Search results

from the dmesg output it looks like the nvidia driver is loaded, can you try to blacklist that and reboot? after a reboot, before the vm start 'lspci -nnk' should say 'vfio-pci' for the used driver (you can also post the 'lspci -nnk' output here if you want)

not sure it's related to the issue, but your screenshot shows that rom-bar is checked, but the error message indicates that it's unchecked ",rombar=0" did you try both? the full vmconfig and the output of dmesg would be interesting: qm config ID dmesg

it's not available by browsing the datastore, and arbitrary paths will probably not be implemented for some time, but the storage can be configured to use different paths than the default one with the 'content-dirs' option (cli only at the moment), see...

sorry that i just now saw the thread, but this seems to be normal behavior for arc: https://www.intel.com/content/www/us/en/support/articles/000094587/graphics.html

yes i assumed as much, thus my question, would it be enough for your use case to use the existing oidc integration ? you can e.g. set up a keycloak instance that uses your ldap as backend and is just used for oidc

ah yes, sorry. I forgot that PMG does not allow using pam users at all (except for root@pam). That this is available in the UI is a bug introduced by the recent addition of OIDC realms. (i created https://bugzilla.proxmox.com/show_bug.cgi?id=6488 for it) would adding the realm as OIDC with...

ah ja das `proxmox-tape` command kann das (noch) nicht (ein feature request hierzu wäre auch super :) ) es gibt aber immer noch as 'proxmox-backup-debug api' tool um das api generell anzusprechen, der entsprechende aufruf wäre dann: proxmox-backup-debug api create /tape/media/move --uuid...

Hi, are you sure the user you're logged in with has enough permissions to create users?

just to clarify, there is an ai enterprise driver that supports the a30 with the "C-Series' vgpus this was previously integrated in the normal vgpu driver, but separated out to the ai enterprise software, for licensing reasons i guess. see also...

nur zur info: in einem Wechsler werden zuerst die bänder verwendet die sich im laufwerk befinden (da sie 'on-site' sind) wenn ich nochmal drüber nachdenke gibt es sogar noch einen weg: die bänder die man nicht verwenden will, kann man als off-site markieren (tape backup -> inventory -> set...

i doubt it's the memory as an A16 will present itself as 4 gpus (as seen in your lspci output), but each of them has just 16G so 4*16 = 64 as a whole i guess this is the problem. I'd try scouting the BIOS menus for options that might fix this or upgrading the bios/efi to the newest versions

ein grund ist damit die retention/allocation policy sinnvoll eingehalten werden kann. je nach dem wie diese policies aussehen kann es zb nicht immer offensichtlich sein welches tape gerade frei ist wenn dann ein tape eingelegt ist welches nicht beschrieben werden darf muss erst wieder ein...

das ist nicht vorgesehen. Im moment wird die reihenfolge der tapes automatisch bestimmt, aber wie bereits gesehen, sagt das web ui bereits vorher welches tape erwartet wird

Hi, please write in English since this is an english forum. Otherwise most people won't be able to help you

no, not with built in methods. the backup works by starting the qemu process, but without actually running the guest. then we can backup the data via qemu and stop the process again. This is done to guarantee consistency and to be able to start the vm during backup (which wouldn't be possible...

i'm not super clear what exactly you are trying to do. Are you trying to connet via your own novnc installation? if yes, did you also set a valid cookie in the browser for the connection? the websocket connection needs a valid cookie set + the vncticket

this seems like it one these things * broken windows iso * broken storage * too small disk (it can of course be a different problem, best would be to post the vm config (qm config ID))

it seems the page cannot load any images, you're also missing the 'x' on the top right of the import window... please check the browser developer tools again, namely the 'network' tab to see if any image loading requests fail are you maybe behind some kind of proxy/reverse proxy?

can you post the output of dmesg nvidia-smi vgpu cat /sys/bus/pci/devices/0000:43:01.2/nvidia/current_vgpu_type cat /sys/bus/pci/devices/0000:43:01.2/nvidia/creatable_vgpu_types ? is there anything sticking out in the journal/syslog on boot with the nvidia cards? if you don't include the...

thanks for noticing! i sent a patch to fix it: https://lore.proxmox.com/pdm-devel/20250616135804.3212548-1-d.csapak@proxmox.com/T/#u