XCP-ng 8.3 betas and RCs feedback 🚀

stormi

@Andrew I now pushed a rebased build of XAPI in the xcp-ng-lab repo.

stormi

@olivierlambert Will XCP-ng 8.3 be using Xen 4.17 for the release (like XenServer)? If so, when will Xen 4.17 be part of the base install?

I'm the one who can answer the second part of this question, actually. It's all built in the xcp-ng-ci repository. We are now testing them in CI, will fix issues if there's any, and then this will reach the xcp-ng-base repository and make the packages in xcp-ng-lab deprecated (and should update them smoothly).

Andrew

@stormi So, is it better/safer to now just use xcp-ng-ci for an actual representation of XCP 8.3/Xen 4.17 beta/RC release?

stormi

@Andrew No, it's not safer, as it's not tested, and there are far more changes than just Xen.

archw

@stormi said in XCP-ng 8.3 beta :

yum update --enablerepo=xcp-ng-lab

After the new updates came out yesterday, I did the update but, this time, I did what you said (" yum update --enablerepo=xcp-ng-lab") and it worked perfectly...thanks!

stormi

New updates are propagating to the mirrors as I write this. Xen 4.17 for everyone, XAPI 24.11, and various other changes bringing us closer to a release candidate for XCP-ng 8.3.

Users of Xen 4.17 from the xcp-ng-lab repository don't need to use this repository anymore, as the new update supersedes it.

ph7

@stormi
I tried it but....
I will try to paste a pic here, don't laugh at me

I have a few weeks ago installed nut client:
yum --enablerepo=* install nut-client.x86_64
I hope that didnt break it

ph7

@ph7
I was running xo from a container in truenas
I shall fire up a fresh xoce

---------- edit -------------------
Here is the output from my 2:nd try, now with a fresh xoce

-1(Command '['yum', 'update', '--disablerepo=*', '--enablerepo=xcp-ng-base,xcp-ng-updates', '-y']' returned non-zero exit status 1, , Traceback (most recent call last):
File "/etc/xapi.d/plugins/xcpngutils/init.py", line 119, in wrapper
return func(*args, **kwds)
File "/etc/xapi.d/plugins/updater.py", line 96, in decorator
return func(args, **kwargs)
File "/etc/xapi.d/plugins/updater.py", line 182, in update
return install_helper(session, args, 'update')
File "/etc/xapi.d/plugins/updater.py", line 153, in install_helper
raise error
CalledProcessError: Command '['yum', 'update', '--disablerepo=', '--enablerepo=xcp-ng-base,xcp-ng-updates', '-y']' returned non-zero exit status 1

ph7

@ph7
I removed Netdata then I was able to update the host from the cli.
I only have one host now so I dont know if it works from within xo.

Houbsi

Can it be that after update and reboot and the bar goes to the left it needs a lot of time to reboot?

On tty 2 it hangs after “stopped the zen xenstore”

ph7

@Houbsi
If that was a reply to me, no
The update didnt even start
but after i removed netdata it seems like the update worked.
I got a 4 hung tasks after the " Smart reboot" i tried before I found the netdata issue,
I have to deal with that tomorrow

good night

Houbsi

So. I will post tomorrow some screenshots. But the install went 100% on first try. But the reboot process got timed out and it killed the job. After reboot everything works fine at the moment.

archw

@stormi
Worked for me at the first site but it did one weird thing that its never done before.

When I applied the patches to the pool master, it disappeared from the XOA GUI. Normally, after you apply the updates, all the hosts (including the master) disappear for a few minutes and then come back but this time they didn't. I had to just reboot it from the Dell Idrac GUI. It rebooted and then all the hosts came back.

stormi

@ph7 I did push a rebuilt netdata for Xen 4.17 yesterday, so it should have worked, but I admit I haven't tested the update process myself so maybe I've missed something.

Can you extract the error messages from the output of yum history?

yum history to get the list of operations, and yum history info NUMBER for the details of a given call.

stormi

@Houbsi What was the update method used?

Houbsi

@stormi
I used "yum update" with the 8.3 repo in the yum.repos.d file activated.

After that i always do a sanity reboot.

stormi

@Houbsi So what job got killed waiting for the reboot? I don't have all context here

Houbsi

@stormi

[ OK ] Stopped The Xen xenstore.
[ !!   ] Forcibly rebooting: job timed out

That's all what i got from the console, if there is a log somewhere with more information, i'll be happy to provide it to you!

stormi

@archw Maybe similar issue to what Houbsi saw, long shutdown due to a stuck task.

stormi

@Houbsi anything above this that would tell us what job was stuck? The xenstore stop ended well, as the [ OK ] message indicates.