XOSTOR appears to be broken on the new XCP-NG May 2026 update

ccooke

After the update, the master server (first to be updated, of course) can no longer attach the xostor SR.
I have verified that the rolling pool update did look for updated packages, and that xostor is still installed and working (thankfully the linstor services are working fine, so there's no risk of data loss here).

When trying to mount the SR, the master host gets this error:

[19:02 goethite ~]# xe pbd-plug uuid=91e19949-cf2c-b1bb-7dda-d2c114c113a1 
Error code: SR_BACKEND_FAILURE_46
Error parameters: , The VDI is not available [opterr=Could not load 5cb360ea-0904-472c-a3f9-1a44b97936c3 because: ['UNKNOWN_XENAPI_PLUGIN_FUNCTION', 'getInfo'] (openers: {'hematite.earth.gkhs.net': {'26922': {'process-name': '+tapdisk', 'open-duration': '612038496'}}, 'goethite.earth.gkhs.net': {}, 'siderite.earth.gkhs.net': {}})],

I would advise nobody updates a host using XOSTOR until this gets fixed!

dthenot

@ccooke Hello,

You should be able to make the XOSTOR SR work again if you update sm and sm-fairlock on the other hosts.

yum update sm sm-fairlock

Then you should be able to re-plug the SR on the master and proceed with the RPU.

ccooke

@dthenot ... The XOSTOR SR is working on the other hosts. It's only the master which updated where it fails to plug the pbd. Given that it can't mount the pbd after doing the update, I am very doubtful about updating another host.
What's more, I can't without shutting down a large number of VMs or doing a long migration off the XOSTOR, because I only have N+1 redundancy on my personal cluster, and we're already one host down because the master can't rejoin the XOSTOR SR.

In short, this looks like a very dangerous suggestion - sorry!

ccooke

@dthenot
To be clear here: The XOSTOR SR is working - but only on the hosts that have not ugraded.
If you can tell me that this is a known issue and that updating sm and sm-fairlock on all of the hosts should still fix it so that the pool master (the only updated host, which does have the updated sm and sm-fairlock) can successfully activate the XOSTOR SR, I'll do a full migration off XOSTOR and give it a try.

Just please tell me that this is the expected case before I have to go and potentially lose all my redundant VM storage and ability to do migrations! Completing the update without migrations will be rather painful.

ccooke

Right, to anyone seeing this - I got the confirmation I needed and can confirm that updating sm and sm-fairlock will allow the master host to mount the XOSTOR SR again properly.

RPU is now resumed and all looks good - Thank you, @dthenot

dthenot

@ccooke Hello,

We have a fix, we are aiming to validate it rapidly so it shouldn't happen for others.
Thank you for reporting the issue. I'll update the thread again when the update is available and it should be safe for other people going through here to update using the RPU.