XCP-ng 8.3 betas and RCs feedback 🚀

stormi

@matt-plan8 I don't understand. Pool and cluster are the same thing. Pool for XCP-ng is what VMWare calls cluster.

matt-plan8

@stormi sorry I consider HA a cluster, pool I consider just a group of machines in the same management domain.

They are in a single pool, they aren't HA

archw

@stormi said in XCP-ng 8.3 beta :

yum downgrade edk2-20180522git4b8552d-1.5.1.xcpng8.3

It didn't work. It just sits there with the attached screen.

stormi

@archw You're quoting the downgrade command. May I assume you tried yum update edk2 --enablerepo=xcp-ng-testing,xcp-ng-ci instead and got version edk2-20220801-1.7.3.xcpng8.3.x86_64?

archw

@stormi
Sorry...I cliked reply to teh wrong post.

Yes, I ran "yum update edk2 --enablerepo=xcp-ng-testing,xcp-ng-cii" and let it install. I then rebooted the VM (several of them) and tried to start them and none worked. I then ran "yum downgrade edk2-20180522git4b8552d-1.5.1.xcpng8.3" and the VM's fired right up.

stormi

I just published new updates for XCP-ng 8.3:

edk2-20220801-1.7.3.xcpng8.3, which some of you already tested. It fixes UEFI pfSense booting.
e2fsprogs-1.47.0-1.1.xcpng8.3 (and subpackages): brought to the same level as latest XCP-ng 8.2.1. If you installed using the latest test installation ISO, you already have it.
r8125-module-9.012.04-1.xcpng8.3 which updates the r8125 driver.

gskger

@stormi While I can not test the changes, the update went well. However, the restart took a few minutes longer than expected (or I was impatient again).

wilsonqanda

@stormi Nice thanks so I can run the complete update on XOA and not worry about having to remove that single edk2 anymore that wasn't working. Will do that later

stormi

So, @ThierryEscande found the issue which made all UEFI VMs crash on @archw's server and patched it.

This was the last issue before we could publish XCP-ng 8.3 Beta 2, so this is good news.

Now, we need you all to test the patched RPM and check your UEFI VMs still work well with it: start, function normally...

To install it:

yum update edk2-20220801-1.7.3.1.xcpng8.3 --enablerepo=xcp-ng-ci

No need to reboot, but UEFI VMs need to be restarted afterwards (when doable. If you have a specfic VM that you can't reboot yet, it's OK to leave it alone).

In case things do not work as expected, you can come back to the previous build with :

yum downgrade edk2

Andrew

@stormi The new version still works for me (linux/freebsd).

bufanda

@stormi Tested the new edk package at looks good to me. was able to reboot linux and freebsd VMs without a hitch with UEFI boot mode.

stormi

Thanks to both of you. I moved the fixed edk2 package to repo xcp-ng-base, so all testers will get it next time they update.

archw

@stormi
Fantastic and way to go...thanks for all your hard work!

olivierlambert

Also thanks to you @archw for your feedback, that's how we like to work

stormi

Everyone: we are about to release XCP-ng 8.3 Beta 2!

It was delayed by one month due to the UEFI issues discussed above, now fixed.

The last thing we need is your feedback on the updated installation ISO: https://updates.xcp-ng.org/tmp/xcp-ng-8.3.0-beta2-test4.iso

I hope I have more success than with the test2 ISO for which I don't remember getting any feedback

[Update: link updated to the test4 ISO. Don't upgrade an existing XCP-ng 8.3 alpha or beta with this: just update with XO or yum instead]

wilsonqanda

@stormi after the update it seem to work well for me. Well the update from edk2 at least is working now. Haven't tried the iso.

wilsonqanda

@stormi said in XCP-ng 8.3 beta :

https://updates.xcp-ng.org/tmp/xcp-ng-8.3.0-beta2-test3.iso

Testing it right now and migrating my VM over to see. My previous install SR was causing issue and disappearing on every restart for some reason lol (don't even know what causes it so decided to start fresh ). So decided to might as well test the "xcp-ng-8.3.0-beta2-test3.iso" new iso you uploaded. Will provide feedback in a few minutes.

wilsonqanda

@stormi
Good things I notice on the new beta2-test3 iso:

"efi: EFI_MEMMAP is not enabled" used to take minutes is now down to 30 sec
host joins other host or pool really quickly (it sometime takes forever or doesn't work for me before until I manually toggle disconnect and reconnect) (hosts in pool always work fine but hosts from different pool was slow previously not sure why when they all start up at the same time is usually the issue especially when network is up slower than the hosts. I assume host search timeout before network is up.)
UEFI is definitely fixed now

All good feedback. No drawback yet.

Also when is NFS v4 available believe it was end of 1st quarter (March???) from what i heard is that still true?

Thanks otherwise it is good on my short test so far.

Tristis Oris

With previous iso i got some weird single problems, but can't reproduce them - mng interface resert after reboot, can't connect to XO, problems with update.
Now it works at first attempt. And it show hostname at ssh instead of ip.

also notice about "efi: EFI_MEMMAP is not enabled".

stormi

"efi: EFI_MEMMAP is not enabled" is a normal message in the context of XCP-ng. It's just a bit annoying because it raises questions for many users.

Regarding NFSv4 (more precisely: support for servers which only offer NFSv4 and not 3+4), isn't support already present?