Proxmox Virtual Environment 9.0 released!

t.lamprecht said:
This is to get in the cleanest state and avoid that the reboot takes longer than anticipated due to potnetially many machines needing to be stopped, and it's also a reminder that those with a cluster should leverage the (live) migration to reduce any impact for running VMs.
Besides that it normally is fine to keep virtual machines and containers running during the upgrade.
Click to expand...
Understood - just made me nervous when running pve8to9 script and saw this warning:

INFO: Checking for running guests..
WARN: 1 running guest(s) detected - consider migrating or stopping them.

Will try the upgrade and leave the OPNsense VM running - see what happens ;-)
 
This release and the beta has an issue with the graphical installer UI not displaying when using IPMI ASPEED 2600 remote video. Just a grey screen and moving cursor. I have to revert to the terminal to install.

1754418491867.png
 
mikeyo said:
This release and the beta has an issue with the graphical installer UI not displaying when using IPMI ASPEED 2600 remote video. Just a grey screen and moving cursor. I have to revert to the terminal to install.

View attachment 88889
Click to expand...

I just tested on an ASUS RS720-E11-RS12U board with ASMB11-iKVM that also got an ASPEED AST2600, it worked fine there.

1754420026659.png

Are you sure you did not switch the "virtual desktop" of the installers open box environment? That might have happened when scrolling to early while the background is still loaded, as getting the initial installer environment can take a bit on those hosts, especially if the ISO is attached as virtual media through the HTML5 remote console, as data is loaded on demand there. I had to wait almost a minute between grey screen and actual installation wizard being fully loaded so that it can show up.

If that wasn't the case: What exact server / mainboard / cpu are you using here? Maybe we have something that's closer to your HW in our test lab. FWIW, while at it, I also tried a AST2500 based IPMI from a Gigabyte MZ32-AR0 based system.

And, does the Terminal UI based environment works?
 
  • Like
Reactions: Stoiko Ivanov
Upgrade was smooth but i found this warning, how can i fix it?

Error: Package 'intel-microcode' has no installation candidate

1754420313231.png
 
t.lamprecht said:
I just tested on an ASUS RS720-E11-RS12U board with ASMB11-iKVM that also got an ASPEED AST2600, it worked fine there.

View attachment 88890

Are you sure you did not switch the "virtual desktop" of the installers open box environment? That might have happened when scrolling to early while the background is still loaded, as getting the initial installer environment can take a bit on those hosts, especially if the ISO is attached as virtual media through the HTML5 remote console, as data is loaded on demand there. I had to wait almost a minute between grey screen and actual installation wizard being fully loaded so that it can show up.

If that wasn't the case: What exact server / mainboard / cpu are you using here? Maybe we have something that's closer to your HW in our test lab. FWIW, while at it, I also tried a AST2500 based IPMI from a Gigabyte MZ32-AR0 based system.

And, does the Terminal UI based environment works?
Click to expand...
I waited a few minutes on the grey screen and nothing popped up. It's an Asrock Rack SIENAD8-2L2T EPYC board. It worked fine on PVE 8 so presumed something new in 9 broke it. Terminal UI works fine.
 
My upgrade did break the system. It is not booting anymore. When starting the PC, I end up on the BIOS screen; similar as described here. The system is located on an nvme ssd in a zfs file system. How can I recover?
 
  • Like
Reactions: vesalius
It's interesting that the memory usage is 101.15% on the pfSense (2.8.0) VM in PVE 9.
On PVE 8, the memory usage was around 70-75%. That was also incorrect, but not over 100%.

The memory balloon is off due to a missing implementation in FreeBSD to calculate the memory usage in QEMU.
Source: https://github.com/aborche/qemu-guest-agent/issues/19

2025-08-05_21-19.png
2025-08-05_21-32.png

monitor:
Code: 
# info memdev
memory backend: pc.ram
  size:  4294967296
  merge: true
  dump: true
  prealloc: false
  share: false
  reserve: true
  policy: default
  host nodes:

# info memory_size_summary
base memory: 4294967296
plugged memory: 0

Memory usage in pfSense:
2025-08-05_21-37.png
 
Last edited:
  • Like
Reactions: macromoon and neverstopdreaming
chillbird said:
Upgrade was smooth but i found this warning, how can i fix it?

Error: Package 'intel-microcode' has no installation candidate

View attachment 88891
Click to expand...
Check your /etc/apt/sources.list.
It should end with "main non-free contrib non-free-firmware" all the debian.org deb section.

for example:
deb http://ftp.de.debian.org/debian trixie main non-free contrib non-free-firmware

deb http://ftp.de.debian.org/debian trixie-updates main non-free contrib non-free-firmware

deb http://security.debian.org trixie-security main non-free contrib non-free-firmware
 
  • Like
Reactions: speedrapide10, chillbird and t.lamprecht
Hello, first post here!
Congratulations for the release.
Have a problem: on a recently installed pve 8.4 (update with apt commands) I am missing pve8to9: where can I find it?
 
albertogscotti said:
Have a problem: on a recently installed pve 8.4 (update with apt commands)
Click to expand...
You must have repos correctly configured either with Enterprise (if you have a subscription) or No-Subscription / Test - as shown here.
You can do all of this from the GUI. After that do the Update & Upgrade also from the GUI.
 
Not seeing the update to 9 yet.

root@myhost:~# pveversion
pve-manager/8.4.8/88beaf427bb10dad (running kernel: 6.8.12-13-pve)
root@myhost:~# apt upgrade && apt dist-upgrade
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
Calculating upgrade... Done
0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
Calculating upgrade... Done
0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
 
gfngfn256 said:
You need to follow the full guide for in-place upgrade, as shown here: https://pve.proxmox.com/wiki/Upgrade_from_8_to_9#In-place_upgrade
Click to expand...
Done. I ran the /usr/share/pve-manager/migrations/pve-lvm-disable-autoactivation command as suggested by the guide. Then ran the upgrade.

None of my containers or VMs start after the upgrade. They all show this when trying to auto start:
TASK ERROR: activating LV 'pve/data' failed: Check of pool pve/data failed (status:64). Manual repair required!

How do I fix this?
 
Last edited:
Jedis said:
Done. I ran the /usr/share/pve-manager/migrations/pve-lvm-disable-autoactivation command as suggested by the guide. Then ran the upgrade.

None of my containers or VMs start after the upgrade. They all show this when trying to auto start:
TASK ERROR: activating LV 'pve/data' failed: Check of pool pve/data failed (status:64). Manual repair required!

How do I fix this?
Click to expand...
ChatGPT helped me fix it. Sharing it here in case someone else needs.


First, you need to repair the broken thin‐pool metadata, then bring it (and its VG) back online, and finally re-enable it in Proxmox:


1. Repair the thin-pool metadata​


The “status:64” error means the pool’s metadata is corrupt or full. Repair it with:
lvconvert --repair /dev/pve/data
This reads the damaged metadata LV, writes a repaired copy from the pmspare LV, and swaps it back in place.

2. Activate the VG and pool​


Once the repair finishes without errors, activate your volume group and the pool itself:
vgchange -ay
lvchange -ay /dev/pve/data
This makes /dev/pve/data writable and available again.

You can confirm it’s active when lvs shows w (writable) instead of i in the third attribute column for “data”:
lvs -o lv_name,lv_attr

3. Re-enable the storage in Proxmox​


Finally, clear the “disabled” flag so Proxmox will use it:
pvesm set local-lvm --disable 0
systemctl restart pvedaemon pveproxy pvestatd
This flips local-lvm back on in /etc/pve/storage.cfg and reloads the storage daemons.

Verify with:
pvesm status
You should now see local-lvm listed as active again.
 
  • Like
Reactions: Damon1974, thomascgh, dgsnb and 1 other person
Just hopping in to say I upgraded from 8.4.1 to 9 and it wiped my efi partition. Unlike ulistermclane, I had a default ext4 lvm partition.
 
You must log in or register to reply here.
Top Bottom