On Apr 2, 2015, at 2:23 PM, Cuttler, Brian (HEALTH) <[email protected]> wrote:
> When I've killed amcheck, or for that matter any command that exercises the > robot that I have to wait for the often disconnected MTX job to complete, > which can take several minutes, depending I suppose on the speed of the robot > and the command in progress. I'm sure most load commands run a lot longer > than the eject commands. Yeah, but isn’t it still faster than letting the robot cycled through ALL of its slots? Yes, it takes some time to kill the job, and some effort to put the pieces back to normal. And no, a summer student probably doesn’t have the privs to do this. Deb > > -----Original Message----- > From: [email protected] [mailto:[email protected]] On > Behalf Of Debra S Baddorf > Sent: Thursday, April 02, 2015 3:19 PM > To: Jason L Tibbitts III > Cc: Debra S Baddorf; AMANDA users > Subject: Re: Behavior when no tape in the library is free > > Ooooo. OK, next answer might have to be from an expert. > I'm guessing that amanda thinks "well, maybe he already DID put another > tape in a slot somewhere, and I just have to go look for it". > I *would* think it would scan the barcodes first, though. > > I DO think you can control-C an amcheck job (or KILL it if it's a cron > job). > I know I have control-C'd a lengthy tape scan. > Though it will have lost track of what the correct CURRENT slot is. > You'll have to tell it, with > amtape [config] slot N > after you've inserted your correct tapes. > You might have to add an mtx eject to eject it too, if amanda got > really confused. > > Experts? Comments on killing the amcheck, as well as why it reads all the > tapes > instead of scanning the barcodes? > Deb > > > On Apr 2, 2015, at 2:05 PM, Jason L Tibbitts III <[email protected]> wrote: > >>>>>>> "DSB" == Debra S Baddorf <[email protected]> writes: >> >> DSB> Are you using the "robot" type changer? >> >> Yes: >> >> define changer magnum { >> tpchanger "chg-robot:/dev/sg8" >> property "tape-device" "0=tape:/dev/nst0" >> device-property "LEOM" "TRUE" >> device-property "BLOCK_SIZE" "512k" >> changerfile "/var/lib/amanda/chg-robot-dev-sg8" >> } >> >> DSB> The barcode match-up to tape name gets stored in that changerfile >> DSB> (whatever name YOU put there; that's my file name) and it might >> DSB> not need to actually mount and scan the tapes each time. >> >> Yes, the barcodes are all there as well as in the tapelist, and >> they're all correct. Pretty easy since the amanda labels and the >> barcode labels are exactly the same. >> >> DSB> Can you perform the following command? >> DSB> amtape [config] inventory >> >> Yes, it works and runs quickly, reporting the tape labels and barcodes: >> >> AMANDA:backup2:~> time amtape math inventory slot 1: label C00023 >> barcode C00023 slot 2: label C00024 barcode C00024 slot 3: label >> C00025 barcode C00025 slot 4: label C00026 barcode C00026 slot 5: >> label C00027 barcode C00027 slot 6: label C00028 barcode C00028 slot >> 7: label C00029 barcode C00029 slot 8: label C00030 barcode C00030 >> slot 9: label C00031 barcode C00031 slot 10: label C00032 barcode >> C00032 slot 11: label C00033 barcode C00033 slot 12: label C00034 >> barcode C00034 slot 13: label C00035 barcode C00035 slot 14: label >> C00036 barcode C00036 slot 15: label C00037 barcode C00037 slot 16: >> label C00038 barcode C00038 slot 17: label C00039 barcode C00039 slot >> 18: label C00040 barcode C00040 slot 19: label C00041 barcode C00041 >> (in drive 0) (current) slot 20: label C00042 barcode C00042 slot 21: >> label C00043 barcode C00043 slot 22: label C00044 barcode C00044 slot >> 23: empty (import/export slot) amtape math inventory 0.18s user 0.03s >> system 9% cpu 2.051 total >> >> (the missing slot is the dedicated cleaning tape) >> >> - JM< > >
