I have a question regarding multiple drives and bitmap sync, here is the use case:
Assuming a file server with 2 disks replicated on-prem to another on-prem environment over a limited bandwidth link.
Users regularly writes small files on both disks but it fits on the existing link and replication is OK during regular activity period.
Now let’s say one user writes a huge file over one of the disks (disk1) while other users keep writing regular small files on disk2.
As my bandwidth is limited, the VPG will enter in bitmap sync state for disk 1.
Now let’s say I need to trigger my failover while disk 1 is still bitmap syncing but disk 2 is properly replicating.
Is there a way to failover to a checkpoint during the bitmap sync (so I get my latest datas on disk 2) or it must be necessarily before the bitmap sync (hence if one drive bitmap sync, the whole VM is impacted)?
Or failover one drive to a specific checkpoint and the other one to another (I don’t think so but you never no 🙂 )?
In order to run recovery operations, the VPG must not be performing other tasks such as a bitmap sync. Please let me know if you have any additional questions.
Thanks for your answer Joe, but I’m a bit confused with it as I can see the opposite in the documentation:
“If a disaster occurs requiring a failover during a bitmap synchronization, the VPG status changes to Recovery Possible and you can recover to the last checkpoint written to the journal.”
It would not really make sense from my perspective if you could not trigger a failover during a network interruption I believe…
When a sync operation is in progress your VPG will not be available for failover. In your statement above, the VPG comes out of bitmap sync and goes into Recovery possible during a disaster, therefore the VPG will be available for failover. Please let me know if you have any additional questions.