amiaopensource
optimize naming of mountpoint
Currently mountlto uses /Volumes/${BARCODE} as a default mount point but if it can not determine that then it uses /Volumes/${DECK_SERIAL}. The process of getting these values involves running an lfts command without a mountpoint, which the intent that it fails but provides deck and tape metadata along the way to use in a second correct lfts command to do the mount.
This process requires a lot of time since it has to wait for an ltfs command to fail until running it correctly. I've been considering have the mountpoint always be /Volumes/${DECK_SERIAL} and to skip the first ltfs command (the one that gathers data) and do the second one directly. However I'm finding that it's not easy to make a quick request of the attached LTO deck's serial.
So far I have this command:
ioreg -c IOSCSITargetDevice | grep "INQUIRY Unit Serial Number" | cut -d"=" -f2
This is close but grabs a bunch of other serial numbers that aren't needed. Any better way to gather some form of deck specific identifier?
We do work internally with the tape ID which for us is the same than the barcode, and for some of our clients the same than the six first chars of the barcode.
We have some setups with LTO-5 Tandberg decks running http://www.tandbergdata.com/default/assets/File/Downloads/ltfs300/LTFS_3.0.0_Mac_OEM.zip and with this version of the utility there appears to be no means to query the barcode/tape_id which is why we fall back to the deck serial.
Thank you! I’ll check tomorrow or on Friday this – and how we do it. We have a Tandberg LTO-5 deck.
OK, I'd be curious what utility source you use with it and if ltfs -o devname=0 on it gives you back a tape id.
The utility is derived from HP’s. It works for all decks we have in production (except Tandberg’s 24-slot which crashes when mixing up LTO-4 cartridges with LTO-5 and/or LTO-6, i.e. LTFS with pre-LTFS proprietary solutions).