SOBR tiering job failure

[selva]

Case # 0469396

I had to migrate the B&R server and had to use an old Windows 2008 R2 machine -- was earlier on a Windows 10 VM. The migration worked fine and jobs targetting local repos are completing without errors.

But SOBR tiering job (to backblaze) tries to do repository cleanup and ends in an error after trying for exactly 1 hour. The error is:

17/06/2021 8:34:33 PM :: Object storage cleanup failed: HTTP exception: WinHttpSendRequest: 12175: A security error occurred
, error code: 12175

As per MS, this has something to do with the certificate returned by the server, but that doesn't make a lot of sense as B&R tries the cleanup for 1 hour before erroring out.

At this point there are no new restore points to upload to capacity tier, so not sure whether that would error too, or whether this cleanup error will block any uploads. This repeats every 4 hours, so its not a one time error.

While migrating I also upgraded to v11 --- in retrospect I should have done the upgrade at a later time but not clear whether that has anything to do with this error. More likely something about Server 2008 R2 which is fairly old? Anyone seen similar errors?

[Gostev]

Right, from what I remember Server 2008 R2 does not support some modern security standards, at least out of the box.

[selva]

This never got resolved (looks like support guys abandoned me as no response in 5 days, I cant just keep knocking at their doors).

But I managed to find another Windows 10 host to move the backup server to. And, all is well again. I wont push this any further as, IMO, it may be better to just remove 2008 R2 SP1 from the list of supported platforms for B&R.

[Gostev]

This will be done in V12 but for V11 it remains fully supported and still used by a few percent of our customers (which is A LOT in absolute numbers).

[selva]

In v12 I thought we are getting support for B&R on Linux -- come on Veeam, you can do it

[Gostev]

Which Linux? One does not just make binaries for Linux

[selva]

Pick a couple of (or even one) distribution -- I would be happy with any.

Wishful thinking apart, I have run into a problem after moving B&R back to Win10. When I moved to 2008 R2, SOBR cleanup "error" caused last 5 days of backup not copied to capacity tier. Now after moving back to Win10, tiering job completes without error, but its not copying those 5 daily backups or new ones being created. I always had and still have "copy backups to object storage as soon as they are created" turned on.

Its like "copy backup to capacity tier" got disrupted 5 days ago and no attempt is being done to catch up. My operational restore window is long, so I can't wait for the "move" operation to kick in. Manually triggered or automatic tiering jobs complete quickly with "0" backups to copy to capacity tier --- while it should be copying 5 or 6 restore points.

Is there something I can do to force it to copy those?

Earnestly hoping I can somehow nudge it to do the right thing instead of having to go back to support.

[selva]

Solved. The capacity tier had gone into sealed state. Not sure whether that happened when I migrated to a new server and back, if so how. I don't recall sealing the extent manually. Anyway, good news is that the pending backups are all getting copied now.

[veremin]

It should not have gone into maintenance mode automatically, so it must have been manual or scripted run. If you do not believe so or the situation repeats, kindly open a new ticket with our support team. Details of this event are likely still in the logs. Thanks!