Hallo zusammen,
auf unserem Proxmox-Cluster haben wir 3x Windows-Server 2019 verteilt für einen Failovercluster.
Bei zweien geht die installation der neuen Version 0.1.271 (virtio-win-0.1.271.iso, virtio-win-gt-x64.exe) gut.
Bei einem geht das...
Ok, thanks. Didn't know that.
https://security-tracker.debian.org/tracker/CVE-2026-20337 and https://security-tracker.debian.org/tracker/CVE-2026-20338 are empty. Both are causing a DoS, so I would assume a fix should be important, though.
Hi @waynej
thanks for providing the additional insight and scripts!
I have validated the server and it is affected by the vulnerability and i also see the underflow you are reporting in kernel 7.0.14.
On kernel 6.17 the bpftrace script does...
As promised a short update: The patch got applied and will be included in the 1.1.4 update for proxmox-datacenter-manager [1]. The package has not yet been published, it will likely become available shortly. Once it is available you should be...
In order to upgrade I had to re-apply the commands you provided on a previous post:
rm /run/proxmox-datacenter-manager/api-cache/remote-*/subscription-state.json &&\
proxmox-datacenter-manager-admin support-status update &&\
apt update...
It's rather easy to find that information if you know your way around Debian's infrastructure a bit: https://security-tracker.debian.org/tracker/source-package/clamav
The "No DSA" status means that the package maintainers and/or the Debian...
There seems to have been an update as the date has changed in this output, but the version number itself is still the old one.
# clamd --version
ClamAV 1.4.4/28111/Wed Sep 2 08:24:01 2026
Would be nice to get some information if ClamAV on PMG...
Update:
Found it. It was core files dumped from a crash which I had not caught. They totaled ~3.4TB but didn't end up in df when checking. Removed these and I'm good to go.
pct exec 105 -- find / -xdev -type f -size +500M -printf '%s\t%p\n'...
We are pleased to share two major milestones that will help us better serve our customers and partners worldwide.
Global 24/7 enterprise support
Starting October 19, 2026, Proxmox Enterprise Support will be available around the clock, 365 days a...
I've tried searching but didn't really find a great reply.
I have an lxc with a 60GB rootfs mount, however the backup -> Size tells me the size is a whopping 3.45TB(!) and I cannot for the life of me understand why/how. I noticed that it takes...
We are pleased to share two major milestones that will help us better serve our customers and partners worldwide.
Global 24/7 enterprise support
Starting October 19, 2026, Proxmox Enterprise Support will be available around the clock, 365 days a...
from another post from me:
https://forum.proxmox.com/threads/on-limitations-and-maximums.179850/#post-834673
so currently there is a memory requirement of ~1MiB per resource + some overhead for the base system
there is also the disk space...
So, you are:
Running a OS that is EOL since two years
Exposing it to the internet
Disabling the firewall
If this is not a troll, I hope the lesson was learned...
correct, the token does not use/inherit the users tfa. in fact, tokens can't have a second factor at all.
This kind of use case is exactly why tokens exist in the first place to give access to users resources in an automated fashion that can be...
We are pleased to share two major milestones that will help us better serve our customers and partners worldwide.
Global 24/7 enterprise support
Starting October 19, 2026, Proxmox Enterprise Support will be available around the clock, 365 days a...
note that historic rrd/task data (etc.) will not be migrated over, so you might loose some rrd/task history (though i'm not sure if it's safe to just simply rename those folder/filenames too, they would be in /varlib/proxmox-datacenter-manager/ )
We are pleased to share two major milestones that will help us better serve our customers and partners worldwide.
Global 24/7 enterprise support
Starting October 19, 2026, Proxmox Enterprise Support will be available around the clock, 365 days a...
I can't say anything about the other parts of the writeup, but the "recommendation" for cleanup/rebuild irritates me:
I can't agree with this. This is a lot of work while you still could overlook something or the bad actor (now that the attack...
I would always recommend to stay within the bumpers of the system. Especially when an easy to use procedure is available.
IMHO this is such a rare situation that complicating UI with handling this case is counterproductive. That said - feel free...