Data Backup and Recovery
Data Backup and Recovery
I have three 8.3.2 cdot clusters in three locations, lets call them A, B, and C. A replicates content to both B and C. I need to set up two jobs that replicate content from A to B and then from B to C.
I've set up the cluster and SVM peering, but when I try to create the snapmirror reliationship from B to C in System Manager it doesn't list the volumes I need to replicate as choices. it indicates that the volumes must be read-write, which of course they aren't since they are snapmirror destination volumes in an A to B relationship.
I've not set up a cascade relationship before and don't see much documentation on how to actually do it. I assumed I could just set up a second relationship. Any insights on how to accomplish this?
Solved! See The Solution
Thomas -
Just do the 'snapmirror create' command for vol B > C.
It works - no special flags needed.
I just tried it on my simulator:
Progress
Source Destination Mirror Relationship Total Last
Path Type Path State Status Progress Healthy Updated
----------- ---- ------------ ------- -------------- --------- ------- --------
vs1:vola DP vs1:vs1_vola_mirror
Snapmirrored
Idle - true -
vs1:vs1_vola_mirror
DP vs1:volc Snapmirrored
Idle - true -
2 entries were displayed.
I hope this response has been helpful to you.
At your service,
Eugene E. Kashpureff, Sr.
Independent NetApp Consultant http://www.linkedin.com/in/eugenekashpureff
Senior NetApp Instructor, FastLane US http://www.fastlaneus.com/
(P.S. I appreciate 'kudos' on any helpful posts.)
Thomas -
Miror-Mirror cascade is supported in 8.3.2, but I don't think it's in System Manager.
You'd want to go to the command line.
You may want to review the Data Protection Guide:
I hope this response has been helpful to you.
At your service,
Eugene E. Kashpureff, Sr.
Independent NetApp Consultant http://www.linkedin.com/in/eugenekashpureff
Senior NetApp Instructor, FastLane US http://www.fastlaneus.com/
(P.S. I appreciate 'kudos' on any helpful posts.)
Thanks ekashpureff. I've read the section you referenced in the Data Protection Guide but it gives no specfic steps.
Anyone else?
To comment further, the man page for snapmirror create says:
"The source volume should be in the online state and a read-write (RW) type."
Of course this isn't true for snapmirror destinations.
As already noted, the "how a mirror-mirror cascade works" section in documentation describes how it's supposed to work but doesn't describe how to set it up.
Thomas -
Just do the 'snapmirror create' command for vol B > C.
It works - no special flags needed.
I just tried it on my simulator:
Progress
Source Destination Mirror Relationship Total Last
Path Type Path State Status Progress Healthy Updated
----------- ---- ------------ ------- -------------- --------- ------- --------
vs1:vola DP vs1:vs1_vola_mirror
Snapmirrored
Idle - true -
vs1:vs1_vola_mirror
DP vs1:volc Snapmirrored
Idle - true -
2 entries were displayed.
I hope this response has been helpful to you.
At your service,
Eugene E. Kashpureff, Sr.
Independent NetApp Consultant http://www.linkedin.com/in/eugenekashpureff
Senior NetApp Instructor, FastLane US http://www.fastlaneus.com/
(P.S. I appreciate 'kudos' on any helpful posts.)
Thank you! That worked. Looks like documentation was incorrect.
Hi
Greetings to you.
I have established cascaded replicaiton (A > B > C ).
Now we have requirement to do reverse resync from cascaded mirror leg to primary source (C > A).
I have planned to create peer relation between cascaded SVM to Primary source SVM. Later Break the replication between Primary replication ( A > B) and proced with reverse resync ( C > A ).
Due to business requirement, I cannot do reverse resync from primary destination leg (B > A).
Can you please help me and share your inputs, if any additional pre-checkes need to be verified.
Thanks
Hi Sir,
Cascaded replication i have successfully configured. Below are the learning points
1. Cascaded DP volumes should be created with same attributes as source volume. Else, svm language settings will be
picked during volume creation.
2. Need to calculate the sync time being taken for incremental update of source replication pair. Based on that, we have to
add cascaded replication schedule.
3. If we have to re-establish the replication back to A from C, then we have to create adhoc snapshot and after it is
appeared on 3 Replication LEGs, we have to terminate the replication. Later we can re-establish the replication back to
A.
Thank you