Failure with Endpoint Backup after Upgrade B&R to v9

[ds2]

Hello,

after Upgrading B&R to V9 all my Endpoint Backups fails with Error "Full backup file merge failed Error: Not transformed OIB was not found. CreationTime: "XXXXXXX".

Endpoints are all at Version 1.1.2.119.

Greetnigs ds2

Case ID #01419013

[Dima P.]

Hi ds2,

Have you reconfigured the target repository by any change? I've forwarded your case directly to QA team, so stay tuned.

[ds2]

Hello Dima,

no there are no changes for respositorys.

[ds2]

Here are some more Informations:

- Creating new Restorepoint is okay
- Merging the oldest runs up to 99% and then comes the failure
- The oldest Restorepoint is deleted if i take a look in B&R Console (Porperties)
- BackupCopy-Job for Endpoints is okay
- A second run of Endpoint Job (without merge) is okay

For me it looks like the oldest Restorepoint ist deleted after merging and there is a second try to delete the file wich then doesn't exist.

[LorettoCNY]

I am also having the same problem with all of my endpoint jobs to V9. Everything were perfectly in V8.
Error: Full Backup file marge failed Error: Not transformed OIB was not found. Creation Time: XXXX

I'm targeting an iSCSI repository that is attached directly to the Veeam B&R server as a local disk to the server.

[Dima P.]

ds2,
Thanks for the heads up - we are looking into this.

LorettoCNY,
Please open a support case, so we could get access to the logs. Thanks.

[ds2]

My repository for the EP's is an USB-3.0-Raid-Disk attached local to the B&R Server (physical W2K12 R2).

[Dima P.]

Thanks – we are investigating. For now, it looks like the first job run after upgrade is affected and then it should work properly.

Can you clarify, does the job retry kick in after the first job fail?

[ds2]

Hello Dima,

the retry is the the normal retry after 10 minutes. Two of the Enponits has no retry forced after the failure (don't ask me why).

If it is realy only the first run after upgrade, then we could see it later this day, because the first endpointjob after the upgrade runs today at 17:50 for one of the endpoints for the second time.

If it works proper at the second run, what about the backupchain then? Is it valid?

Greetings

P.S.: For the support-case i should upload the logs, but actually I'm out of office, so I could only upload the logs later this day.

[Dima P.]

That’s ok - share the logs at your own convenience. Yes, next job run should be valid.

[ds2]

That’s ok - share the logs at your own convenience. Yes, next job run should be valid.

And what about the complete chain?

[Dima P.]

Same - since there is a built in health check. Otherwise, you'd get a new full backup and new backup chain

[Dima P.]

Regarding the missing retries – was this first failed job run a manual one, scheduled or triggered by backup event. Thanks

[ds2]

Hello Dima,

some of the jobs are triggered by VEB it self (Servers at defined time) and some are triggered by tasksheduler (Endpoints at defined time and run task as soon as possible after a scheduled start is missed if the Endpoint isn't on at the defined time). One Endpoint I've tested by a manual run.

All have the same issue. I don't know why 2 (the manual one and one triggered by tasksheduler) of the 6 jobs don't force a retry.

greetings

[Dima P.]

I got it. Script trigger acts the same way as a manual backup start (to avoid the interference with the regular settings, retires and backup no more often safety trigger switch), so it does not have a retry cycle.

[ds2]

One quick question before I upload the logs later. Through the support was requested the logs from B&R and VEB. Are the VEB-Logs from client enough?

[Dima P.]

Since the issue is related to VBR and VEB please upload both.

[ds2]

My question is - B&R and one client VEB or B&R and all client VEB logs. It isn't realy easy to access to all clients an collect the logs from VEB.

[Dima P.]

One affected VEB and target VBR should be enough - no need to gather all the VEB's logs.

[ds2]

Hello,

the logs are uploaded now.

It seems realy that just the first run is affected.

Right now there was a Jobrun tiggered by VEB an one Jobrun tiggered by tasksheduler on an other client. Both merged the oldest Restorepoint without any error or warning!

Please take a look to the uploaded logs despite of. I think others would run into same issue. If i could help with more informations or logs, then contact me please.

Thanks an greetings from Germany!

[futureweb]

Just to let you know, had the same issue, 10 minutes later the job finished normally without any error.
As there are already support cases open I believe I won't need to open one, I will see the next days once all Endpoints are through, if the error is gone.

Thanks
Patrick

[ds2]

Additional Information:

All VEB-Clients have completed the second run now and all finished without any error or warning.

I hope you can figure out the probleme for the first run with my logs, bevore all the others also run in this issue and cares about it.

Greetings

[Dima P.]

Hi guys,

I am glad to hear that you are up and running and thanks for all your help! We are still investigating and I’ll update this thread with the results asap. Thanks again!

[Dima P.]

Follow up: we have confirmed that this is just a glitch in the backup ‘pre-check’ logic. As a result, VEB job may fail only once even before starting the actual merge process. No data loss detected and all the restore points are valid. We will add the fix to the VBR’s Update 1. Thanks for reporting and sorry for any inconvenience caused.

[lp@albersdruck.de]

Follow up: we have confirmed that this is just a glitch in the backup ‘pre-check’ logic. As a result, VEB job may fail only once even before starting the actual merge process. No data loss detected and all the restore points are valid. We will add the fix to the VBR’s Update 1. Thanks for reporting and sorry for any incontinence caused.

Thanks for the information.
And I really hope this error did not cause any incontinence

[Dima P.]

Thanks! These Russian spellcheckers...

[isle-gmbh]

Just to inform you,
I'm getting the error message after upgrade of B&R to v9 SP1 as well (9.0.0.1491).
So this is still an issue although SP1 of B&R has been installed.

Code: Select all

Full backup file merge failed Error: Not transformed OIB was not found.
CreationTime: "25.04.2016 06:01:03"

BR, Oliver S. ISLE GmbH.

[Dima P.]

Hello Oliver,

Can you please open a case and copy the case ID to this thread? Thank you.

[isle-gmbh]

Case # 01818504

[Dima P.]

Oliver,

QA checked your logs - it looks like you are still using old VEB version. Kindly, update it as the said issue was fixed in the latest version 1.5. Thanks!