Unsolved
This post is more than 5 years old
1 Rookie
•
28 Posts
0
1796
May 10th, 2017 13:00
ISILON SYNCIQ PERMANENT FAILOVER
Hi Folks,
We are doing the fail over from one site to other site, it is going to be a Permanent Fail over using synciq.
We know the fail over process (just Write enable the Target Copy ), but my query is Permanent Failover.
For Permanent Failover do we need to wait untill the Sync is good between two sites and Power down the server and Delete the SYNCIQ Policy and map the Target Drive to the Production Server that has been Migrated to Secondary Site and Create a New SYCIQ Policy, IS THIS PROCESS CORRECT????????
The issue with this is If we want to Rollback, do we have a Consistent Copy there ?
Kindly tell me the correct process for Permanent Failover of the SYNCIQ??????????
Regards,
JAY
chughh14
1 Message
1
May 12th, 2017 03:00
Hello Jaii,
You do not need to shutdown primary cluster. Stop all SMB NFS HTTP & FTP services. SyncIQ should complete from source to target.
Allow writes on target cluster change dns-recors, clients will connect to target cluster.
For rollback there is a standard procedure for consistent copy.
Refer to guide for more clarification.
https://www.emc.com/collateral/hardware/white-papers/h8224-replication-isilon-synciq-wp.pdf
Brian_Coulombe_
1 Rookie
•
107 Posts
1
May 12th, 2017 04:00
Jay,
1. Yes, you need to wait until the data is 100% synced. If you are still writing to the source, you'll probably never have a 100% sync. At some point you'll have to stop writes to the source, let SyncIQ finish, break sync, then make the target r/w.
2. If you delete the SyncIQ policy and data, to fail back you'll have to perform another complete Sync from source to target (for consistency). You will not have a consistent copy if new data is being written to the target (now source) so I'd just run a SyncIQ job and wait for it to catch up, stop writes to the source, let SyncIQ finish, then break sync again.
Be careful how you do this....