XCP-ng 8.3 betas and RCs feedback 🚀

yann

I realize one thing is not clear to me: is the "unreachable master" in your situation the one you switched to, or the original master? And how many hosts in total are in this pool when you launch this operation?

The logs we'll need to analyze will be the xensource.log from the first occurrence of designate_new_master on both old and new master, until the time of that exception.

Tristis Oris

@yann new one.

new part1 https://pastebin.com/ij0B7KHy
new part2 https://pastebin.com/0tQyvesY
old https://pastebin.com/dvAkEEbY

Tristis Oris

on 8.3 pool more and more VMs became attached to Control Domain. Backups fail due VDI_IN_USE.

can't unplug vdi:
xe vbd-list vm-uuid=***
xe vbd-unplug uuid=***
The server failed to handle your request, due to an internal error. The given message may give details useful for debugging the problem.
message: Expected 0 or 1 VDI with datapath, had 5

can't migrate to 8.2 pool due incompatible versions.

and already got coalesce on VM without backup.

upd
removed all stuck vdi, but sr coalesce won't move.

Tristis Oris

@Tristis-Oris after i fixed all dom0 stuck VDIs, backup succeed once and now stuck again.

yann

@Tristis-Oris very strange, your log seems to show that 2 designate_new_master requests were handled one after the other (at 12:11:33 and 12:12:45), both to switch to the same host - and we see in the logs that while the 1st one got all phases executed (1, 2.1, 2.2), the second one starts to have issues during "Phase 2.1: telling everyone but me to commit". Sending a second request should indeed not trigger the whole thing again, so something apparently went quite wrong, but what is indeed not obvious.

Tristis Oris

@yann so, i don't get indication about running master change and was able to run it again.

yann

@Tristis-Oris Oh OK. Had a try to run it several times myself (though on a 2-host pool), and I was able to see the operation performed twice, though apparently the second op did finish.
Running from the shell, if launch a second xe pool-designate-new-master while the first has not returned yet, it gets a DESIGNATE_NEW_MASTER_IN_PROGRESS error, but once it has returned there seem to be a window to do strange things.

yann

@Tristis-Oris and with a 3-host pool I can reproduce you issue on 2nd attempt: new master loses its xapi process

When the master is changed, the xapi service is stopped and then restarted but something seems to get wrong this time.

Among issues I realize that my former-master shows this at the time of the failing switch:

Oct 14 15:20:50 xcpng83-bzkcpvhy xsh: [ warn||0 ||xsh] TLS verification is disabled on this host: /var/xapi/verify-certificates is absent

(while both other hosts do have that file)

daemon.log on new-master shows systemd desperately trying to restart xapi:

Oct 14 15:20:59 xcp-ng-hqerhcgv xapi-init[1244028]: Stopping xapi: [  OK  ]
Oct 14 15:20:59 xcp-ng-hqerhcgv systemd[1]: Unit xapi.service entered failed state.
Oct 14 15:20:59 xcp-ng-hqerhcgv systemd[1]: xapi.service failed.
Oct 14 15:20:59 xcp-ng-hqerhcgv systemd[1]: xapi.service holdoff time over, scheduling restart.
Oct 14 15:20:59 xcp-ng-hqerhcgv systemd[1]: Cannot add dependency job for unit lvm2-activation.service, ignoring: Unit is masked.
Oct 14 15:20:59 xcp-ng-hqerhcgv systemd[1]: Cannot add dependency job for unit lvm2-activation-early.service, ignoring: Unit is masked.
Oct 14 15:20:59 xcp-ng-hqerhcgv systemd[1]: Starting XenAPI server (XAPI)...
Oct 14 15:20:59 xcp-ng-hqerhcgv systemd[1]: Started XenAPI server (XAPI).
Oct 14 15:20:59 xcp-ng-hqerhcgv xapi-init[1244047]: Starting xapi:
Oct 14 15:21:00 xcp-ng-hqerhcgv systemd[1]: xapi.service: main process exited, code=exited, status=2/INVALIDARGUMENT
Oct 14 15:21:00 xcp-ng-hqerhcgv xapi-init[1244078]: Stopping xapi: [  OK  ]

and xensource.log is very similar to yours. Congrats, that's a nice bug

Tristis Oris

@yann yay!

Strebor

Just a note to say thanks to all involved! Excellent job!

I yesterday upgraded our 8.3 release pool to 8.3 production. 2x Intel NUC11TNKi5. Running very nicely! The upgrade from release candidate with the ISO written to USB media went smoothly. Also the yum updates after the upgrade.

Now looking forward to see XO Lite be completed!
Keep up the great work!

yann

@Tristis-Oris I also reproduced the issue on 8.2.1, and record the issue. As for your pool left in a tricky state, would it be reasonable for you to reinstall the impacted hosts?

Tristis Oris

@yann yep, already reinstall it.

gb.123

Installed in my Homelab. Install went smoothly.

Small bugs I have noticed :
1.
XCP-NG Install Error.jpg

This ONLY happens when using KVM and 'scaling' the display to fit window.

XCP-NG Update Error.jpg

This happens ONLY on the first boot after applying patch ( 6 programs ) once release iso is installed.
This goes away from the second boot.

Everything seems to be working smoothly for now. Kudos to the Vates Team !

gb.123

This post is deleted!

stormi

Regarding xsconsole, I think it always had this issue where at some point it thinks (rightfully or not) that the screen is too small for it to run.

Tristis Oris

Already every VM at 8.3 pool stuck at control domain. Can't detach them without pool reboot. Any workaround?

vbd.delete
{
  "id": "7af4b39d-d6aa-2589-509f-f082a1f0f2bb

"
}
{
  "code": "OPERATION_NOT_ALLOWED",
  "params": [
    "VBD '7af4b39d-d6aa-2589-509f-f082a1f0f2bb

' still attached to 'e9e4e600-b180-481d-9a1b-a45be8fc192e

'"
  ],
  "call": {
    "method": "VBD.destroy",
    "params": [
      "OpaqueRef:65481d14-d3de-82d7-fa7d-95131f14a209

"
    ]
  },
  "message": "OPERATION_NOT_ALLOWED(VBD '7af4b39d-d6aa-2589-509f-f082a1f0f2bb

' still attached to 'e9e4e600-b180-481d-9a1b-a45be8fc192e

')",
  "name": "XapiError",
  "stack": "XapiError: OPERATION_NOT_ALLOWED(VBD '7af4b39d-d6aa-2589-509f-f082a1f0f2bb

' still attached to 'e9e4e600-b180-481d-9a1b-a45be8fc192e

')
    at Function.wrap (file:///opt/xo/xo-builds/xen-orchestra-202410181215/packages/xen-api/_XapiError.mjs:16:12)
    at file:///opt/xo/xo-builds/xen-orchestra-202410181215/packages/xen-api/transports/json-rpc.mjs:38:21
    at runNextTicks (node:internal/process/task_queues:60:5)
    at processImmediate (node:internal/timers:454:9)
    at process.callbackTrampoline (node:internal/async_hooks:130:17)"
}

Ajmind 0

I have tried to install an old Intel Server (SR2625/S5520UR / Dual Xeon E5645) running XCP-NG XCP-NG 8.3 beta with 8.3 release ISO:

The Server was first running XCP-NG 8.2.1 and was upgraded with 8.3 beta 2 ISO.

I could run the installation proccess until

completing installation 50%

And nothing happens further:
PXL_20241021_125706035.MP (Groß).jpg

Install media was freshly created and booted in EFI mode.
Any idea how to finish this process?

Finallf

@Ajmind-0
try running the installation in safe option.

jhansen

Version XCP-NG 8.3 RC
XO lastest version

Hello,
I use XO with Delta Backup to back up my VM's. I have activated NBD and CBT in the backup. The backups run via the management interface with NBD activated.
The first full backup runs error-free.
The second backup with delta gives me the error message cleanVm: incorrect backup size in metadata and VDI_IN_USE and the backup ends with an error.
After the delta backup, I then have snapshots on the storage that are blocked with "Control domain on host xxx". I can then no longer delete these. The only option is to reboot all Xen servers in the pool, after which the snapshots are still there, but can then be disconnected and forget to get them deleted.
I was able to reproduce this error on different pools with different servers, all 8.3. It is always different VMs that trigger the error and others that are backed up without errors.
If I delete all control domain blocked snapshots and also the purged snapshot from the storage, the next full backup runs error-free again.
Does anyone have the same problem and maybe a solution?
regards Joerg

jhansen

I saw that Tristis Oris seems to have a similar problem with snapshots.

Something else, when updating from 8.2.1 to 8.3 RC.
My 8.2.1 servers were still running on BIOS boot. When updating to 8.3 RC I got the message that there was still a DOS table on the hard drive and the update from the USB stick was aborted.
A newly installed 8.3 server with UEFI cannot be integrated into the old 8.2 pool.

Here is a small workaround:
2 USB sticks, one with 8.2.1 and one with 8.3 RC.
Take one server out of the pool, not the master.
Reinstall with UEFI on 8.2.1.
Integrate server into the old pool.
make server as new pool master.
The server can now be updated to 8.3 without any problems, the master is on UEFI and all pool information and VM are on 8.3.
Now just take all other servers out of the pool, do a clean 8.3 UEFI installation and put them back into the pool.

Maybe that will help others with the same problem.

regards Joerg