Error: Snapshot volume does not exist

[Mircea Giurgeu_1]

I have two MSA2050 connected through FC switches. I'm the following error when trying to run a replication task:

Scheduler: The scheduler was unable to complete the replication. - Snapshot volume does not exist. (task: Repl_VOL1, task type: Replicate)

EVENT ID:#A1774

EVENT CODE:362

Both units have the same firmware, I have enough space on the pools, I can successfuly run the replication manually. I can provide any information needed to troublesthoot this issues. Any help appreciated.

[sbhat09]

Hello @Mircea Giurgeu_1 

The error is due to the replication scheduler failed. Please follow the below checkpoints and update me:

Is the connection between the source and target systems working fine?
When was the last time the replication was running successfully?
Are there are any other volumes between the same source and target running successfully?
Do you correlate this issue with any other changes (DST change, networking, firewall, upgrade etc) with your systems?
Can you add a test volume to replication from the same source to target and check how it goes?
Have you tried to re-run the replication task? what is the outcome?

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.

[Mircea Giurgeu_1]

Thank you for your suggestions. Here are my responses to your questions, in order:

The connection between source and target systems seems to be working fine. During initial replication I got one or two notifications that the process has been suspended but then it restarted on its own and completed successfuly.

The intial and subesquent replications are always successful if I run them mannually but have never worked for this particular volume when I had it configured to run under the scheduler.

No, I did not have any other replicated volumes between these two systems

I cannot think of any correlation or system changes related to this failed replication. The systems are connected through FC switches, and like I said the replication is always successful when I initiate it manually, both intial and subsequent ones, but the subsequent replications are always failing with the "Snapshot volume does not exist"

I have added a small (100GB) test volume and I am able to replicate it successfuly to the other system both manually and using the scheduler. The only thing different with this replication set is the names I have used for the local and remote volumes and the size of the volume.

I have removed and re-created the replication set for the production volume that was failing and restarted the replication but it will take a while for the initial stage to complete. I have tried this procedure before posting in this forum and it didn't work. If it does work this time it must be because of the different names that I had used.

I will keep you posted with the result as soon as the initial replication completes and cand run the first scheduled one.

[Mircea Giurgeu_1]

Hi Srinivas,

I just posted a reply to your suggestions below.

[ArunKKR]

Hi,

Recreating the replication set and deleting old snapshots should help.

Could you try the below command and check whether there are any snapshots with status "unavailable"? 

show snapshots type all

Note: Do not delete any internal snapshots ending with the extension .S1, .S2 , .S3

[Mircea Giurgeu_1]

I am happy to report that after recreating the replication set and replication schedule, now the replication works as it should. The only thing different is the name of the replicaiton set. 

Thank you very much for your suggestions