When a running a `pbm restore`, it is possible that the restore fails for several reasons. During our testing we have experienced both out-of-space and out-of-memory events.
The problem is that there is no way to automatically determine if the restore has completed successfully, or failed for some reason. We are polling the output of `pbm status --sections=running --out=json`, and the failed job just disappears with no way to determine there was an error.
In the case of the out-of-space problem, there was at least an error recorded in the logs, but as I can't depend on an error making it to the database (if pbm-agent crashes, for example), at the moment it appears I have to poll `pbm logs` for a success message.
Environment
None
Smart Checklist
Activity
Aaditya Dubey December 10, 2023 at 8:11 AM
Hi ,
Closing the report, no activity for a long!
Aaditya Dubey July 19, 2022 at 11:43 AM
Hi ,
Thank you for the update. Will wait for your confirmation.
Daniel Oliver July 19, 2022 at 11:41 AM
Hi Aaditya Dubey,
I'll be working on an update backup and restore procedures of the next few weeks, so I will re-test with the latest version then.
Aaditya Dubey July 12, 2022 at 9:27 AM
Hi ,
Thank you for the report. please let me know if issue still persists.
When a running a `pbm restore`, it is possible that the restore fails for several reasons. During our testing we have experienced both out-of-space and out-of-memory events.
The problem is that there is no way to automatically determine if the restore has completed successfully, or failed for some reason. We are polling the output of `pbm status --sections=running --out=json`, and the failed job just disappears with no way to determine there was an error.
In the case of the out-of-space problem, there was at least an error recorded in the logs, but as I can't depend on an error making it to the database (if pbm-agent crashes, for example), at the moment it appears I have to poll `pbm logs` for a success message.