XCP-ng 8.2 updates announcements and testing

nuts23

@stormi
upon update, I got a
Transaction check error:
installing package openvswitch-2.5.3-2.2.3.3.xcpng.x86_64 needs 1 inodes on the /var/log filesystem

Which translated that the ovsdb-server logs were too many and had filled up my log storage.
I had to run:
find . -name "ovsdb*" -delete
because rm couldn't handle so many ovsdb log files to delete.

stormi

@nuts23 Thanks for the feedback. Indeed, when the /var/log partition is out of inode, you may need to delete the files before installing the update.

RelNetDavidMerrill

Check out:

https://xcp-ng.org/forum/topic/1044/var-log-has-run-out-of-inodes

stormi

New update candidates are available to fix the recently disclosed Xen security issues (see http://xenbits.xen.org/xsa/ and https://support.citrix.com/article/CTX246572).

Please test as soon as possible. Objective of the test: make sure that there is no obvious regression. Just tell us that you installed the update candidate, the version of XCP-ng, and whether it works as well as before or not after a reboot.

XCP-ng 7.6

xen-4.7.6-6.4.1.xcpng

Installation of the update candidate:

yum install --enablerepo='xcp-ng-updates_testing' xen-dom0-libs xen-hypervisor xen-tools xen-dom0-tools xen-libs

Reboot required after the installation.

XCP-ng 7.5

xen-4.7.5-5.8.1.xcp

Installation of the update candidate:

yum install --enablerepo='xcp-ng-updates_testing' xen-dom0-libs xen-hypervisor xen-tools xen-dom0-tools xen-libs

Reboot required after the installation.

stormi

Anyone? This update candidate should be as safe as could be and internal tests are OK, but I don't want to push it before I get feedback from a few testers.

AllooTikeeChaat

Updated one host without issue and currently testing.

nuts23

security fix installed. Does it require a reboot of the system?

stormi

@nuts23 yes

borzel

@stormi testhost updated, restored some part of our infrastructur (DC, Mail, ...) on local ZFS pool, connected the VMs to a test network. Runs. A bit slow (more vCPU used than the PCPU can delivier), but runs.

rizaemet 0

@stormi said in Updates announcements and testing:

Anyone?

I updated one host in a pool and reboot. It is working as before. Seems no problem. I can't see any diffrence vcpu usage as @borzel say.

AllooTikeeChaat

So far so good .. I haven't noticed any visible change in performance like @borzel but still monitoring.

stormi

Since there's an update for both XCP-ng 7.5 and 7.6, please all tell me which version you tested

borzel

@AllooTikeeChaat there was no performance change, was the first use of that test host

@stormi XCP-ng 7.6, just HVM, no PV

stormi

If all goes well I'll push the updates to everyone on monday. Meanwhile, please go on with the testing if you can.

rizaemet 0

@stormi XCP-ng 7.6

stormi

Thanks to everyone who tested the update candidate. I've now pushed the security update for XCP-ng 7.6. I'm holding that of XCP-ng 7.5 until at least one person confirms that it's working for them.

AllooTikeeChaat

Sorry been away this weekend .. @stormi XCP-ng 7.6

Ultra2D

I've installed the updates on a pool. Now every time I migrate a pfSense HVM VM within the pool the console is gone and memory usage is at 100%. I did do the following on the pool master: yum update xenopsd xenopsd-xc xenopsd-xenlight --enablerepo='xcp-ng-updates_testing'. The package version also differs from other hosts (0.66.0-1.1.xcpng versus 0.66.0-1.el7.centos). Would that be the cause?

stormi

@Ultra2D Could you downgrade to the previous version with yum downgrade xenopsd xenopsd-xc xenopsd-xenlight, restart the toolstack and tell us if migration works better?

Also make sure that you already have the latest updates for xcp-emu-manager, which fixed many migration issues about two months ago. See https://github.com/xcp-ng/xcp/wiki/Updates-Howto#a-special-word-about-xcp-ng-75-76-and-live-migrations.

stormi

@Ultra2D In fact, you may be right, differing versions of those packages may cause an issue, because both hosts may behave differently regarding VMs that have no platform:device-id set.

So another test to do would be install the update candidate for those packages on all hosts and restart their toolstack, then test migration of that VM again.