Scanning tape labels in a robot

[longwave2]

Hello Forum,

What is the method to inventory a robot by only reading tape labels after retracting and inserting tapes?

We eject weekly tapes with:

# vmchange -res -w -verbose -multi_eject -rn 0 -rt tld -ml <tapes_to_extract>

The operator opens the CAP door, pulls out the tapes found and then puts in the tapes that were stored in the vault and closes the door.

Then we do a

# vmupdate -rt tld -rn 0 -empty_map

to insert the tapes in the library. But, when pulling out 2 tapes, for example 000018 and 000019 and then putting in 4 tapes, say 000020, 000021, 000022, 000023 the robot gets confused and doesn't show the right contents. It says the 000018 and 000019 are in and that the tapes 000022 and 000023 are out, which isn't true.

As I deduce it the robot 'thinks' it has the tapes in since it doesn't scan the tape labels when picking the tapes from the CAP.

So, the question is: how do I force the robot to scan its contents, preferrably when picking the tapes from the CAP, without going through the mount and scan of the tape header as the vmphyinv command would do?

Thanks.

 

[Qdog]

I think you're looking to perform an inventory. The following command lists the media contents of TLD robot 1 and the volume
configuration for that robot on the host named niagra, along with a list of any mismatches that are detected:
vmcheckxxx -rt tld -rn 1 -rh niagra -h niagra

You should be able to just inventory the robot. During the inventory, the robot scans the barcodes and updates the volDB if necessary. You could inject the tapes, but the inventory works better.