HPE Storeonce VTL

[ssevasta]

Hi
I created a new VTL on my storeonce and managed to add it to the proxmox backup server. However none of the tapes is visible in inventory tab and i have this error: error "reading element status: decode element status for MediumTransport on 0 failed - failed to fill whole buffer". Do i need to install any store once drivers on the backup server?

 

[dcsapak]

hi, what versions do you have?

proxmox-backup-manager versions --vebose

also can you post the output of the following commands?:

sg_raw -r 64k <PATH_TO_CHANGER_DEVICE> B8 11 00 00 ff ff 00 ff ff ff 00 00
sg_raw -r 64k <PATH_TO_CHANGER_DEVICE> B8 12 00 00 ff ff 00 ff ff ff 00 00
sg_raw -r 64k <PATH_TO_CHANGER_DEVICE> B8 13 00 00 ff ff 00 ff ff ff 00 00
sg_raw -r 64k <PATH_TO_CHANGER_DEVICE> B8 14 00 00 ff ff 00 ff ff ff 00 00

and replace <PATH_TO_CHANGER_DEVICE> with the path from /dev/tape/by-id that you get from

proxmox-tape changer list

 

[ssevasta]

Hi Dominik

see attached.

 

[ssevasta]

Hi Dominik, good morning, any updates re above issue please?

 

[dcsapak]

hi, sorry for the late response, i'll check it out soon (hopefully today)

 

[dcsapak]

hi,

i looked at it, and it seems the library returns some unexpected data (it sets the PVolTag flag for the medium transfer call, but does not include the volume tag then). Sadly i cannot find the scsi reference for that changer type (storeonce) so
i cannot say if this is expected behaviour for that type of changer or not. In any case, i sent a patch for our code to not treat this as an error, but only as a warning:

https://lists.proxmox.com/pipermail/pbs-devel/2024-August/010693.html

if that patch gets applied and you install the next bumped version after that, this error should not occur on your side anymore