INVALID Solaris Backup Jobs - do not work/start

[uk1984]

We have an urgent issue with the Veeam Agent for Solaris on 1 very important machine.

We have several different job types for the same machine (backups different folders).

Backup is created off of a Protection group and my solaris admin has installed the package
and also imported the xml file on the machine.
The Solaris server is also registered within VBR server correctly.

We have these 2 Prio 1 cases and need urgent help here!
Case #07510829 — 2 of 3 jobs show status "invalid" and do not work
Case #07514540 — Backupjob running since over 18 hours and at 99% for hours

-----

Job1 - Custom scope "/" (but with folder exlusions)
-> works!

Job2 Custom scope "one folder"(no exclusions)
/zones/pkvtest/root/usr/local/application/.zfs/snapshot/veeam/
with pre- and post-script
-> shows invalid on solaris server cli output and does not work/start!
And if I make changes at the job at some point they are not shown on the solaris side...
Syncnow does not "catch" them anymore..

Job3 Custom scope "one folder" (no exclusions)
/zones/pkvprod/root/usr/local/application/.zfs/snapshot/veeam/
with pre- and post-script
-> shows invalid on solaris server cli output and does not work/start!
And if I make changes at the job at some point they are not shown on the solaris side...
Syncnow does not "catch" them anymore..

Job4 - Entire machine "works" - at least starts...
- Is kind of stuck at 99% and we do not know if it is technically still doing something...

-----
I created Job2 and Job3 like I did with the other 2 jobs which show "ALL system".
Tried also cloning the 2 working jobs and modifying but at some point the are just "invalid"...

Please help!

[PTide]

Hi,

I suspect that Jobs #2 and #3 might be having issues because of the hidden ZFS snapshots.
Have you tried making the directory visible (as describied in the KB article)?

As for Job #4 - what's the overall amount of data that is being backed up?

Thanks!

[VAB]

Note that the commands referenced in the linked KB article are incorrect. They are not applicable to Solaris, but work only on the ZFS storage appliance.

The correct solution on Solaris is to set the dataset property for the affected snapshots. It is called "snapdir" and takes the values "hidden" and "visible", with "hidden" being the default.

[rovshan.pashayev]

Hello,

Thank you for bringing the incorrect link to our attention. We will fix the KB.