Backup error with space available

Bane

@olivierlambert I don't see the drive I'm backing up to when I run mount -l, I do see it under remotes in settings.

olivierlambert

You can spot it here, it's very obvious.

Can you see the amount of Total disk available? The path you entered is used inside Xen Orchestra virtual Appliance, not from the host/XCP-ng perspective.

There's no such device as /dev/sdf/backup in your XOA.

You need to mount it INSIDE your XO VM, then to create a mount point, and finally point to this mountpoint.

This is a kind of advanced scenario, ideally, you should rely on a NFS share (the easiest thing).

Bane

@olivierlambert

Ok, that makes sense.

Thanks

olivierlambert

You are welcome, let us know if you need more assistance on setting the "remote"/backup repo

Bane

@olivierlambert

I have the backup setup now for my NFS share. The NFS remote tests fine and the backup job has some success. I am getting the below failures though on 5 VM's. I am posting one example, let me know if I should post the entire log.

 Security Onion (gamakbase)  
 Snapshot 
Start: Apr 22, 2022, 04:03:48 PM
End: Apr 22, 2022, 04:03:58 PM
 xcp-ng-backup 
 transfer 
Start: Apr 22, 2022, 04:04:02 PM
End: Apr 22, 2022, 04:38:03 PM
Duration: 34 minutes
Error: HTTP connection has timed out
Start: Apr 22, 2022, 04:03:58 PM
End: Apr 22, 2022, 04:38:03 PM
Duration: 34 minutes
Error: HTTP connection has timed out
Start: Apr 22, 2022, 04:03:48 PM
End: Apr 22, 2022, 04:45:18 PM
Duration: 42 minutes
Error: HTTP connection has timed out
Type: full

Danp

@Bane See this thread for solution to HTTP timeouts -- https://xcp-ng.org/forum/topic/5769/keep-getting-http-connection-has-timed-out-on-my-delta-backup?_=1650710290869

Bane

@Danp

I applied the fix mentioned in that thread, and now I have more VM's that succeed. However I still have a few that fail with the same HTTP timeout. Is there anything else I should look at?

olivierlambert

Maybe the huge number of DNS requests could also cause issues. What about testing it? (it's merged on master since recently, be sure to be on the latest commit!)

Bane

@olivierlambert Can you point me to instructions for that?

Thanks

olivierlambert

If you installed XO from our doc, then it's trivial: https://xen-orchestra.com/docs/installation.html#updating

Bane

@olivierlambert I updated to the master as you mentioned and I still get the http timeout error on 4 of my VM's.

olivierlambert

Okay so you are on latest commit? (the commit should be 3ecf099)

Bane

@olivierlambert

I'm using the trial with pro support included right now. So, I don't see the commit string in my about section. Is there another way to verify the commit string?

olivierlambert

Okay, so go on latest release channel and wait for tomorrow's release to make the update

Bane

@olivierlambert

When I run the below, I get destination 'xen-orchestra' already exisits.

git clone -b latest https://github.com/vatesfr/xen-orchestra

I'm not an expert so forgive me if this seems like a basic question.

olivierlambert

If you use XOA, then you just have to click on "Update" by using the "latest" release channel. No command line.
If you use XO from the source, then please follow the instructions: https://xen-orchestra.com/docs/installation.html#fetching-the-code

Bane

@olivierlambert

Ok, after updating now the backup fails with with:

SESSION_AUTHENTICATION_FAILED(root, Authentication failure)

I can log into the xcp-ng server with root from the XOA VM. So, I don't understand what is taking place here.

olivierlambert

Are you sure about the password? This error is clearly an auth issue

Bane

@olivierlambert

I deleted my server out and re-added it. I no longer have the authentication error. So back to the backup...

Bane

@olivierlambert

I still have HTTP timeouts on the backup. The quantity is less however.