Posgresql - The SQL Server machine hosting the configuration database is currently unavailable

[redhorse]

Hello,

from time to time a job failed because of this message:

„ Posgresql - The SQL Server machine hosting the configuration database is currently unavailable“

I usually see it when a health check is running and then the sql transaction log backup fails. The backup server has 128 GB RAM and 2 CPUs, I don‘t see a performance issue. What could cause this problem?

[PetrM]

Hello,

Please provide a support case ID as requested when you create a new topic about technical issue. As explained there, we cannot troubleshoot technical issues over the forum posts. All posts about technical issues without case ID will be deleted by forum moderator eventually. You should also upload debug logs for our support as per the instruction of this KB before they will be able to assist.

Thanks!

[redhorse]

Yes, I will do this. We've modified the AV settings and I hope it helps, otherweise I'll crate a case.

[karsten123]

did you install postgre standalone or was it installed with the VBR setup?

[redhorse]

It's installed with the VBR setup

[FrancWest]

We had the same issue, support suggested modifying a specific setting in PostgreSQL, after that the PostgreSQL was stable and we didn't had the issue again. In our case it had nothing to do with AV. The issue happened when backups were running with synthetic fulls and occured during the weekend on Saturdays. For your reference, our case number was #07186664.

[redhorse]

Thank you, may be I have to come back to this. Right now it seems to work with the new AV settings, but I think I have to observe this a few days.

[redhorse]

Good morning,

so we had this issue again and I can confirm that this happens during the health check job. No performance issue with the server and no parallel jobs, so I had to stop the Veeam jobs.

Support case: #07243054

#MOD: replaced contract number #03 with case number #07

[FrancWest]

Let support try changing the PostgreSQL setting as they did in our case.

[redhorse]

Yes, I'll attach your case number to the case as reference. Thanks!

PS: Sorry, I've copied the wrong number for the case, thanks for correction.

[redhorse]

The support advised us to set the work_mem parameter in the VeeamBackup Database to 131072kb and that solved the issue.

[FrancWest]

Same as in our case. Glad it worked.