Search results

You can use the Proxmox Offline Mirror [1] to provide an offline repository for your host. Then you can set up Ceph as you usually would via the GUI or the pveceph command by selecting the manual repository. [1]: https://pom.proxmox.com/

Given that Proxmox VE 9.1 provides support for OCI images as application containers, you can now run Docker containers through Proxmox VE directly. They get more or less “translated” to LXC containers. This feature is still a tech preview, though.

The currently planned feature set is available in the roadmap [1]. It is subject to change. If you find something missing, you can open a feature request over in our Bugzilla [2]. [1]: https://pve.proxmox.com/wiki/Proxmox_Datacenter_Manager_Roadmap [2]: https://bugzilla.proxmox.com/

Generally this causes issues between our use of Apparmor and other parts of our code base over and over again. So we heavily discourage it. However, with the release of Proxmox VE 9.1 you can use OCI templates for application containers on Proxmox VE. This means that you can run Docker...

Yes, sorry. Yep, looks OK I'd say, the patch my colleague send will fix the issue in question, and you should be good to go :)

I think this is just an oversight on our end. To be sure it'd be useful if you could run the follow command: openssl x509 -in /etc/pve/local/pve-ssl.pem -noout -text | grep "Algorithm" and post it's output here. Context: The security level of a signature key depends on the amount of entropy it...

These two logs are fine, it would have been interesting to see whether there are any authentication related issues here, but it doesn't look like it. I think at this point logs will not be super helpful as we don't log contents of the requests (for good reason) and tend to return 200 error cases...

Hm I'd need to look into this further, but my current guess would be that PVE added a parameter in certain API endpoints that PDM doesn't respect yet. Can you check the logs of your PVE and PDM hosts to get some more insight into what might be happening here and which requests exactly are...

Since nobody picked up the issue, it is still not implemented and nobody is currently working on it. If you feel like this should be more important, add a comment to the Bugzilla issue. Thank you.

We do not support Docker containers on top of LXC containers. If you want to use a Docker container, you should do so in a VM. Hence, we probably won't add such a feature any time soon as it would just encourage a use case we want to avoid.

Most of the cluster requirements are there for a reason. However, you might be interested in PDM [1] if you want a common management plane for multiple PVE instances without clustering them. [1]: https://forum.proxmox.com/threads/proxmox-datacenter-manager-0-9-beta-released.171742/

There are some ISOs on or download host for older releases still [1]. But beware that everything but Proxmox VE 8 & 9 is now EOL and receives no support. [1]: https://enterprise.proxmox.com/iso/

Hm ok da sollte das eigentlich schon funktionieren, außer es wurde absichtlich in den Einstellungen abgedreht oder so. Die Desktop UI kann auch auf das mobile UI umgestellt werden in dem eins oben rechts auf den aktuellen Username klickt und im Menü, dass dann ausklappen sollte "Switch to...

Das neue mobile Front-End verwendet unser neues Rust-basiertes Front-End. Wenn der mobile Browser Web Assembly (WASM) nicht unterstützt, dann kann die UI nicht angezeigt werden. Der Simulator scheint da nicht korrekt das neue Front-End zu laden.

Do you have a proxy in between you and the host? Also could you post the output of the following command openssl s_client -connect '45.84.67.184:443'? Ideally in a block? Thanks!

Das sollte auch schon mit PVE 8 gehen, allerdings empfehlen wir im Upgrade Guide [1], diese erst nachträglich zu modernisieren. Hauptsächlich weil der entsprechende apt Subcommand erst mit Debian 13 Trixie verfügbar ist und daher unter PVE 8 die Änderungen manuell durchzuführen wären. [1]...

Jup dann war /etc/pve an der stelle nicht korrekt am Laufen bzw. evtl. nur read-only gemountet. In dem Fall ist ein Login über unsere API nicht möglich, über PAM (CLI) aber schon. Das lässt sich an der Stelle nur schwer verhindern, wenn TFA verwendet werden soll.

Das Journal speichert das :) Also, wenn der Zeitraum noch ca. bekannt ist einfach dorthin scrollen oder z.B. mit --from und --since einschränken. Ja so weit ich das gerade im Kopf hab, sollten alle relevanten Config-Files hier unter /etc/pve liegen. Das sollte bei einem funktionalen Cluster auf...

Hallo, gibt es eventuell mehr Infos zu den gescheiterten Logins im Journal (journalctl)? Außerdem wäre es praktisch zu wissen welche Cookies gesetzt waren, wenn das noch nachvollziehbar ist.

Proxmox Backup Server needs at least relatime for the garbage collection to work correctly (see [1]). So yes, the atime property is needed, but for performance reasons only relatime is used (not standard atime). From man 7 zfsprops: [1]: https://pbs.proxmox.com/docs/maintenance.html#gc-background