Remote copy groups// capacity issue on the secundary Storage on VMware enviroment

[francisco82]

hello community, we have 2 Storage Primera for more than 2 years with 3 remote copy groups for a VMware environment with Peer persistence, since 2 months we are facing a capacity problem in secondary storage:

Storage01 is 89 porcentage full but the secondary is 97 porcentage full and in theory we should have more or less the same capacity on both Storages

 the  firmware version that we have in these Primera is 4.4.1

i opened a case with HPE for this issue and the solution was.. make a switchover of the all remote copy groups to be able to reclaim the space in the second Storage, i did that   and now al is good (same space in both Storages).... but my question is:

What is the difference between these last 2 months and the last year working with the same configuration on this storage?

Why do we have this problem now? and not before?

Is it possible to run some command on secondary storage to avoid periodically switchover remote copy groups to reclaim space on secondary storage?

Thanks in Advance.

[support_s]

System recommended content:

1. HPE OmniStack 4.1.3 for vSphere Release Notes

Please click on "Thumbs Up/Kudo" icon to give a "Kudo".

Thank you for being a HPE valuable community member.

[francisco82]

@support_sthanks for your answer but... i don't understand the relationship of this document and the issues that i am facing, could give me something more of information?

[sbhat09]

Hello @francisco82,

Was that the difference in virtual size only or physical (actual space) as well?
If it was actual space, the difference that occurred (between the primary and secondary) is an exception.
The reasons we can suspect are due to the difference in the deduplication/compression ratio, excessive snapshots stalled up, or housekeeping not running as expected on secondary etc. But anyway it is just guesswork.
To find the actual reason, it requires historical log analysis and a detailed 1-1 volume comparison.
in this case, it is difficult find the root cause here. Probably you can ask HPE support to explain that.
I don't think it will recur frequently. The simpler way to keep a watch on this is to keep monitoring the space utilization on primary and secondary once in a week or fortnight. When you get to know the exact reason causing the difference, we can look for the fix.

Regards,
Srinivas Bhat

If you feel this was helpful please click the KUDOS! thumb below!
Note: All of my comments are my own and are not any official representation of HPE.

[lober]

hi @sbhat09  thanks for checking my message, i ran shovv -s command on both storages and i have very similar values on all VVs but HPE says it's a space reclamation problem, but i don't understand why we have this problem now and not before, and the The solution was made the switchover of the remote copy groups to reclaim space on secondary storage.

Have you heard of any similar case?

[sbhat09]

Hello @lober, 

I had seen a similar issue around 4-5 years ago. We had done snapshot cleanup and compactcpg. Not sure if switchover was also done.

I don't think you will have to keep doing switchovers on regular basis. It may be a one-time issue. In case it recures, run compactcpg -dr <cpg_name> to understand if there is any space that can be compactable. Later you can run the command without 'dr' at your convenience for the actual compact.

If you can share the HPE case number with me in private message, I will try to get more details and get back to you if there are any measures you can take in case the issue recures.

Regards,
Srinivas Bhat

If you feel this was helpful please click the KUDOS! thumb below!
Note: All of my comments are my own and are not any official representation of HPE.

[sbhat09]

Hello @lober,

I checked further details about this. This issue occurues due to a bug.
It happens sometimes due to remote volume fails temporarily and the replication initiates a new full copy on the secondary along with a recovery snapshot. So, a lot of space will be consumed on the secondary by the un-necessary copies and snapshots.

When the issue occurs, switchover is not the only workaround. You can remove all the unwanted snapshots and copies on the secondary and let the remote copy continue. This may not happen very frequently though.

The issue is identified as a bug and fixed in the Primera version 4.5 or later. So, please consider to upgrade the OS for permanent  fix.

Regards,
Srinivas Bhat

If you feel this was helpful please click the KUDOS! thumb below!
Note: All of my comments are my own and are not any official representation of HPE.

[lober]

hi @sbhat09  top thanks for your comments, we are working in upgrade our Storages Primera thanks

[sbhat09]

Great.! Hope the issue will be fixed permanently.

Regards,
Srinivas Bhat

If you feel this was helpful please click the KUDOS! thumb below!
Note: All of my comments are my own and are not any official representation of HPE.