Oracle ZFS Storage Appliance: Replication issue with VMWARE SRM (Site Recovery Manager) and Oracle SRA (Storage Replication Adapter)

Asset ID:	1-72-2095611.1
Update Date:	2018-01-08
Keywords:

Solution Type Problem Resolution Sure

Solution 2095611.1 : Oracle ZFS Storage Appliance: Replication issue with VMWARE SRM (Site Recovery Manager) and Oracle SRA (Storage Replication Adapter)

Applies to:

Oracle ZFS Storage ZS3-2 - Version All Versions and later
Sun ZFS Storage 7420 - Version All Versions and later
Sun Storage 7110 Unified Storage System - Version All Versions and later
Sun Storage 7410 Unified Storage System - Version All Versions and later
Sun Storage 7310 Unified Storage System - Version All Versions and later
7000 Appliance OS (Fishworks)
Environment:
ZFSSA Clustered
VMWARE SRM (Site Recovery Manager) 5.5 NFS
Oracle SRA (Storage Replication Adapter) 5.0.0.13

Symptoms

Replication within VMWARE SRM (Site Recovery Manager) and Oracle SRA (Storage Replication Adapter) failing.

The SRA Errors "Invalid Peer Array ID".

An error in SRM "Fail to sync data on replica devices".

SRA command 'syncOnce' Failed with Error message "Invalid Peer Array ID Check if the Peer ArrayId is valid".

Changes

Resources moved within the cluster peer node. Takeover or Failback initiated in the replication target cluster.

Cause

Cause 1:

Problem:Invalid Peer Array ID
Error code: 1016

This error occurs when Oracle SRA is not able to match the peer array ID in the command to the array ID returned by the appliance in the discovery phase.

Check if the Peer Array Id is valid

Cause 2:

It is a limitation and/or verification check by Design.

Oracle SRA assumes by default that the cluster heads are in a healthy cluster state.

If any head fails, and/or the resources are taken over by the other head this 'errors out and replication refused.

Solution

Oracle SRA assumes by default that the cluster heads are in a healthy cluster state.

If any head fails, and/or the resources are taken over by the other head (peer), the SRM administrator must manually intervene and run the batch file 'RunMe.bat' provided with the downloaded package (plugins) from the Oracle ZFS Storage Appliance Plug-in Downloads page on the Oracle Technical Network website.

The script will take cluster peer node names as inputs

For example:

If peer a-h1 is taken over by a-h2, provide a-h1 and a-h2 as inputs. The script should be run on both sites.

After the failed head comes back up, the administrator should run the batch file again to remove the heads from the list of clustered appliances.

Note that only nodenames should be used while running the script.

Do not enter a-h1.domainName.

References

https://docs.oracle.com/cd/E51475_01/html/E52872/integration__vmware_site_recovery_manager__sun_zfs_storage_7000_storage_replic.html
http://www.oracle.com/technetwork/server-storage/sun-unified-storage/downloads/zfssa-plugins-1489830.html

Attachments

This solution has no attachment