Search results

hm yeah, this seems to be an exception to an unwritten rule kind of situation. anyway, i send a patch that should fix this [1]. [1]: https://lore.proxmox.com/pve-devel/20250312083819.15379-1-s.sterz@proxmox.com/T/#u

can you post the full output of udevadm info -p /sys/block/nvme0n1 --query all? the regex in question does seem to expect there to be no leading spaces [1]. however, there also needs to be a leading E:, that is missing from your output too. [1]...

I am not sure why this can't be incorporated into a PBS setup, do you have a specific reason why you can't use a local PBS for that? Either a machine or even a VM could work. You could use a removable datastore, ifyou need to be able to detach the media sometimes [1]. [1]...

The installer complains about a checksum issue. This indicates that the installer was probably not downloaded correctly or the file was corrupted in some other way. You should try to download the ISO from here again [1]. And if it happens again, try to verify the ISO by following this guide [2]...

Hey, have you tried the opt-in 6.11 kernel [1]? [1]: https://forum.proxmox.com/threads/opt-in-linux-6-11-kernel-for-proxmox-ve-8-available-on-test-no-subscription.156818/

No, they won't be restarted automatically, that would be the point of HA. So if you don't use HA for your guests, they won't be restarted automatically. You still have a distributed file system with redundancy. So you can manually restart the VMs without starting the missing node again. It also...

Hey, yeah, the regex gets “anchored” by PMG. Meaning that it always adds a “^” at the beginning and a “$” at the end. “^” matches the beginning of a line and “$” matches the end of a line. So the regex you provide matches a string that starts with “@schimmel” followed by an arbitrary character...

Well if you want proper fault tolerance of a node, I'd highly recommend three nodes as a three node cluster can handle the loss of a single node much better. Both in-terms of the way that Corosync clustering works and also, because the HA resources can be distributed better. Think of it this...

Hallo, hast du den Upgrade Guide zu 8 gut durchgelesen und die Schritte dort befolgt [1]? Sonst fällt in dem Screenshot auf, dass das aktive Proxmox Repo noch das für bullseye, also Proxmox VE 7 ist. Bitte dieses deaktivieren und das für bookworm aktivieren. Sonst kann das Upgrade nicht...

Hi, since you are hosting Proxmox VE on top of Virtualbox, the most likely cause to me here, would be that Virtualbox does not pass through some necessary extension to PVE. Since the screenshot is in Russian, not sure if I can help you much further there, though.

If you are using the controller in HBA mode, you should be fine to use ZFS via the installer's advanced options [1]. [1]: https://pve.proxmox.com/pve-docs/pve-admin-guide.html#advanced_zfs_options

We do not package Proxmox VE for ARM based ISAs. You seem to be using Pimox [1] which is a community project that is not officially supported by us [1]. [1]: https://github.com/pimox/pimox7

Hi, do you mean nothing gets restored at all order files that are downloaded have a weird name? Could you provide an example of such a file name? Also, I am assuming that you are only talking about file restore. So when you restore the entire VM, files are till in place, right? Could you...

Hi, this is a feature of your browser that hides the full path from our JavaScript components. This improves privacy and security [1,2]. The most we can realistically do is strip the “C:\fakepath\” part to only display the filename. You could open an enhancement request over in the bugzilla...

That looks correct, those permissions are actually 755, you missed an “r” there (you can check with `stat /etc/apt/trusted.gpg.d`). The issue remains strange, though. I'm guessing a curl --silent http://download.proxmox.com/debian/dists/bullseye/InRelease | gpgv --keyring...

Hey, yes you'd need two subscription keys if you want the whole cluster to be covered.

Well, usually you wouldn't populate two sockets with odd numbered CPU cores in real-life either. The socket option has two main uses-cases (from my knowledge at least): Helping pass through a more sensible NUMA topology to improve performance. This requires the NUMA option to be set as well...

This means that the number of sockets you configure for the VM should match the number of NUMA nodes of the host system the VM is running on. You can use `lscpu | grep "NUMA node(" to figure out how many NUMA nodes your host system has. Typically, the number of NUMA nodes of the host system...

Interesting, did you also configure the bookworm repos? What does `apt policy` report in terms of configured repos? Other than our and the regular Debian keys, is there anything else configured there that might be relevant?

Do you mean in your VM config? Usually not much, if your host system has two sockets, though, it may benefit from that if you also activate NUMA [1]. [1]: https://pve.proxmox.com/pve-docs/pve-admin-guide.html#_numa