want to make storage efficient, use the volume efficiency command with the start and -scan-old-data parameters to start a scan of the volume.

33


MANAGING STORAGE EFFICIENCY FOR SNAPVAULT SECONDARY VOLUMES Guidelines for managing storage efficiency for SnapVault backups If both the primary and secondary volumes in a SnapVault relationship have storage efficiency enabled, then data transfers to the SnapVault secondary volume preserve storage efficiency. If the primary volume does not have storage efficiency enabled, you might want to enable storage efficiency only on the secondary volume. Because SnapVault secondary volumes typically contain a large amount of data, it is important to maximize storage efficiency on SnapVault secondary volumes. If storage efficiency is enabled on the primary volumes If the primary volume in a SnapVault relationship is enabled for storage efficiency, all data backup operations preserve the storage efficiency. If storage efficiency is enabled only on the secondary volume If the primary volume in a SnapVault relationship does not have storage efficiency enabled, you might want to enable storage efficiency for the secondary volume, because it is likely to contain a large amount of data over time. You can use the volume efficiency command to start a scan on the volume, if data is already present on the volume from transfers. If this is a new relationship with no transfers, then you do not need to run the scan manually.

5-42




Changes to the volume's efficiency schedule do not take effect for a SnapVault secondary volume. Instead, when storage efficiency is enabled, the SnapVault relationship manages the schedule. When a data transfer begins, the storage efficiency process automatically pauses until the transfer is finished and automatically begins again after the data transfer is complete. Because data transfers to a SnapVault secondary volume might include more than one Snapshot copy, the storage efficiency process is paused for the entire duration of the update operation. After the transfer is finished, and the post-transfer storage efficiency process is complete, the last Snapshot copy that was created in the secondary volume is replaced by a new, storageefficient Snapshot copy. If the last Snapshot copy that is created in the secondary volume is locked before it can be replaced by a new, storage-efficient Snapshot copy, then a new, storage-efficient Snapshot copy is still created, but the locked Snapshot copy is not deleted. That Snapshot copy is deleted later during the storage-efficient cleanup process, after a subsequent update to the SnapVault secondary volume and after the lock is released. A Snapshot copy in a SnapVault secondary volume might be locked because the volume is the source in another relationship, such as a data protection mirror relationship. If the secondary volume has additional compression enabled, storage efficiency is not preserved. Storage efficiency on all data transfers in SnapVault relationships is not preserved when the secondary volume has additional compression enabled. Because of the loss of storage efficiency, a warning message is displayed when you enable compression on a SnapVault secondary volume. After you enable compression on the secondary volume, you can never have storage-efficient transfers.

5-43




Lesson 5

Monitoring and Managing SnapVault and SnapMirror

34


LESSON 5: MONITORING AND MANAGING SNAPVAULT AND SNAPMIRROR

5-44




Managing Replication Updates If you want to do this…

Use this command.

Get SnapVault status.

snapmirror show

Manually update a SnapVault relationship.

snapmirror update

Modify SnapVault relationship properties.

snapmirror modify

Modify a mirror policy or SnapVault policy.

snapmirror policy modify

Modify an existing rule in a SnapVault policy.

snapmirror policy modify-rule

Remove a rule in a SnapVault policy.

snapmirror policy remove-rule

Delete a mirror policy or SnapVault policy.

snapmirror policy delete

SnapMirror commands are used to manage SnapMirror and SnapVault relationships. 35


MANAGING REPLICATION UPDATES Commands for managing mirror and SnapVault policies Cluster s can use the snapmirror policy commands to create and manage all data protection mirror and SnapVault policies. SVM s can use the same commands to create and manage all data protection mirror and SnapVault policies within SVMs.  

5-45

All policy-management commands (except for the snapmirror policy show command) must be run on the SVM that contains the destination volume. Commands for SnapVault policies are ed only by FlexVol volumes.




Get the SnapVault or SnapMirror Relationship Status  Use these commands on the destination cluster: cluster2::> snapmirror show

 Get more details by using the –destination-path or the –instance (or –fields) options: cluster2::> snapmirror show -destination-path svm1_2:svm1_2_vol1

 To get more granular data, use the –fields option.

36


GET THE SNAPVAULT OR SNAPMIRROR RELATIONSHIP STATUS You can use the snapmirror show command to get various levels of detail about a SnapVault relationship. Use these commands on the destination cluster: cluster2::> snapmirror show Progress Source Path

Type Updated

--------

----

Destination Path ----------

Mirror State

Relationship Status

------- ------------

Total Progress

Last Healthy

---------

-------

svm1:svm1_vol1 XDP


-

true

Get more details by using the –destination-path or the –instance options: cluster2::> snapmirror show -destination-path svm1_2:svm1_2_vol1 Source Path: svm1:svm1_vol1 Destination Path: svm1_2:svm1_2_vol1 Relationship Type: XDP

5-46




-

Relationship Group Type: none SnapMirror Schedule: SnapVaultJob SnapMirror Policy Type: vault SnapMirror Policy: svm1_2_vault-policy Tries Limit: Throttle (KB/sec): unlimited Mirror State: Snapmirrored Relationship Status: Idle File Restore File Count: File Restore File List: Transfer Snapshot: Snapshot Progress: Total Progress: Network Compression Ratio: Snapshot Checkpoint: Newest Snapshot: snapmirror.1383c700-4345-11e497f5-0050560140c1_2147484678. 2014-09-23_182150 Newest Snapshot Timestamp: 09/23 16:16:29 Exported Snapshot: snapmirror.1383c700-4345-11e497f5-0050560140c1_2147484678. 2014-09-23_182150 Exported Snapshot Timestamp: 09/23 16:16:29 Healthy: true Unhealthy Reason: Constituent Relationship: false Destination Volume Node: cluster2-01 Relationship ID: 466c0484-434e-11e4-97f50050560140c1 Current Operation ID: Transfer Type: Transfer Error: Current Throttle: Current Transfer Priority: Last Transfer Type: initialize Last Transfer Error: Last Transfer Size: 21.31KB Last Transfer Network Compression Ratio: 1:1

5-47




Last Transfer Duration: 0:0:19 Last Transfer From: svm1:svm1_vol1 Last Transfer End Timestamp: 09/23 18:22:09 Progress Last Updated: Relationship Capability: 8.2 and above Lag Time: 2:13:5 Number of Successful Updates: 0 Number of Failed Updates: 0 Number of Successful Resyncs: 0 Number of Failed Resyncs: 0 Number of Successful Breaks: 0 Number of Failed Breaks: 0 Total Transfer Bytes: 21820 Total Transfer Time in Seconds: 19 Use the –fields option to get more granular data: cluster2::> snapmirror show -fields ? source-path

Source Path

source-cluster

Source Cluster

source-vserver

Source SVM

source-volume

Source Volume

destination-path

Destination Path

destination-cluster

Destination Cluster

destination-vserver

Destination SVM

destination-volume

Destination Volume

type

Relationship Type

relationship-group-type

Relationship Group Type

vserver

Managing SVM

schedule

SnapMirror Schedule

policy-type

SnapMirror Policy Type

policy

SnapMirror Policy

tries

Tries Limit

throttle

Throttle (KB/sec)

current-throttle Throttle (KB/sec)

Current Transfer

state

Mirror State

status

Relationship Status

file-restore-file-count

File Restore File Count

5-48




file-restore-file-list

File Restore File List

transfer-snapshot

Transfer Snapshot

snapshot-progress

Snapshot Progress

total-progress

Total Progress

network-compression-ratio Ratio

Network Compression

snapshot-checkpoint

Snapshot Checkpoint

newest-snapshot

Newest Snapshot

newest-snapshot-timestamp Timestamp

Newest Snapshot

exported-snapshot

Exported Snapshot

exported-snapshot-timestamp Timestamp

Exported Snapshot

healthy

Healthy

relationship-id

Relationship ID

current-operation-id

Current Operation ID

current-transfer-type

Transfer Type

current-transfer-error

Transfer Error

last-transfer-type

Last Transfer Type

last-transfer-error

Last Transfer Error

last-transfer-size

Last Transfer Size

last-transfer-network-compression-ratio Compression Ratio

Last Transfer Network

last-transfer-duration

Last Transfer Duration

last-transfer-from

Last Transfer From

last-transfer-end-timestamp Timestamp

Last Transfer End

unhealthy-reason

Unhealthy Reason

progress-last-updated

Progress Last Updated

relationship-capability

Relationship Capability

lag-time

Lag Time

current-transfer-priority Priority

Current Transfer

is-smtape-op

SMTape Operation

is-constituent

Constituent Relationship

destination-volume-node Name asynchronous

5-49

Destination Volume Node Synchronous/Asynchronous




identity-preserve

Identity Preserve SVM DR

update-succ-cnt Updates

Number of Successful

update-fail-cnt

Number of Failed Updates

resync-succ-cnt Resyncs


resync-fail-cnt

Number of Failed Resyncs

break-succ-cnt Breaks


break-fail-cnt

Number of Failed Breaks

tot-trans-bytes

Total Transfer Bytes

tot-trans-time-secs

Total Transfer Time in Seconds

5-50




Perform a SnapVault Manual Update The snapmirror update command must be used from the destination cluster: cluster2::> snapmirror update -source-path svm1:svm1_vol1 -destination-path svm1_2:svm1_2_vol1 -source-snapshot Emergency_SnapVault_Update

You can choose exactly which Snapshot copy on the source volume to use.

37


PERFORM A SNAPVAULT MANUAL UPDATE If you need to perform a SnapVault update in between regularly scheduled updates or instead of scheduled updates, use the snapmirror update command.   



The snapmirror update command updates the destination volume of a SnapMirror relationship. The snapmirror update command behaves differently for data protection, vault, and load-sharing relationships. For vault relationships, the snapmirror update does not take a Snapshot copy on the source volume but transfers only selected Snapshot copies that are newer than the common Snapshot copy to the destination volume. (Copies that are older than the common copy can be transferred by using the source-snapshot parameter.) Snapshot copies are selected by matching the value of -snapmirror-label of a Snapshot copy with the value of -snapmirror-label of one of the rules from the corresponding SnapMirror policy that is associated with the SnapMirror relationship.

The snapmirror update command must be used from the destination cluster: cluster2::> snapmirror update -source-path svm1:svm1_vol1 -destinationpath svm1_2:svm1_2_vol1 -source-snapshot Emergency_SnapVault_Update

5-51




Modify a SnapVault Relationship  Use the snapmirror modify command to change an existing SnapVault relationship.  Change the update schedule to use the 5min schedule: cluster2::> snapmirror modify -destination-path svm1_2:svm1_2_vol1 -schedule 5min

38


MODIFY A SNAPVAULT RELATIONSHIP The snapmirror modify command enables you to change one or more properties of SnapMirror relationships. The key parameter that identifies any SnapMirror relationship is the destination volume. Changes that are made by the snapmirror modify command do not take effect until the next manual or scheduled update of the SnapMirror relationship. Changes do not affect updates that are started but are not yet finished. Change the update schedule to use the 5min schedule. cluster2::> snapmirror modify -destination-path svm1_2:svm1_2_vol1 schedule 5min Operation succeeded: snapmirror modify for the relationship with destination "svm1_2:svm1_2_vol1".

5-52




Modify a SnapVault Policy Use the snapmirror policy modify command to modify the policy attributes: -comment -tries -transfer-priority -ignore-atime -restart

39


MODIFY A SNAPVAULT POLICY You can use the snapmirror policy modify command to modify the policy attributes: -comment

Specify a text comment for the policy.

-tries

Specify the maximum number of times to attempt each manual or scheduled transfer.

-transfer-priority

Normal or low, normal transfers are scheduled before low-priority transfers.

-ignore-atime

True or false. Ignores the file access time.

-restart

Always, never, or default. Applies only to data protection relationships. A setting of “default” is interpreted the same as a setting of “always.” Vault transfers always resume based on a restart checkpoint, if the source Snapshot copy still exists.

5-53




Modify and Remove SnapVault Policy Rules

40

-keep

This rule specifies the maximum number of Snapshot copies that are retained on the SnapMirror vault destination volume for a rule.

-preserve

This rule can be true or false. It specifies the behavior when the Snapshot copy retention count is reached on the SnapMirror vault destination for the rule. The default value is false.

-warn

This rule specifies the warning threshold count for the rule. The default value is 0.

-vserver

This rule specifies the SVM for the SnapMirror policy.

-policy

This rule specifies the SnapMirror policy name.


MODIFY AND REMOVE SNAPVAULT POLICY RULES You can use the snapmirror policy modify-rule command to modify the retention count, preserve setting, and warning threshold count for a rule in a SnapMirror policy. If you reduce the retention count or disable the preserve setting for a rule in a SnapMirror policy, it might result in the deletion of Snapshot copies on the vault destination when the next transfer by the snapmirror update command occurs. -keep

Specifies the maximum number of Snapshot copies that are retained on the SnapMirror vault destination volume for a rule. The total number of Snapshot copies that are retained for all the rules in a policy cannot exceed 251.

-preserve

True or false. Specifies the behavior when the Snapshot copy retention count is reached on the SnapMirror vault destination for the rule. The default value is false, which means that the oldest Snapshot copy is deleted to make room for new copies, only if the number of Snapshot copies has exceeded the retention count that is specified in the "keep" parameter. When set to true, an incremental SnapMirror vault update will fail, when the Snapshot copies reach the retention count.

-warn

Specifies the warning threshold count for the rule. The default value is 0. When set to a value greater than zero, an event is generated after the number of Snapshot copies (for the particular rule) that are retained on a SnapMirror vault destination reaches the specified warn limit. The preserve parameter for the rule must be true, to set the warn parameter to a value that is greater than zero.

5-54




The snapmirror policy remove-rule command removes a rule from a SnapMirror policy. On the vault destination, all Snapshot copies with a SnapMirror label matching the rule are no longer processed by the snapmirror update command and might need to be deleted manually. A SnapMirror policy that is associated with a SnapMirror vault relationship must have at least one rule. -vserver

Specifies the SVM for the SnapMirror policy.

-policy

Specifies the SnapMirror policy name.

5-55




References  NetApp University courses: Data ONTAP 8.3 istration

 NetApp Documentation:  Clustered Data ONTAP 8.3 Data Protection Guide  Clustered Data ONTAP 8.3 Data Protection Tape Backup and Recovery Guide  NetApp Technical Reports:

 Operational How-To Guide: NetApp Snapshot Management TR-4210  NetApp Data Compression and Deduplication Deployment and Implementation Guide TR-3966

41


REFERENCES

5-56





Now I know the differences between SnapVault and SnapMirror technology. I learned how to design and implement a SnapVault solution. I know how to recover my data using SnapVault software, and I learned how to monitor and manage my SnapVault and SnapMirror relationships.

42



5-57




Exercise

Module 5: SnapVault Backup and Recovery


43



5-58




Module 6

Data Protection Using NDMP

1


MODULE 6: DATA PROTECTION USING NDMP

6-1

Data ONTAP Data Protection istration: Data Protection Using NDMP




 Describe how clustered Data ONTAP uses the NDMP protocol  Describe design criteria for NDMP solutions

 Implement the NDMP protocol to back up clustered Data ONTAP 8.3 storage systems  Monitor and manage NDMP-based operations from the perspective of clustered Data ONTAP 8.3

2


MODULE OBJECTIVES

6-2




Lesson 1

NDMP Technology

3


LESSON 1: NDMP TECHNOLOGY

6-3




NDMP Technology NDMP is an industry-standard protocol that can control backup, recovery, and data transfer between primary and secondary storage devices. All communications occur over T/IP or T/IPv6. NDMP provides low-level control of tape drives and medium changers.

4


NDMP TECHNOLOGY NDMP is an industry-standard protocol for controlling backup, recovery, and data transfer between primary and secondary storage devices, including storage systems and tape libraries. Enabling the NDMP protocol on a NetApp storage system enables that storage system to communicate with NDMP-enabled backup applications.

6-4




Ken Asks

I don’t understand the NDMP terminology. Can you explain it?

5


KEN ASKS

6-5




NDMP Key and Concepts  Data management application (DMA)

 Direct access recovery (DAR)  Cluster Aware Backup (CAB) extension  Connection address extension (CAE)  Affinity

6


NDMP KEY AND CONCEPTS Data management application In the context of the NDMP protocol, DMA refers to your backup application. Direct access recovery DAR enables quick access to the secondary media during a recovery operation. In Data ONTAP 8.3, enhanced DAR functionality is enabled by default. Enhanced DAR allows for directory DAR and DAR of files with NT streams. You can enable or disable enhanced DAR in both node-scoped and storage virtual machine (SVM)-scoped NDMP modes. CAB extension The CAB extension is an NDMPv4 protocol extension. This extension enables the NDMP server to establish a data connection on a node that owns a volume. This extension also enables the backup application to determine whether volumes and tape devices are located on the same node in a cluster. To enable the NDMP server to identify the node that owns a volume and to establish a data connection on such a node, the backup application must the CAB extension. The CAB extension requires the backup application to inform the NDMP server about the volume to be backed up or restored, before establishing the data connection. This requirement enables the NDMP server to determine which node hosts the volume and to appropriately establish the data connection. When the backup application s the CAB extension, the NDMP server provides affinity information about volumes and tape devices. Using this affinity information, the backup application can perform a local backup instead of a three-way backup, if a volume and tape device are located on the same node in a cluster. Connection address extension The CAE is used for IPv6 .

6-6




Affinity When the backup application s the CAB extension, the NDMP server provides unique location information about volumes and tape devices. Using this affinity information, the backup application can perform a local backup instead of a three-way backup, if a volume and a tape device share the same affinity. If the volume moves from Node 1 to Node 2, affinity information about the volume and tape device changes. Hence, for a subsequent backup, the DMA performs a three-way NDMP backup operation. This ensures continuity of the backup policy for the volume, irrespective of the node to which the volume is moved.

6-7




NDMP Modes, Connections, and Variables  Node-scoped NDMP mode

 SVM-scoped NDMP mode  NDMP control connection  NDMP data connection  NDMP environment variables

7


NDMP MODES, CONNECTIONS, AND VARIABLES Node-scoped NDMP mode Node-scoped NDMP mode enables you to perform tape backup and restore operations at the node level. You must establish the NDMP control connection on a logical interface (LIF) that is hosted on the node that owns the volume or tape devices. NOTE: Node-scoped NDMP mode is deprecated, and it will be removed in a future major release of Data ONTAP. SVM-scoped NDMP mode SVM-scoped NDMP mode enables you to back up and restore all volumes that are hosted across different nodes in an SVM, as long as the backup application s the CAB extension. If your backup application s the CAB extension, and if a volume and tape device share the same affinity, then the backup application can perform a local backup or restore operation instead of a three-way restore operation. NOTE: The NDMP control connection can be established on a data or LIF only if the NDMP service is enabled on the SVM that owns the LIF. SVM-scoped NDMP mode and Data ONTAP upgrades or installations An upgrade of Data ONTAP from 8.1 to 8.3 causes NDMP to follow the node-scoped behavior. You can explicitly disable node-scoped NDMP mode, so that tape backup and restore operations are performed in SVM-scoped NDMP mode. A new installation of Data ONTAP 8.3 causes NDMP to follow SVM-scoped mode by default. You can perform node-scoped NDMP operations if you explicitly enable node-scoped NDMP mode. NDMP control connection The NDMP control connection is used to manage NDMP backup and restore requests and replies. 6-8




NDMP data connection The NDMP data connection is used only to transfer data. NDMP environment variables NDMP environment variables are used to communicate information about a backup or restore operation between an NDMP-enabled backup application and a storage system. Typically, the backup application sets the environment variables automatically. However, to unique circumstances, the backup can set some environment variables manually. A backup rarely specifies environment variables; however, you might want to change the value of an environment variable in order to characterize or work around a functional or performance problem. Many backup applications provide a means to override or modify environment variables or to specify additional environment variables. For information, see your backup application documentation. Data ONTAP s environment variables that have an associated default value. However, you can manually modify these default values. For a complete list of environment variables that are ed for SMTape and dump operations, see the Clustered Data ONTAP 8.3 Data Protection Tape Backup and Recovery Guide.

6-9




Node-Scoped NDMP  Behaves much like NDMP in Data ONTAP operating in 7-Mode  Requires NDMP to be enabled and configured per node: The NDMP must be set and defined on each node.

 Uses only physical resources and is not aware of other nodes or any logical management: Node-scoped NDMP uses clusters or SVMs.

 Requires the DMA to use a LIF on the node that hosts the desired volume: The DMA backup policy must be updated if the volume or LIF move to a different node.

 Cannot use the NDMP CAB extension Node-scoped NDMP mode is deprecated, and it will be removed in a future major release of Data ONTAP. 8


NODE-SCOPED NDMP

6-10




SVM-Scoped NDMP  NDMP is now enabled and configured per SVM: NDMP is in the “allowed protocols” list by default.

 Authentication is now integrated with role-based access control (RBAC): This can be a single cluster-wide (like “”).

 SVM-scoped NDMP introduces LIF limitations and restrictions:*  NDMP must use an appropriate LIF type for some resources.  Some resource limitations are negated by a CAB extension.

 With CAB, NDMP can back up any volume on any node and use any tape device in the entire cluster.

9


SVM-SCOPED NDMP *Refer to NetApp KB ID: 1014598: How to identify which resources are available through NDMP based on LIF type

6-11




Lesson 2

Design Criteria for NDMP Solutions

10


LESSON 2: DESIGN CRITERIA FOR NDMP SOLUTIONS

6-12




NDMP Design Considerations  File history

 Firewall policy  LIF connections  CAB extension

 Format of the NDMP backup path  Use of the ndmpcopy command  Backup of SnapMirror destination data 11


NDMP DESIGN CONSIDERATIONS When you start the NDMP service on the storage system, note the following design considerations. File history NDMP services can generate file history data at the request of NDMP backup applications. File history is used by backup applications to enable optimized recovery of selected subsets of data from a backup image. File history generation and processing might be time-consuming and U-intensive for both the storage system and the backup application. NOTE: SMTape does not file history. If your data protection is configured for disaster recovery—where the entire backup image is recovered—you can disable file history generation to reduce backup time. See your backup application documentation to determine if it is possible to disable NDMP file history generation. Firewall policy The firewall policy for NDMP is enabled by default on all LIF types. In node-scoped NDMP mode, to back up a FlexVol volume, you must use the backup application to initiate a backup on a node that owns the volume. However, you cannot back up a node root volume. LIF connections You can perform NDMP backup from any LIF, as permitted by the firewall policies. If you use a data LIF, you must select a LIF that is not configured for failover. If a data LIF fails over during an NDMP operation, the NDMP operation fails and must be rerun. CAB extension In node-scoped NDMP mode and SVM-scoped NDMP mode with no CAB extension , the NDMP data connection uses the same LIF as the NDMP control connection. 6-13




Format of the NDMP backup path The NDMP backup path uses the format IP_address:/vserver_name/volume_name/path_name, where path_name is the path of the directory, file, or Snapshot copy. Using the ndmpcopy command The ndmpcopy command transfers data between storage systems using the NDMP v4 protocol. Both full and incremental data transfers can be performed. You can transfer full or partial volumes, qtrees, directories, or individual files. You can run ndmpcopy at the command line of the source or destination storage systems, or of a storage system that is neither the source nor destination of the data transfer. You can also run ndmpcopy on a single storage system that is both the source and destination of the data transfer. For the syntax and examples of the ndmpcopy command, see the Clustered Data ONTAP 8.3 Data Protection Tape Backup and Recovery Guide. Backing up SnapMirror destination data When a SnapMirror destination is backed up to tape, only the data on the volume is backed up. The SnapMirror relationships and the associated metadata are not backed up to tape. Therefore, during restore, only the data on that volume is restored; the associated SnapMirror relationships are not restored.

6-14




NDMP Backup Models  Clustered Data ONTAP s three models for NDMP backups:  Direct (local)  Indirect (remote)  Three-way

 The backup model is important because it defines “who is responsible for what.”

12


NDMP BACKUP MODELS

6-15




Local NDMP Backup DMA

Control Connection NDMP Control Connection LIF

Tape and Data Service

Cluster

Node 1

Node 2

Data Connection

13


LOCAL NDMP BACKUP

6-16




Remote NDMP Backup DMA Data Connection

Control Connection NDMP Control Connection LIF

Tape and Data Service

Cluster

Node 1

14

Node 2


REMOTE NDMP BACKUP

6-17




Three-Way NDMP Backup DMA

Control Connection Tape Service

Data Service

Cluster

NDMP Control Connection LIF

Node 1

NDMP Data Connection LIF

Node 2

Data Connection

15


THREE-WAY NDMP BACKUP

6-18




ed NDMP Authentication Methods  Plaintext

 Challenge

16


ED NDMP AUTHENTICATION METHODS In node-scoped NDMP mode, both authentication methods are enabled by default: challenge and plaintext. You can disable plaintext, but you cannot disable challenge. In the plaintext authentication method, the is transmitted as clear text. In SVM-scoped NDMP mode, the default authentication method is challenge; you can choose to enable or disable plaintext or challenge. However, one authentication mode must be enabled.

6-19




Authentication in SVM-Scoped NDMP Mode NIS LDAP

NDMP name and RBAC

17

SVM

Cluster

vs

vs-backup

backup


AUTHENTICATION IN SVM-SCOPED NDMP MODE In SVM-scoped NDMP mode, NDMP authentication is integrated with RBAC. In the SVM context, the NDMP must have either the “vs” or “vs-backup” role. In a cluster context, the NDMP must have either the “” or “backup” role. In SVM-scoped NDMP mode, you must generate an NDMP for a given , which is created based on RBAC roles. Cluster s in an or backup role can access a node-management LIF, cluster-management LIF, or an intercluster LIF. s in a vs-backup or vs role can access only the data LIF for that SVM. Therefore, depending on the role of a , the availability of volumes and tape devices for backup and restore operations vary. This mode also s authentication for Network Information Service (NIS) and Lightweight Directory Access Protocol (LDAP) s. Therefore, NIS and LDAP s can access multiple SVMs with a common ID and . However, NDMP authentication does not Active Directory s. In SVM-scoped NDMP mode, a must be associated with the Secure Shell (SSH) application and the “ ” authentication method.

6-20




Scalability Limits for NDMP Sessions

System Memory of a Storage System

18

Maximum Number of NDMP Sessions

Less than 16 gigabytes (GB)

8

Greater than or equal to 16 GB but less than 24 GB

20

Greater than or equal to 24 GB

36


SCALABILITY LIMITS FOR NDMP SESSIONS You must be aware of the maximum number of NDMP sessions that can be established simultaneously on storage systems of different system memory capacities. This maximum number depends on the system memory of a storage system.

6-21




Lesson 3

Managing Node-Scoped NDMP

19


LESSON 3: MANAGING NODE-SCOPED NDMP

6-22




Commands for Managing Node-Scoped NDMP Mode Use This Command

To Do This Action

system services ndmp on*

Enable NDMP service.

system services ndmp off*

Disable NDMP service.

system services ndmp show

Display NDMP configuration.

system services ndmp modify

Modify NDMP configuration.

system services ndmp version*

Display the default NDMP version.

system services ndmp status*

Display all the NDMP sessions.

cluster1::> system services ndmp version Node Version *Commands that are marked with an --------------------asterisk on this slide are deprecated and will be removed in a future major cluster1-01 4 release. cluster1-02 4 20


COMMANDS FOR MANAGING NODE-SCOPED NDMP MODE You can use the system services ndmp commands to manage NDMP at a node level. On the slide, the commands with an asterisk are deprecated and will be removed in a future major release. For more information, see the man pages for the system services ndmp commands.

6-23




Commands for Node-Scoped NDMP Mode Advanced Privilege Level

Use This Command

To Do This Action

system services ndmp terminate

Forcefully terminate all the NDMP sessions.

system services ndmp start

Start the NDMP service daemon.

system services ndmp stop

Stop the NDMP service daemon.

system services ndmp log start*

Start logging for the specified NDMP session.

system services ndmp log stop*

Stop logging for the specified NDMP session.

cluster1::*> system services ndmp start -node cluster1-01 *Commands that are marked with an

asterisk on this slide are deprecated and will be removed in a future major release.

21


COMMANDS FOR NODE-SCOPED NDMP MODE: ADVANCED PRIVILEGE LEVEL You can use the following NDMP commands only at the advanced privilege level:     

system system system system system

services services services services services

ndmp ndmp ndmp ndmp ndmp

terminate start stop log start log stop

On the slide, the commands with an asterisk are deprecated and will be removed in a future major release. For more information, see the man pages for the system services ndmp commands.

6-24




Commands for Node-Scoped NDMP Mode Use This Command

To Do This Action

system services ndmp probe*

system services ndmp kill-all*

Display detailed information about all the NDMP sessions. Terminate the specified NDMP session. Terminate all the NDMP sessions.

system services ndmp *

Change the NDMP .

system services ndmp node-scope-mode on* system services ndmp node-scope-mode off* system services ndmp node-scope-mode status*

Enable node-scoped NDMP mode.

system services ndmp kill*

Disable node-scoped NDMP mode.

Display the node-scoped NDMP mode status.

cluster1::*> system services ndmp node-scope-mode status NDMP node-scope-mode is disabled. 22


COMMANDS FOR NODE-SCOPED NDMP MODE The commands with an asterisk are deprecated and will be removed in a future major release. For more information, see the man pages for the system services ndmp commands.

6-25




Lesson 4

Managing SVM-Scoped NDMP

23


LESSON 4: MANAGING SVM-SCOPED NDMP

6-26




SVM-Aware NDMP Remote NDMP Server

CAB-Enabled DMA Data or Intercluster LIF

Control Connection

Control Connection Data or Intercluster LIF

Data Connection

Target Volume

24


SVM-AWARE NDMP With Data ONTAP 8.3, you can choose to perform tape backup and restore operations at the SVM level. For NDMP to be aware of an SVM, the NDMP data management application software must be enabled with CAB extensions, and the NDMP service must be enabled on the SVM. After the feature is enabled, you can back up and restore all volumes that are hosted across all nodes in the SVM. An NDMP control connection can be established on different LIF types. An NDMP control connection can be established on any data or intercluster LIF that is owned by an SVM that is enabled for NDMP and that owns the target volume. If a volume and tape device share the same affinity, and if the data management application s the CAB extensions, then the backup application can perform a local backup or restore operation. Therefore, you do not need to perform a three-way backup or restore operation.

6-27




Commands for Managing SVM-Scoped NDMP Mode

25

Use This Command

To Do This Action

vserver services ndmp on

Enable NDMP service.

vserver services ndmp off

Disable NDMP service.

vserver services ndmp show

Display NDMP configuration.

vserver services ndmp modify

Modify NDMP configuration.

vserver services ndmp version

Display the default NDMP version.

vserver services ndmp status

Display all NDMP sessions.

vserver services ndmp probe

Display detailed information about all NDMP sessions.


COMMANDS FOR MANAGING SVM-SCOPED NDMP MODE You can manage NDMP on a per-SVM basis by using the NDMP options and commands. You can modify the NDMP options by using the vserver services ndmp modify command. In SVM-scoped NDMP mode, authentication is integrated with the RBAC mechanism. You can add NDMP in the allowed or disallowed protocols list by using the vserver modify command. By default, NDMP is in the allowed protocols list. If NDMP is added to the disallowed protocols list, NDMP sessions cannot be established. You can control the LIF type on which an NDMP data connection is established by using the -preferredinterface-role option. During an NDMP data connection establishment, NDMP chooses an IP address that belongs to the LIF type as specified by this option. If the IP addresses do not belong to any of these LIF types, then the NDMP data connection cannot be established. You can use the vserver services ndmp commands to manage NDMP on each SVM.

6-28




Try This  In your lab kit, to cluster 1.  Enter: vserver services ndmp version Which NDMP version does the cluster ?

 Enter: vserver services ndmp show  Is NDMP enabled for the cluster?  What is the current authentication type?

 Enter: system services ndmp Why are many of these commands deprecated?

 Enter: system services ndmp version Why is this output different from the vserver command that was entered

26


TRY THIS

6-29




earlier?

Commands for Managing SVM-Scoped NDMP Mode cluster1::*> vserver add-protocols -vserver svm1 -protocols ndmp Add the NDMP protocol to the SVM.

cluster1::*> vserver services ndmp on -vserver svm1 Enable the NDMP service for the SVM.

cluster1::*> vserver services ndmp show VServer Enabled Authentication type ------------- --------- ------------------cluster1 false challenge Check that NDMP is enabled for the SVM. svm1 true challenge 27


COMMANDS FOR MANAGING SVM-SCOPED NDMP MODE

6-30




Commands for Managing SVM-Scoped NDMP Mode (Cont.) Use This Command

To Do This Action

vserver services ndmp kill

Terminate a specified NDMP session.

vserver services ndmp kill-all

Terminate all NDMP sessions.

vserver services ndmp generate

Generate the NDMP .

vserver services ndmp extensions show (advanced)

Display the NDMP extension status.

vserver services ndmp extensions modify (advanced)

Modify (enable or disable) the NDMP extension status.

vserver services ndmp log start (advanced)

Start logging for the specified NDMP session.

vserver services ndmp log stop (advanced)

Stop logging for the specified NDMP session.

*Commands that are marked with an asterisk on this slide are available only at the advanced privilege level.

28


COMMANDS FOR MANAGING SVM-SCOPED NDMP MODE (CONT.) Commands that are marked as (advanced) on this slide are available only at the advanced privilege level.

6-31




References  NetApp University course: Clustered Data ONTAP 8.3 istration

 NetApp Documentation:  Clustered Data ONTAP 8.3 Data Protection Guide  Clustered Data ONTAP 8.3 Data Protection Tape Backup and Recovery Guide

 NetApp KB ID: 1014598. How to identify which resources are available through NDMP based on LIF type.

29


REFERENCES

6-32





I had no idea I could use the NDMP protocol in so many useful ways. I’m looking forward to getting my applications set up for SVM-scoped backup.

30



6-33




Course Post-Test Please take some time now to complete the course post-test.

Taking the same exam again measures how much you learned. All scores will remain private.

31


COURSE POST-TEST

6-34




Your Learning Journey Continues Visit Regularly to Stay Informed

NetApp University

NetApp

NetApp University Overview

New to NetApp Webcast

 Find the training you need.

Ensure a successful experience.

 Explore certification.

NetApp

 Follow your learning map.

NetApp University Community the discussion.

Customer Success Community Engage with experts.

NetApp University

Welcome to Knowledgebase

the team.

32

Access s, tools, and documentation.

Access a wealth of knowledge.


YOUR LEARNING JOURNEY CONTINUES: VISIT REGULARLY TO STAY INFORMED The NetApp University Overview page is your front door to learning. Find training that fits your learning map and your learning style, learn how to become certified, link to blogs and discussions, and subscribe to the NetApp newsletter Tech OnTap. http://www.netapp.com/us/services-/university/index.aspx The NetApp University Community page is a public forum for NetApp employees, partners, and customers. NetApp University welcomes your questions and comments! https://communities.netapp.com/community/netapp_university The NetApp University page is a self-help tool that enables you to search for answers and to the NetApp University team. http://netappu.custhelp.com Are you new to NetApp? If so, for the New to NetApp Webcast to acquaint yourself with facts and tips that ensure that you have a successful experience. http://www.netapp.com/us/forms/webcastseries.aspx?REF_SOURCE=new2ntapwl-netappu The NetApp page is your introduction to all product and solution : http://my.netapp.com. Use the Getting Started link (http://my.netapp.com/info/web/ECMP1150550.html) to establish your and hear from the NetApp CEO. Search for products, s, tools, and documentation, or link to the NetApp Community (http://community.netapp.com/t5/Products-and-Solutions/ct-p/products-and-solutions). the Customer Success Community to ask -related questions, share tips, and engage with other s and experts. https://forums.netapp.com/ 6-35




Search the NetApp Knowledgebase to leverage the accumulated knowledge of NetApp s and product experts. https://kb.netapp.com//index?page=home

6-36




Appendix A

Open Systems SnapVault

1


APPENDIX A: OPEN SYSTEMS SNAPVAULT



 Describe how Open Systems SnapVault integrates with Data ONTAP SnapVault  Describe Open Systems SnapVault advanced features

 Configure and ister Open Systems SnapVault  Perform Open Systems SnapVault backup and restore operations  Troubleshoot and resolve Open Systems SnapVault transfer failures 2


MODULE OBJECTIVES

2

Appendix A: Open Systems SnapVault



Lesson 1

Open Systems SnapVault Fundamentals

3


LESSON 1: OPEN SYSTEMS SNAPVAULT FUNDAMENTALS

3




Open Systems SnapVault  The primary storage system is a Microsoft Windows system or a UNIX-based open system.  The Open Systems SnapVault agent is installed on the primary system. The agent enables the system to back up its data to the secondary storage system.  The secondary system uses Data ONTAP SnapVault technology to back up the primary system data. 4

ESX Server iSCSI Linux Server

Windows Server

UNIX Server Primary Systems Running Open Systems SnapVault

FC

SnapVault Secondary Storage System


OPEN SYSTEMS SNAPVAULT Open Systems SnapVault is a disk-to-disk data protection solution that uses NetApp SnapVault technology to protect data that resides on these platforms:       

Microsoft Windows Red Hat Enterprise Linux Novell SUSE Linux Enterprise Server Sun Solaris IBM AIX HP HP-UX VMware ESX Server

For a complete list of currently ed versions of these platforms, refer to the Open Systems SnapVault 2.6 Installation and istration Guide on the NetApp site. The Open Systems SnapVault agent software module is installed on the primary storage system. The agent enables the system to back up its data to the secondary storage system. The secondary system is a NetApp storage appliance, such as a NearStore storage system. The secondary system uses Data ONTAP SnapVault technology to back up the data from the primary storage system.

4




Theory of Operation  The directory is the basic unit of Open Systems SnapVault backup.

 An incremental update is based on file modification time. If blocklevel incremental (BLI) backup is enabled, only changed blocks are transferred.  Creation of the Snapshot copy for archiving takes place on the SnapVault secondary storage system. Primary Systems Windows Server

SnapVault Secondary Storage System

C:\data

/vol/vol1 qtree_data

UNIX Server

5

qtree_usr

/usr

Online and Writable Volume Read-Only Qtrees Replica


THEORY OF OPERATION Each directory, file, or drive that you want to back up from a primary system is backed up to its own qtree on the SnapVault secondary storage system. For the first backup, you perform an initial baseline transfer of the identified directory or file. This transfer establishes the SnapVault relationship between the Open Systems SnapVault primary directory or file and its mapped SnapVault secondary qtree. You can initiate subsequent transfers manually, or you can configure transfers for automatic scheduled updates. Using the Data ONTAP snapvault command, you specify the schedules on the secondary system. You can also use an NDMP-based management application, such as Protection Manager or a thirdparty NDMP-based application that is ed. When an incremental update request occurs, the Open Systems SnapVault primary storage system determines whether the directory or file changed since the last successful transfer. This determination is made by examining the file modification time. If BLI is enabled (for Windows platforms only), Open Systems SnapVault determines on which blocks data changed. Then, Open Systems SnapVault sends only those modified blocks to the secondary system. After the secondary qtree is updated, SnapVault creates a Snapshot copy of the volume for archiving. If a directory or file data needs to be restored to the primary system, SnapVault retrieves the data from one specified Snapshot copy and transfers the data back to the primary system that requests it.

5




Central Management Use these NDMP-based applications to centrally manage Open Systems SnapVault backup schedules, retention policies, backup control, and monitoring:  NetApp OnCommand Unified Manager  BakBone NetVault  Syncsort Backup Express  CommVault Galaxy

6


CENTRAL MANAGEMENT Open Systems SnapVault can be managed from a variety of management applications. These applications use the NDMP protocol and T port 10000 for communication between an Open Systems SnapVault client and the storage systems, over a T/IP network. Backup schedules, retention policies, backup control, and monitoring are centrally configured on applications such as NetApp SnapProtect management software, NetApp OnCommand Unified Manager, or third-party management programs.

6




Integrated Backup and Disaster Recovery Solution  You can use SnapMirror software to replicate SnapVault backup copies to a disaster recovery site.  If the SnapVault secondary storage system becomes unusable, you can redirect the subsequent Open Systems SnapVault transfers to the tertiary storage system. Data Center

Disaster Recovery Site

Primary Open Systems SnapVault Systems

SnapVault

Volume SnapMirror

SnapVault 7


INTEGRATED BACKUP AND DISASTER RECOVERY SOLUTION The Open Systems SnapVault primary systems provide the same advantages as other primary systems provide in a data protection and disaster recovery scenario. You can use the SnapMirror feature to protect the SnapVault secondary system from disasters. To configure, set up SnapMirror relationships from the volumes on the SnapVault secondary system to volumes on a remote (tertiary) NetApp storage system, as shown. If the SnapVault secondary system becomes unusable because of a disaster, you can manually redirect the subsequent Open Systems SnapVault transfers to the tertiary system. In effect, the tertiary system becomes the new SnapVault secondary system. The Open Systems SnapVault transfers use the most recent Snapshot copy that is common to both the primary and tertiary storage systems.

7




Lesson 2

Open Systems SnapVault Features

8


LESSON 2: OPEN SYSTEMS SNAPVAULT FEATURES

8




Snapshot Copy Management  Open Systems SnapVault can retain Snapshot copies.

 Snapshot copy management commonly uses two possible configurations:

MaxRestartWaitTime  Sets the maximum waiting time a Snapshot copy is retained after a transfer failure; the default is 10 minutes FailRestartOnNewSnapshot  If a Snapshot copy is not available during restart, the following occurs:  When set to FALSE, a new Snapshot copy is created, and the transfer continues.  When set to TRUE, the transfer is aborted.

 To change these values, use the svsetstanza command or edit the snapvault.cfg file on the Open Systems SnapVault primary system. 9


SNAPSHOT COPY MANAGEMENT Common Snapshot copy management ensures that the same Snapshot copy is used for backup. In releases earlier than Open Systems SnapVault 2.5, whenever the transfer of files failed, the Snapshot copies were deleted. A new Snapshot copy was created during the transfer restart. Open Systems SnapVault 2.5 can retain old Snapshot copies and use them during subsequent transfer restarts.

9




Checkpoints for Restart  Enable a failed transfer (baseline or update) to resume from the last valid recorded checkpoint  Are not used to resynchronize a broken relationship or to resume operations after a restore  checkpoints at block levels inside files  Enable a failed transfer to restart even from the middle of a file

 Enable configuration of the checkpoint interval:  Default is 300 seconds (5 minutes).  Minimum is 60 seconds. 10


CHECKPOINTS FOR RESTART Checkpoints are recorded by the Open Systems SnapVault primary system when certain predetermined conditions or periodic intervals are met. The Open Systems SnapVault primary system records the checkpoints and sends the checkpoints to the SnapVault secondary system. In releases earlier than Open Systems SnapVault 2.5, checkpoints were taken at five-minute intervals or at the end of a file. If a transfer failure for a large file occurred, this configuration made it necessary to send all the file data again. Open Systems SnapVault 2.5 and later releases the following improvements in the checkpoint mechanism:   

10

Checkpoints are enabled at block levels inside files: This enhancement is useful when the dataset contains large files (greater than 100 megabytes [MB]). Checkpoints are allowed inside files: Therefore, you can restart the transfer even from the middle of a file. Checkpoint interval configuration is enabled: You can configure the checkpoint interval by editing the snapvault.cfg file on the Open Systems SnapVault primary system.




Block-Level Incremental Backup  From the information in the timestamp and checksum algorithm, BLI backup recognizes that a file changed.  Name-based BLI backup creates file modification issues.  Open Systems SnapVault agents address known issues of file modification.

11


BLOCK-LEVEL INCREMENTAL BACKUP  





11

BLI backup determines exactly which blocks in the file are changed, then backs up only those blocks to the SnapVault secondary system. Checksum database files are stored in the internal database directory of Open Systems SnapVault. Each Open Systems SnapVault relationship contains its own checksum file directory. Disk storage requirements to maintain the computed checksum values on the primary storage system are expected to be approximately 2% of the baseline backup size. Name-based applications, such as Microsoft Word, Excel, and PowerPoint, modify files by inserting new data blocks in the file and rewriting all subsequent data blocks in the file to new positions in the file. Because the modified file is considered new, a backup of all the rewritten blocks and a recalculation of checksum are required. Open Systems SnapVault agents work around this issue by recognizing files by name and by the filesystem location. You can use the Open Systems SnapVault Configurator to disable BLI backups for certain name-based applications.




Block-Level Incremental Modes  BLI backup operates in three possible modes:  OFF: No checksums are computed. Full files are transferred after they are identified as changed files.  HIGH (default): Checksums are computed on the initial transfers and incremental update.  LOW: Checksums are computed on the first and subsequent incremental updates.

 You can configure the BLI backup values in the Open Systems SnapVault Configurator.

12


BLOCK-LEVEL INCREMENTAL MODES

12




Exclusion Lists and Encrypted Files  Backup exclusion lists exclude specified files, directories, subdirectories, and paths from backups.  Open Systems SnapVault can back up and restore Encrypted File System (EFS) files on Windows platforms:  BLI backup does not the backup of EFS files.  Sufficient free space must be available in the target volume to restore EFS files.

13


EXCLUSION LISTS AND ENCRYPTED FILES The Open Systems SnapVault agent uses backup exclusion lists to exclude specified files, directories, subdirectories, and entire paths from backups. Open Systems SnapVault s two types of exclusion lists:  

File exclusion list entries consist of single-path elements. A file or directory is excluded if the file name or any path element matches a file exclusion entry in the list. The file exclusion list is in the install_dir/etc/file-exclude.txt file. Path exclusion list entries consist of complete file system paths to either a directory or a file. The path exclusion list is in the install_dir/etc/path-exclude.txt file.

Refer to the Open Systems SnapVault Installation and istration Guide for details on ed characters and wildcards for those file entries. EFS-encrypted file restore requires the Open Systems SnapVault agent to create a temporary file. The file must be equal to or greater than the size of the EFS-encrypted file that is being replaced.

13




Open Systems SnapVault Database  The Open Systems SnapVault database is a set of files that describe the relationship between primary and secondary systems:  History file  BLI checksums file (if BLI transfer is enabled)  Checkpoint file (if a backup transfer has failed with a checkpoint)

 On the primary system, the database resides in install_dir\snapvault\db\QsmDatabase.  On the secondary system, database backup is stored under the qtree root as OSSV_DATABASE_BACKUP.

14


OPEN SYSTEMS SNAPVAULT DATABASE Three backup levels of the Open Systems SnapVault database:   

14

BLI (default): Backs up the history file and its corresponding BLI checksum file DB only: Backs up only the history file NONE: Disables the automatic database backup




Restoring the Open Systems SnapVault Database  If the Open System SnapVault database becomes corrupt or gets out of sync with the secondary system, data transfers fail. If you have a valid backup, you can restore the database.  To restore the database, issue the snapvault restore command from the Open Systems SnapVault primary system, and include the file name Open Systems SnapVault_DATABASE_BACKUP in the path.  The Open Systems SnapVault database can be restored without the need to initiate a baseline transfer.  If the file system was modified between backup and restore, you must resynchronize the relationship. 15


RESTORING THE OPEN SYSTEMS SNAPVAULT DATABASE After a database file is restored, Open Systems SnapVault decompresses the database file and places the files where Open Systems SnapVault database files are located for the relationship. Data transfers can be performed from this point onward. However, if the file system was modified between the backup and the restore operations, you must resynchronize the relationship.

15




System State and Event Log Backup  Open Systems SnapVault s Windows 2000 Server, Windows Server 2003, and Windows 2008 system state backup and restore:  The system can be restored to an earlier state.  Subsequent backups use BLI, when it is enabled.

 System state backup can include a Windows Event Log:  Logs can be used for troubleshooting problems or for capacity planning.  Configurable backup options are Application, Security, and System event log backup.  Incremental updates of Windows Event Log files are not ed. 16


SYSTEM STATE AND EVENT LOG BACKUP You can back up and restore various components of the Windows 2000 Server and Windows Server 2003 system state:        

Registry COM+ Class Registration database System files and settings, including the boot files Certificate Services database IIS Metadirectory System files that are under Windows file protection Performance counters System state data on domain controllers (Active Directory and SYSVOL data)

You can add backups of Windows system state data to existing Open Systems SnapVault backup schedules. Then you can use the backups to restore a system to an earlier state. You can also use an Open Systems SnapVault system state data backup along with complete file system backups. These backups can be part of a disaster recovery plan. With Open Systems SnapVault for Windows Event Log, you can maintain the records of all the events that occur in the system. Such record keeping helps you troubleshoot problems or do capacity planning. You can use Open System SnapVault Configurator to enable or disable for Windows Event Log. Application, Security, and System event log backups are configurable options. Incremental updates of Windows Event Log files are not ed. Every update does a full backup of event logs.

16




Restoring Open Systems SnapVault  To restore a directory or a file, use these methods:  Use CIFS or NFS to copy the file or directory.  Use the snapvault restore command.  Use the OnCommand Unified Manager server.

 To restore an entire primary system, an operating system must reside on the disk.  To restore files to a primary system from tape, perform these steps: 1.Restore the data from the tape to the secondary system. 2.Restore from that secondary system to the Open Systems SnapVault primary system. 17


RESTORING OPEN SYSTEMS SNAPVAULT If data loss or corruption occurs on the Open Systems SnapVault primary system, the can restore a directory or a file, the entire system, or files from a tape device. Restoring a directory or a file When you use NFS or CIFS to copy files from the secondary system to the primary system, some Windows and UNIX attributes are not preserved, such as Windows sparse files, Windows EFS data, and UNIX access control lists. Use the snapvault restore command that is issued from the Open Systems SnapVault primary CLI: snapvault restore –s sv_snapshot –S sec_hostname:/vol/sec_vol/sec_qtree/file.doc prim_hostname:dirpath You can also use the OnCommand Unified Manager restore wizard or ed NDMP management software. Restoring an entire primary system To restore an entire primary system from a SnapVault secondary system, perform these steps: 1. 2. 3. 4.

Reinstall the operating system on the primary system. Reformat the file system just as the original file system was formatted. Install the Open Systems SnapVault agent. (Optional) If you backed up the primary system, restore the Windows system state data of the Open Systems SnapVault primary system. 5. Use the snapvault restore command to restore the backed-up directories.

17




Restoring to a primary system from tape To restore from tape to a primary system, perform these steps: 1. Mount the tape with the files that need to be restored on the secondary system. 2. Use the Data ONTAP restore command to restore from the tape to the SnapVault secondary system. 3. Use the snapvault restore command to restore the files from the secondary system to the NFS or CIFS primary system.

18




Restore Limitations  You cannot use the snapvault restore command to restore files to a directory that is part of a SnapVault relationship. To restore these files, do these steps: 1.Release the SnapVault backup relationship from the primary system before you perform the restore to the same location. 2.Restore the files to a new location.

 NTFS-compressed files and directories lose compressed attributes when the files and directories are restored.

18


RESTORE LIMITATIONS Restoring to a directory that is part of a SnapVault relationship If you attempt to use the snapvault restore command to restore files to a directory that is involved in a SnapVault relationship, the attempt fails, and you see this message: Invalid Qtree/Snapshot requested Directory in wrong phase Error performing restore. Check snapvault log. The following information is logged in the SnapVault log files: Invalid Qtree/Snapshot requested Directory in wrong phase To restore to such a directory, first release the SnapVault backup relationship. To release the relationship, use the snapvault release command that is issued from the primary system: snapvault release pathname sec_hostname:/vol/sec_vol/sec_qtree Alternatively, you can restore the files to a new location on the primary system. Restoring NTFS-compressed files and directories Open Systems SnapVault does not restore the compressed attribute on a Windows file if both of these conditions are true:  

19

The file has a compressed attribute set. The file has a sparse or compressed alternate data stream that is attached.




Resynchronization  To resynchronize a broken relationship, use the snapvault start –r command. This command does not resynchronization of restored subdirectories and single files.  To resynchronize after you use snapvault restore, perform these steps: 1.Select the Enable restart/resync on restore option on the primary system before you perform the restore. 2.Use the snapvault start –r command to resynchronize the relationship.

19


RESYNCHRONIZATION To be able to continue incremental backups, you must resynchronize the relationship when the Open Systems SnapVault primary and secondary systems become unsynchronized. An Open Systems SnapVault relationship can get out of sync in these cases:   

When an older version of the Open Systems SnapVault database is restored to the Open Systems SnapVault primary system When data is restored by using the snapvault restore command When the state of the destination qtree in an Open Systems SnapVault relationship is changed to readwritable (such an operation breaks the Open Systems SnapVault relationship)

Open Systems SnapVault resynchronization is ed from Data ONTAP 7.2 and Open Systems SnapVault 2.2 and later releases. Resynchronizing a broken relationship To resynchronize a broken Open Systems SnapVault relationship, on the SnapVault secondary system, enter this command: Secondary> snapvault start -r -S prim_hostname:dirpathsec_hostname:/vol/sec_vol/sec_qtree NOTE: Open Systems SnapVault does not resynchronizing restored subdirectories and single files.

20




Resynchronizing after a SnapVault restore To perform subsequent incremental backups from the restored location to the same qtree on the secondary storage system, resynchronize the relationship by performing these steps: 1. Before you restore the data, select the Enable restart/resync on restore check box in the Open Systems SnapVault Configurator. By default, this option is not selected. 2. If you selected this option before you performed the SnapVault restore, use the snapvault start – r command to resynchronize the SnapVault backup relationship after the restore is completed. 3. If you did not select this option before you performed the SnapVault restore, you must initialize the relationship after the restore. Issue the snapvault start command from the SnapVault secondary system.

21




Lesson 3

Open Systems SnapVault Deployment

20


LESSON 3: OPEN SYSTEMS SNAPVAULT DEPLOYMENT

22




Installing Open Systems SnapVault  Install the Open Systems SnapVault agent on Windows systems or UNIX-based systems.  An automatic postinstallation check occurs at the end of the installation: svinstallcheck.  Unattended installation requires minimal intervention:

 Is ed on all platforms on which Open Systems SnapVault is ed  Uses svconfigpackager to create the install image

 Open T ports 10000 and 10566 before installation.

21


INSTALLING OPEN SYSTEMS SNAPVAULT You can install the Open Systems SnapVault agent from the CD-ROM or from the Software page on the NetApp site. Refer to the Open Systems SnapVault Release Notes and the Installation and istration Guide for details on the installation process for Windows-based and UNIX-based systems. The Open Systems SnapVault agent installation stores these components on the primary system:    

Open Systems SnapVault database A set of executables Logs file Exclude Lists files

On a Windows primary, the default installation path is the C:\Program Files directory. On UNIX systems, the default installation location is the /usr directory. During the installation process, you are prompted to enter the NDMP to connect to the primary system. You are also prompted to enter the NDMP listening port and the host name or IP address of the secondary system. At the end of the installation routine, the svinstallcheck utility verifies successful installation and ensures that the services are running properly. Unattended installation  



23

Unattended installation enables you to install or upgrade Open Systems SnapVault software over a large number of Open Systems platforms with minimal intervention. To perform an unattended installation of Open Systems SnapVault, an installation script and other ing files are required. A utility called svconfigpackager is available in the Open Systems SnapVault software. When you run this utility on an Open Systems SnapVault primary system, the utility saves the current configuration settings to a file. Also, this utility can create an installation script that performs unattended installations or upgrades, in conjunction with the configuration settings file and other files. Ensure that the T port 10000 (for central management using NDMP) and T port 10566 (QSMSERVER) are open before Open Systems SnapVault is installed. For NetApp Host Agent, HTTP port 4092 and HTTPS port 4093 must be open as well. Appendix A: Open Systems SnapVault



Windows Server 2008 R2 Firewall Configuration

Windows Server 2008 R2 includes a built-in firewall that must be configured to allow SnapVault traffic. Create an inbound rule and add the appropriate ports.

22


WINDOWS SERVER 2008 R2: FIREWALL CONFIGURATION

24




Open Systems SnapVault Primary  Use the Open Systems SnapVault Configurator UI to and modify Open Systems SnapVault parameters.  Use the svsetstanza command as an alternative to the Open Systems SnapVault Configurator utility.

23


OPEN SYSTEMS SNAPVAULT PRIMARY Use the Open Systems SnapVault Configurator utility to and modify the Open Systems SnapVault configuration parameters. The Configurator utility is automatically installed during the Open Systems SnapVault agent installation. The utility can be invoked from the Windows > Start > Programs menu or from the CLI. The svconfigurator.exe program is located in the Install_dir/bin directory. From the Configurator, you can perform actions from these tabs:     

Machine tab: Displays information about the version of Open Systems SnapVault agent software and primary system machine information, such as the IP address and OS version Service tab: Enables you to stop and start the Open Systems SnapVault services General tab: Enables you to generate debugging files and to modify the default db, tmp, and trace directory locations Trace Level tab: Is used (with the General tab) to choose which process to take traces of and chooses the trace level SnapVault tab: Enables you to specify the SnapVault secondary hostname or IP address and to modify important parameters such as the BLI level, Open File Manager (OFM), Volume Shadow Copy Service (VSS), and NDMP parameters

Alternatively, you can use the svsetstanza utility from the primary system for configuration purposes instead of using the Configurator. The svsetstanza command is located in the install_dir/util directory.

25




Configuring SnapVault Secondary SnapVault secondary configuration steps are identical to the steps that are used in a SnapVault storage system deployment: 1. Install the required licenses (not on the primary system). 2. Enable the NDMP service.

3. Specify the Open Systems SnapVault primary systems to back up. 4. Initialize the baseline transfer. 5. Schedule updates.

24


CONFIGURING SNAPVAULT SECONDARY SnapVault secondary configuration steps: 1. Install the required licenses. sec> license add 2. Ensure that the NDMP service is enabled. sec> options ndmpd.enable on 3. Specify the names or IP addresses of the Open Systems SnapVault primary systems to back up and restore. sec> options snapvault.access host=prim_host1, prim_host2 4. For each Open Systems SnapVault platform directory or file that you want to back up to the SnapVault secondary, execute an initial baseline transfer. sec> snapvault start -S prim_hostname:dirpath sec_hostname:/vol/sec_vol/sec_qtree 5. Schedule incremental updates. For example: sec> snapvault snap sched -x vol1 sv_hourly 11@mon-fri@7-18 The -x parameter causes SnapVault to copy new or modified files from the Open Systems SnapVault system directory or file to their associated qtrees on the secondary. After all the secondary qtrees on the specified volume are updated, the SnapVault secondary creates a Snapshot copy of this volume for archiving.

26




Monitoring Transfers Use the snapvault status command to monitor the transfer progress, status, and lag. C:\> snapvault status -l c:\data Source: W2K3-Client:c:\data Destination: system:/vol/vol1/backup-windows Status: Transferring State: Source Lag: Mirror Timestamp: Base Snapshot: Current Transfer Type: Contents: Last Transfer Type: Last Transfer From: Bytes transferred so far: 18 MB Transfer Time Elapsed: 00:00:21 Total files to transfer: 12 Total files transferred: 7 Current File Size: 8323815 Current File Progress: 2097152 Current File Name: c:\data\file1 Transfer Error ID: Transfer Error Message: -

25


MONITORING TRANSFERS Use the snapvault status command to these things:   

Whether a transfer is in progress The state of the destination How long ago the last successful transfer took place

If the -l option is given, the output displays more detailed information. The information includes these things:    

The total files to transfer How many files were transferred Which file is currently transferring The transfer error ID, if any error occurred

For details on the snapvault status command output and options, refer to the Data ONTAP Data Protection Online Backup and Recovery Guide.

27




Open Systems SnapVault Commands The SnapVault commands that are available from the Open Systems SnapVault primary CLI are identical to the commands that are ed on a SnapVault storage primary system: C:\Program Files\netapp\snapvault\bin> snapvault The following commands are available; for more

information type "snapvault help " abort

destinations

help

release

restore

status

service

diag

NOTE: The service and diag commands are specific to Open Systems SnapVault primary systems.

26


OPEN SYSTEMS SNAPVAULT COMMANDS SnapVault commands are stored in the Install_dir/bin directory. Refer to the SnapVault module of this course for details of SnapVault command syntax:       

Use the status command to monitor the status of an Open Systems SnapVault backup or restore. Issue the destinations command to list all the known destinations for this Open Systems SnapVault primary system. The abort command enables you to stop a restore transfer from the secondary to the primary system. The release command enables you to release relationships that are stopped on the secondary system. If data loss or corruption occurs on the primary system, use the restore command to restore the affected file, directory, drive, or even the entire system. Use the service command to manage the Open Systems SnapVault service. Issue the diag command to display details of a given error number.

NOTE: The service and diag commands are specific to Open Systems SnapVault primary systems.

28




Log Files  Open Systems SnapVault primary log files:  Windows primary system: install_dir\netapp\snapvault\etc  UNIX-based primary system: /usr/snapvault

 A new file that is created daily at midnight:  Filename: snapvault.yyyymmdd  No extension for the current file

 SnapVault secondary log files:

root volume: /etc/log/snapmirror

27


LOG FILES You can find the operational status and problem reports of the Open Systems SnapVault primary system in the log file named snapvault. On the Open Systems SnapVault primary, these log files are found:  

On a Windows primary system, log files are in the C:\Program Files\netapp\snapvault\etc directory. On a UNIX-based primary system,log files are in the /usr/snapvault directory.

A new file is created daily at midnight or as soon after midnight as the first subsequent activity on the system takes place. The existing file is not archived until a new file is created. The current file has no extension. The archived files use the .yyyymmdd extension, where yyyy is the year, mm is the month, and dd is the date when the file was created. The log file is created during the initial baseline transfer, not by the install process. Therefore, if you search for the log file after you complete the installation, but before you back up data for the first time, you do not find the log file.

29




Lesson 4

Best Practices and Troubleshooting

28


LESSON 4: BEST PRACTICES AND TROUBLESHOOTING

30




Considerations  Open Systems SnapVault does not back up NFS mount points and CIFS shares.  Open Systems SnapVault built-in databases require free space. Use the Free Space Estimator utility.  OFM requires at least 15% disk space in the drive that is being backed up.

 Open Systems SnapVault s a maximum of 16 concurrent transfers.  The presence of a large number of small files can degrade performance.  The application database must be dismounted before Open Systems SnapVault backups are initiated. 29


CONSIDERATIONS Open Systems SnapVault does not back up remote NFS or CIFS file systems that are mounted on or mapped to UNIX or Windows primary systems. The Open Systems SnapVault database space requirements depend on several things:   

BLI level, the number of files Average file size Number of directories to back up

Use the Free Space Estimator utility to ensure that there is sufficient space on the Open Systems SnapVault install drive to perform an incremental update. If OFM (Windows 2000 Server) is used, the file systems that are being backed up must contain at least 15% of free disk space in the drive that is being backed up. If free space is not available, disable OFM for those drives. Open Systems SnapVault s a maximum of 16 simultaneous transfers from a primary system. Plan your backup schedules to ensure that 16 or fewer transfers occur at the same time. When you use Open Systems SnapVault, some overhead is always transferred for the files in the Open Systems SnapVault relationship that were modified. The Open Systems SnapVault primary system sends one 4-kilobyte (KB) header for every file and directory that exist in the relationship. In addition, for files and directories that are larger than 2 MB, an additional 4-KB header is transferred for every 2 MB. A large number of small files can degrade performance. A large number of files can result in a large amount of overhead data. If a large number of files is not likely to be modified, consider changing the BLI level to LOW. Open Systems SnapVault is not integrated with any database backup APIs. The database files need to be dismounted before you use Open Systems SnapVault to back up the files. To use the “hot backup mode” method, s need to script the procedure and test the script to ensure that the procedure works reliably in their environment. 31




Error Messages  Collect error messages that are logged on the Open Systems SnapVault primary and the SnapVault secondary: snapvault status –l /etc/log/snapmirror C:\Install_dir\snapvault\etc (Windows example)

 On the primary, use the snapvault diag <err_num> to display details for a given error number.  Error messages, causes, and actions are listed in the Open Systems SnapVault Installation and istration Guide.

30


ERROR MESSAGES To diagnose and troubleshoot Open Systems SnapVault issues, collect and analyze the error messages that are logged on these systems:  

Open Systems SnapVault primary (Install_dir/snapvault) SnapVault secondary (/etc/log/snapmirror) systems

On the Open Systems SnapVault primary CLI, use the snapvaultdiag <err_num> command to display details for a snapvault status –l given error code. The error code number is displayed in the command output. Example: $ snapvaultdiag 3016 Type: Error Message: A network error has occurred. Possible cause: The network socket was closed unexpectedly or the transfer was aborted by the . Possible action to take: network connectivity between the Open Systems SnapVault primary system and the secondary storage system. Error codes ranges:   

1000 to 1999: Information messages 2000 to 2999: Warning messages 3000 to 3999: Errors causing an abort of the transfer

Not all values in the range are used. A complete list of error codes, strings, causes, and actions is available in the Open Systems SnapVault Installation and istration Guide. 32




Data Collection  Open Systems SnapVault INFO tool:  Collects a complete set of information, command outputs, and (optionally) ChangeLog and trace files  Is packaged with the Open Systems SnapVault agent:  OSSVINFO.exe for W2K and W2K3  OSSVINFO.pl for Linux-based and UNIX-based systems

 Generate debug information:

Traces are compressed into a .bin file.

31


DATA COLLECTION When you the NetApp site, use either of two methods to provide information:  

Provide a set of files and command outputs. Run the Open Systems SnapVault INFO tool on the affected Open Systems SnapVault primary to collect a complete set of information from the primary and the secondary systems.

The Open Systems SnapVault INFO tool is packaged with Open Systems SnapVault, and two versions are available:  

Open Systems SnapVault INFO.exe for Windows 2000 Server and Windows Server 2003 Open Systems SnapVault primary systems Open Systems SnapVault INFO.pl for Linux-based and UNIX-based Open Systems SnapVault primary systems

The Open Systems SnapVault INFO tool writes the data to the output directory as a text file in a specific format. Also, the tool collects the ChangeLog and trace files to this output directory if either the -q (for Windows) or -all (for all platforms) option is given, as illustrated in this example: Open Systems SnapVaultINFO.exe [ -s secondary ] [ -l name:] [-q qtreeid][-all] Output_Dir The ChangeLog file captures duplicate inode issues and metadata corruption at run time. The ChangeLog capture is enabled by default in the snapvault.cfg file. NetApp technical might ask you to enable Open Systems SnapVault debugging to troubleshoot the issue. The traces that are collected are automatically compressed into a .bin extension file. When you generate debug files, complete these actions:   33

Disable the generation of these files after you send a batch to technical . To minimize the impact on performance, delete the debug files from the system after you send the files to technical . Appendix A: Open Systems SnapVault



Baseline Transfer Assistance Open Systems SnapVault

 Using a portable device, the logical replication utility (LREP) initializes SnapVault or Open Systems SnapVault data transfers over low-bandwidth connections.  Two utilities are used:  lrep_reader  lrep_writer

 LREP s Open Systems SnapVault data restore, data compression, and encryption.  LREP 2.0 is packaged with Open Systems SnapVault 2.5 and later releases and is able from the NetApp site. 32


BASELINE TRANSFER ASSISTANCE: OPEN SYSTEMS SNAPVAULT The LREP tool uses a portable device to initialize Open Systems SnapVault data transfers without traversing the network:    

You ship the portable device to the location of the SnapVault secondary system. Next, you move the data from the portable device to the secondary system. After the relationship between the primary and secondary systems is established and modified, incremental transfers can occur directly. You can also use LREP to restore data from the secondary storage system to the primary system.

The LREP tool consists of two utilities:  

lrep_reader: used at the remote office to write data from the Open Systems SnapVault primary system to the portable device lrep_writer: used at the location of the SnapVault secondary storage system to write data from the portable device to the secondary storage system

Using a zlib library, LREP enables the compression of data. Compression of LREP data is done in the memory, before the data is written to the disk. The Advanced Encryption Standard algorithm is used to encrypt and decrypt the LREP data. LREP 2.0 is packaged with Open Systems SnapVault 2.5 and later. You can also the LREP tool from the NetApp site. You run the tool on Windows 2000/2003, Linux-based, and UNIX-based platforms. For details on how to use LREP to back up and restore Open Systems SnapVault data, refer to the Logical Replication (LREP) Tool 2.0 Guide on the NetApp site.

34




Baseline Transfer Assistance SnapVault

 Open Systems SnapVault and SnapVault primaries are ed.  LREP can be ed from the NetApp site. Data Center

Remote Office

3

4

1 lrep_reader

5

2

lrep_writer

6 Low-Bandwidth FAS1

Connection

qtree1 qtree1

qtree1 qtree1

33

FAS2

SnapVault


BASELINE TRANSFER ASSISTANCE: SNAPVAULT You can use the LREP tool with a portable device, such as tape or an external drive, to perform SnapVault initial transfers without traversing the network. This process is often referred to as seeding the baseline transfer. LREP applies only to SnapVault software. You can the LREP tool set from the NetApp site. In the illustration, you see the configuration of a SnapVault baseline transfer from FAS1 (the source system at the data center) to FAS2 (the destination system at the remote office). This configuration uses the LREP utilities. 1. Install the lrep_reader utility on Client1 at the data center. 2. Run the lrep_reader utility from Client1 to snapmirror initialize from the source to the portable media. 3. The portable media is then transferred to the remote office. 4. The lrep_writer utility is installed on Client2 at the remote office. 5. Run the lrep_writer utility from Client2 to reconstruct the replication streams by reading from files that were created by the lrep_reader utility. 6. From the destination, run the snapmirror initialize command to read the logical replication stream data from the lrep_writer utility. The SnapVault initial transfer completes. Edit the snapmirror.conf file to reflect the correct source storage system. The LREP tool is ed on the following operating systems:      35

Windows 2000, Windows Server 2003, and Windows Storage Server 2003 on x86 and x86-64/EM64T platforms Red Hat Enterprise Linux 4.0 for x86 and x86-64/EM64T Red Hat Enterprise Linux 5.0 for x86 and x86-64/EM64T SUSE Linux Enterprise Server 9 for x86 and x86-64/EM64T SUSE Linux Enterprise Server 10 for x86 and x86-64/EM64T Appendix A: Open Systems SnapVault



  

36

Solaris 9 and 10 on UltraSPARC systems AIX 5L 5.1, 5.2, and 5.3 on IBM PowerPC and IBM POWER processor-based systems HP-UX 11.23 and HP-UX 11.31 on PA-RISC based systems




References  Open Systems SnapVault Installation and istration Guide  Open Systems SnapVault Release Notes  Enabling Rapid Recovery with SnapVault http://media.netapp.com/documents/tr-3252.pdf  Open Systems SnapVault Best Practices Guide http://media.netapp.com/documents/tr-3466.pdf

34


REFERENCES For more information, see the NetApp site.

37




Module Summary Now that you have completed this module, you should be able to:  Describe how Open Systems SnapVault integrates with Data ONTAP SnapVault  Describe Open Systems SnapVault advanced features  Configure and ister Open Systems SnapVault  Perform Open Systems SnapVault backup and restore operations  Troubleshoot and resolve Open Systems SnapVault transfer failures 35


MODULE SUMMARY

38




36


THANK YOU

39




Cluster Data Ontap 8.3 Data Protection Student Guide 4s6m69

Overview 3e4r5l

More details w3441

Related Documents 3m3m1z

Cluster Data Ontap 8.3 Data Protection Student Guide 4s6m69

Data Domain Student Guide 35w4p

Data Domain Student Guide 35w4p

Vmware Data Protection istration Guide 61 6b1b4t

Safend Data Protection Suite 3.4.5 Installation Guide 2q4d4p

Data Ontap Dsm 3.2 For Windows Mpio 23z10