XCP-ng 8.3 updates announcements and testing

ph7

I also take this opportunity to call for more feedback on the previous batch of updates,

Well I updated a few days ago, tough I dont run much of the updated functions on my simple home lab, it all seems to work fine.
i7 gen 4 and NFS

Now testing the new updates......

ph7

The new template for debian 13 is working in XO-Lite

gskger

@stormi Updated the usual suspects (HP ProDesk 600 G6, Dell Optiplex 9010, Dell R720) with no problem. Host run as expected.

stormi

@acebmxer said in XCP-ng 8.3 updates announcements and testing:

@stormi
How to revert changes if needed to? and/or how to switch back to normal repo?

The command only enables the testing repositories for the time of the update, so no need to disable them afterwards.

Reverting changes can be done with yum downgrade, but it's not always doable. XAPI updates can come with an upgrade of the XAPI database. If you downgrade, then XAPI with detect that the database is too recent and will refuse to start.

So, you can technically downgrade the files, but not the state.

acebmxer

@stormi

Thanks for the reply back. Update when sucessfull. Windows Server 2025 iso now properly installs.

At work I was not able to install default certs for UEFI due to one failing to download. Run these updates and I was able to successfully install the certs to the host.

JeffBerntsen

@stormi My "test/production" system, an HP DL165 is updated and running normally with the updated updates. Not seeing any change with secure boot VMs at all, i.e. working just fine.

gduperrey

New update candidates for you to test! (adding to the previous batch again)

New updates join the previous batch of update candidates. They're the last ones.

A new XSA (Xen Security Advisory) was published on the 21th of October, and updates to Xen address the disclosed vulnerabilities. We also reverted a change in XAPI that we deemed risky.

Additionally, we also publish an updated Intel-Ice alternate driver.

xen:
- XSA-475 - Potential risks include Denial of Service (DoS) impacting the whole host, information exposure, or escalation of privileges. There are two vulnerabilities related to hypercalls in the Viridian code:
  - CVE-2025-58147: Out-of-bounds write in vpmask_set() from hypercalls using the HV_VP_SET Sparse format.
  - CVE-2025-58148: Out-of-bound read in send_ipi() from hypercalls using any format, that could lead to a wild vCPU pointer.
xapi:
- We reverted a change related to how rsyslog configuration is handled. The way XenServer handled the change seemed risky to us, we'll take the time to make it in a safer way.

Optional packages:

Alternate Driver: Updated to newer version.
- intel-ice-alt: Update driver sources to v1.17.2

Test on XCP-ng 8.3

yum clean metadata --enablerepo=xcp-ng-testing,xcp-ng-candidates
yum update --enablerepo=xcp-ng-testing,xcp-ng-candidates
reboot

The usual update rules apply: pool coordinator first, etc.

Versions:

xapi: 25.27.0-2.2.xcpng8.3
xen: 4.17.5-20.2.xcpng8.3

Optional packages:

Alternate drivers:
- intel-ice-alt: 1.17.2-1.xcpng8.3

What to test

Normal use and anything else you want to test.

Test window before official release of the updates

~2 days.

gskger

@gduperrey Works on my play-/homelab (HP ProDesk 600 G6, Dell Optiplex 9010). Can't update my Dell R720s GPU cluster at the moment, though.

JeffBerntsen

@gduperrey This is working without problems on my test lab system, an HP Proliant DL165 G7

gduperrey

Updates published: https://xcp-ng.org/blog/2025/10/23/october-2025-security-and-maintenance-update-for-xcp-ng-8-3-lts/

Thank you for the tests!

manilx

@gduperrey Updated 2 pools/2 servers @home, updated 3 pools/5 servers @ office with RPU.
All fine.

flakpyro

Updates 3 pools. All fine except for the last host on the last pool ran into a yum mirror error and failed. Manually running yum update and rebooting the host worked. Sadly i then had to move VMs back around since the RPU failed and the process of migrating VMs back did not kick off as a result.

Error was

One of the configured repositories failed (Unknown),
 and yum doesn't have enough cached data to continue. At this point the only
 safe thing yum can do is fail. There are a few ways to work "fix" this:

     1. Contact the upstream for the repository and get them to fix the problem.

     2. Reconfigure the baseurl/etc. for the repository, to point to a working
        upstream. This is most often useful if you are using a newer
        distribution release than is supported by the repository (and the
        packages for the previous distribution release still work).

     3. Run the command with the repository temporarily disabled
            yum --disablerepo=<repoid> ...

     4. Disable the repository permanently, so yum won't use it by default. Yum
        will then just ignore the repository until you permanently enable it
        again or use --enablerepo for temporary usage:

            yum-config-manager --disable <repoid>
        or
            subscription-manager repos --disable=<repoid>

     5. Configure the failing repository to be skipped, if it is unavailable.
        Note that yum will try to contact the repo. when it runs most commands,
        so will have to try and fail each time (and thus. yum will be be much
        slower). If it is a very temporary problem though, this is often a nice
        compromise:

            yum-config-manager --save --setopt=<repoid>.skip_if_unavailable=true

xcp-ng-base: Check uncompressed DB failed

acebmxer

Updated AMD Ryzen pool at home and update two Intel Dell r660 and r640 pools at work. No issues to report back.

marcoi

updated my two hosts without issues.

stormi

IMPORTANT NOTICE!

After publishing the updates, we discovered a very nasty bug when using the UEFI certificates that we distribute. Long story short, they're too big, and there's only limited space (57K), and combined to a preexisting bug in varstored, this will cause the VM to stop booting after Windows or any other OS attempts to append to the DBX (revocation database).

We pulled the varstored update, but those who updated can be affected.

There are conditions for the issue:

Existing VMs are not affected, unless you propagated the new certs to them
New VMs are affected only if you never installed UEFI certs to the pool yourself (through XOA or secureboot-certs install), or cleared them using secureboot-certs clear in order to use our default certificates.

If you have the affected version of varstored (rpm -q varstored yields varstored-1.2.0-3.1.xcpng8.3) :

on every host, downgrade it with yum downgrade varstored-1.2.0-2.3.xcpng8.3. No reboot or toolstack restart required.
if you have affected UEFI VMs, that is VMs that meet the conditions above but are not broken yet, don't install updates, turn them off, and fix them by deleting their DBX database: https://docs.xcp-ng.org/guides/guest-UEFI-Secure-Boot/#remove-certificates-from-a-vm. This has to be done when the VM is off. Your OS will add its own DBX afterwards.
If you already have broken VMs (this warning reaching you too late), revert to a snapshot or backup. Other ways to fix them will require a patched varstored currently in the making.

manilx

@stormi Downgraded all hosts @home and @office. I have not done anything to running Windows Server VM's or touch anything regarding certs. So I guess I'm all good.

stormi

@flakpyro Thanks for letting us know. I suppose there was a mirror that was not ready yet, or had a transient issue, and unfortunately XOA's rolling pool update feature is not very resilient to that at the moment.

flakpyro

@stormi

I reverted the package however i initially followed the directions provided by vates in the release blog post and ran "secureboot-certs clear" then on each VM with Secure boot enabled i clicked "Copy the pools default UEFI Certificates to the VM".

After reverting the updates and running secureboot-certs install again i went back and clicked "Copy the pools default UEFI Certificates to the VM" again thinking it would put the old certs back.

It sounds like this may not be enough and i need to remove the dbx record from each of these VMs. Am i correct or was that enough to fix these VMs?

Per the docs:

varstore-rm <vm-uuid> d719b2cb-3d3a-4596-a3bc-dad00e67656f dbx

Note that the GUID may be found by using varstore-ls <vm-uuid>.

When i run the command i see

Exmaple:

varstore-ls f9166a11-3c3f-33f1-505c-542ce8e1764d
8be4df61-93ca-11d2-aa0d-00e098032b8c SecureBoot
8be4df61-93ca-11d2-aa0d-00e098032b8c DeployedMode
8be4df61-93ca-11d2-aa0d-00e098032b8c AuditMode
8be4df61-93ca-11d2-aa0d-00e098032b8c SetupMode
8be4df61-93ca-11d2-aa0d-00e098032b8c SignatureSupport
8be4df61-93ca-11d2-aa0d-00e098032b8c PK
8be4df61-93ca-11d2-aa0d-00e098032b8c KEK
d719b2cb-3d3a-4596-a3bc-dad00e67656f db
d719b2cb-3d3a-4596-a3bc-dad00e67656f dbx
605dab50-e046-4300-abb6-3dd810dd8b23 SbatLevel
fab7e9e1-39dd-4f2b-8408-e20e906cb6de HDDP
e20939be-32d4-41be-a150-897f85d49829 MemoryOverwriteRequestControl
bb983ccf-151d-40e1-a07b-4a17be168292 MemoryOverwriteRequestControlLock
9d1947eb-09bb-4780-a3cd-bea956e0e056 PPIBuffer
9d1947eb-09bb-4780-a3cd-bea956e0e056 Tcg2PhysicalPresenceFlagsLock
eb704011-1402-11d3-8e77-00a0c969723b MTC
8be4df61-93ca-11d2-aa0d-00e098032b8c Boot0000
8be4df61-93ca-11d2-aa0d-00e098032b8c Timeout
8be4df61-93ca-11d2-aa0d-00e098032b8c Lang
8be4df61-93ca-11d2-aa0d-00e098032b8c PlatformLang
8be4df61-93ca-11d2-aa0d-00e098032b8c ConIn
8be4df61-93ca-11d2-aa0d-00e098032b8c ConOut
8be4df61-93ca-11d2-aa0d-00e098032b8c ErrOut
9d1947eb-09bb-4780-a3cd-bea956e0e056 Tcg2PhysicalPresenceFlags
8be4df61-93ca-11d2-aa0d-00e098032b8c Key0000
8be4df61-93ca-11d2-aa0d-00e098032b8c Key0001
5b446ed1-e30b-4faa-871a-3654eca36080 0050569B1890
937fe521-95ae-4d1a-8929-48bcd90ad31a 0050569B1890
9fb9a8a1-2f4a-43a6-889c-d0f7b6c47ad5 ClientId
8be4df61-93ca-11d2-aa0d-00e098032b8c Boot0003
8be4df61-93ca-11d2-aa0d-00e098032b8c Boot0004
8be4df61-93ca-11d2-aa0d-00e098032b8c Boot0005
4c19049f-4137-4dd3-9c10-8b97a83ffdfa MemoryTypeInformation
8be4df61-93ca-11d2-aa0d-00e098032b8c Boot0006
8be4df61-93ca-11d2-aa0d-00e098032b8c BootOrder
8c136d32-039a-4016-8bb4-9e985e62786f SecretKey
8be4df61-93ca-11d2-aa0d-00e098032b8c Boot0001
8be4df61-93ca-11d2-aa0d-00e098032b8c Boot0002

So the command would be:

varstore-rm f9166a11-3c3f-33f1-505c-542ce8e1764d d719b2cb-3d3a-4596-a3bc-dad00e67656f dbx correct? 

Does "d719b2cb-3d3a-4596-a3bc-dad00e67656f " indicate the old certs have been re-installed?

dinhngtu

@flakpyro It's in the middle. But the "d719b2cb-3d3a-4596-a3bc-dad00e67656f dbx" part is always the same across all VMs.

After reverting the updates and running secureboot-certs install again

This will create a pool-level dbx which does not have the problem seen in varstored-1.2.0-3.1. So as long as you propagated the variables to all affected VMs without any errors, there's no need to delete them manually. But you can always delete them again if you're not sure.

flakpyro

@dinhngtu Ah okay, i was wondering if "d719b2cb-3d3a-4596-a3bc-dad00e67656f" indicated i was back on the safe certs since it is the same on all VMs since reverting and clicking "Copy the pools default UEFI Certificates to the VM"

So i need to run

varstore-rm f9166a11-3c3f-33f1-505c-542ce8e1764d d719b2cb-3d3a-4596-a3bc-dad00e67656f dbx

while powered off to be safe?