That worked! It So I basically need to wipe it and then readd to the bacula by issuing 'label barcodes'? And the problem was that it was corrupt?
JJ -----Original Message----- From: John Drescher [mailto:dresche...@gmail.com] Sent: Friday, July 16, 2010 4:04 PM To: Jeremiah D. Jester Cc: bacula-users Subject: Re: [Bacula-users] Please mount volume... On Fri, Jul 16, 2010 at 6:39 PM, John Drescher <dresche...@gmail.com> wrote: >> How did you determine it is corrupt? This? >> "Could not unserialize Volume" > > It could not find the signature for the tape header see below for my > explanation. > >> >> Here is the output for just KL0339. >> >> | 102 | KL0339 | Append | 1 | 64,512 | 0 | >> 15,552,000 | 1 | 2 | 1 | LTO2 | 0000-00-00 00:00:00 | >> > > Good. There was nothing on the volume anyways. > >> Not sure how to verify if /dev/nst0 is valid. This is how I originally >> determined to use /dev/st0. >> >> >> [r...@monk bacula]# dmesg | grep -B 2 -A 2 "tape" >> e1000: 0000:06:07.0: e1000_probe: (PCI:66MHz:32-bit) 00:11:43:e1:0f:a1 >> st: Version 20070203, fixed bufsize 32768, s/g segs 256 >> st 1:0:1:0: Attached scsi tape st0 >> st0: try direct i/o: yes (alignment 512 B) >> st 1:0:2:0: Attached scsi tape st1 >> st1: try direct i/o: yes (alignment 512 B) >> e1000: eth0: e1000_probe: Intel(R) PRO/1000 Network Connection >> >> Although, nst0 file does exist. >> >> >> [r...@monk bacula]# ls /dev/nst0 >> 0 /dev/nst0 >> > The difference between /dev/st0 and /dev/nst0 is that /dev/nst0 does > not rewind after operations while /dev/st0 does. With /dev/st0 there > is a slight chance of bacula corrupting tapes if the tape was rewound > and bacula expected the tape to be positioned at the end. > > Now looking at the list media result I do not think that is what > happened because list media said that bacula never wrote to this > volume. Even with that I still recommend on changing your config to > use /dev/nst0 instead. > > Is it possible that there was data on this volume when you first used > it with bacula? I think somehow the label barcodes failed because > there was data on the tape but it still put it in the database. > > BTW, I have to leave soon. Its 6:36 PM here and I am still at work.. I > will try to help you as much as you can before I go though. > I am leaving in less than 5 minutes so I doubt we will be able to continue this today. One thing to try first before I explain how to fix the corrupted tape is to do the following mt -f /dev/st0 rewind then ./bin/bls -c /etc/bacula/bacula-sd.conf -V* -L /dev/st0 to verify that the tape was at the beginning when bls looked for it. I believe bls does that anyways but it will not hurt to check. And yes in this case /dev/st0 is fine since we are not writing.. If you still get these errors. Then do the following. NOTE: Make sure you have the KL0339 tape still in the drive as this will erase the tape.. mt -f /dev/st0 rewind mt -f /dev/st0 weof mt -f /dev/st0 rewind then unload the tape to a slot. then start bacula-sd then use bconsole to delete KL0339 delete volume and follow the prompts then label barcodes and follow the prompts. Do not worry about already labeled volumes bacula will ignore them when it labels. John ------------------------------------------------------------------------------ This SF.net email is sponsored by Sprint What will you do first with EVO, the first 4G phone? Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
