MESSAGE_METHOD_UNKNOWN(OpaqueRef:***.get_record)

[olivierlambert]

Because your XO isn't connected anymore to the pool when you started the backup.

[runfi85]

No, that's not the problem. XO can reach the servers without problems, I see the list of VMs

What else could be?

[runfi85]

Any suggestion??

DELTA Backups are working without any problem, Disaster Recovery backups are giving Error: no opaque ref found

[Danp]

Sounds like a bug to me. Does it also occur if you run a backup on a new job instead of an existing one?

[runfi85]

It happens in both situations. If I run a new backup job, I get that error on most VMs, on other I get "write ECONNRESET" error.
The strange thing is that the schedulet delta backups on the same storage worked without issues.
There seems to be a problem with Disaster Recovery backups.. And it all happened after I updated to the latest master commit and restarting..

[Danp]

There are a few things that you could try --

• Try to locate the problematic commit using git bisect. See this post for more details.

• Try to replicate the problem using XOA. I'm sure @olivierlambert would extend your trial if needed. Just ask nicely.
  .

• Build a new XO VM and retest to see if the problem persists.

FWIW, I ran a DR backup on a single VM the other day and it worked as expected. In my case, the source and destination SR was the same.

[lawrencesystems]

@runfi85

I was getting this same error for full backups, updated to latest version in our lab when doing full backups. I updated to the latest commit db99a and the problem was solved.

[runfi85]

@lawrencesystems I’m already at the latest commit

Thanks anyway

[lawrencesystems]

@runfi85
Are you using this https://github.com/ronivay/XenOrchestraInstallerUpdater to do the builds and if so have you pulled the latest as that also has some recent updates for the build.

[runfi85]

@lawrencesystems yes I’m using that bash script and the build is up to date to the latest commit

[caponate]

I was getting the same thing, after updating to the latest XO the issue was resolved. I will note my 8.3 beta system did not have this problem. Only our production 8.2 servers in the office.

[runfi85]

XO is updated and rebooted, but still getting the same error:

[runfi85]

Hello, I updated XO to the latest commit 7af08 but still getting the same error.

Is this really happening only to me?

[olivierlambert]

Seems like, yes. That's why I bet on a build issue on environment issue on your side. Please try to build it following our documentation, from a fresh folder to be sure you are directly on the latest commit (you can check that in the About view)

[runfi85]

@olivierlambert Done but still not working. Any way to install a trial of xenorchestra to troubleshoot the problem?

[runfi85]

I think I found the issue, it seems that xen cannot connect to the SMB Storagebox used for backup destination. I need to investigate further, thank you

[olivierlambert]

Let me know if you need a free XOA trial

[runfi85]

Thank you @olivierlambert, after disconnecting and recreating the SMB Storage I could successfully complete the full backup.
Just one note: I suggest to show more detailed errors, for example: "backup error, Target SR not working"

[olivierlambert]

I'm not sure it's trivial. In which state was your BR/remote in Settings/remote before removing it? Because if it's still view as connected, it's hard to know the real problem. If you can reproduce, that would be helpful.

[runfi85]

@olivierlambert it was connected, but when trying to rescan the virtual disks I got an error, I tried to disconnect and could not reconnect the servers.
Maybe you could test the Target SR before starting the backup and check if is reachable?