/dev/tape/by-id/scsi-10WT044700-sg failed - decode mode sense failed - wrong mode_data_len

[FelixJ]

Hi, I want to try out the new tech-preview on the PBS Tapebackup - so far I did not come very far, as I get the above message, when I try to label a tape.

This is my tapedrive (tapeinfo -f /dev/st0):

Product Type: Tape Drive
Vendor ID: 'IBM '
Product ID: 'ULTRIUM-HH7 '
Revision: 'G9Q1'
Attached Changer API: No
SerialNumber: '10WT044700'
MinBlock: 1
MaxBlock: 8388608
SCSI ID: 0
SCSI LUN: 0
Ready: yes
BufferedMode: yes
Medium Type: 0x78
Density Code: 0x5c
BlockSize: 0
DataCompEnabled: yes
DataCompCapable: yes
DataDeCompEnabled: yes
CompType: 0xff
DeCompType: 0xff
BOP: yes
Block Position: 0
Partition 0 Remaining Kbytes: -1
Partition 0 Size in Kbytes: -1
ActivePartition: 0
EarlyWarningSize: 0
NumPartitions: 0
MaxPartitions: 3

lsscsi -v:
[1:0:0:0] tape IBM ULTRIUM-HH7 G9Q1 /dev/st0
dir: /sys/bus/scsi/devices/1:0:0:0 [/sys/devices/pci0000:00/0000:00:1c.0/0000:0a:00.0/host1/port-1:0/end_device-1:0/target1:0:0/1:0:0:0]

What is the problem?
Thanks, Felix

 

[dcsapak]

can you post the complete task log ? (to see where exactly the error occurs)

edit: also can you post the output of

sg_raw  -r 64k <PATH_TO_DRIVE> 5A 08 0F 00 00 00 00 10 00 00

(best attach as a file)

 

[FelixJ]

Good Morning,
sorry for the delay, an other backup was in progress. Attached are the screenshot from the task as well as the output from sg_raw.
regards,
Felix

 

[FelixJ]

Ok, I just realized, the above sg_raw.out.gz was made without tape loaded. So attached to this post is now the sg_raw output with tape loaded.

 

[dcsapak]

thanks for the output, it's clear now why it fails (the drive returns more data than we expect)
i'll send a patch

 

[FelixJ]

No doubt! my tapedrive... I'm also chatty! I'll wait for the patch.

 

[dcsapak]

fyi: the patch is here https://lists.proxmox.com/pipermail/pbs-devel/2021-April/002904.html
it will probably included (pending review) in one of the upcoming pbs updates

 

[FelixJ]

could you compile the patch for me, so I can try it right a way?

 

[dcsapak]

could you compile the patch for me, so I can try it right a way?

mhmm.. we do not have an automated build system, but the patch was applied and will be in proxmox-backup-server 1.1.5-1 (probably soon on pbs-test repository)

 

[FelixJ]

What is the estimated time-span from the testing to the "production" repo?
Related question: Do you rely on /etc/stinit.def to initialize the tape-drive correctly or do you use an pbs internal way to initialize the drive?

 

[tom]

What is the estimated time-span from the testing to the "production" repo?

Just to note, Tape Backup is a technical preview feature, not meant for production use.

 

[FelixJ]

Just to note, Tape Backup is a technical preview feature, not meant for production use.

I know, but for the sake of the product, one has to test it! And what better way to test it from the prospective, as if it would be already enterprise-ready... then one can say, what's still missing!

 

[dcsapak]

just fyi, the package is available on the pbs-test repository

 

[dcsapak]

Related question: Do you rely on /etc/stinit.def to initialize the tape-drive correctly or do you use an pbs internal way to initialize the drive?

no we do not, we initialize the drives ourselves