This is a discussion on Tapedrive not working... within the Gentoo Linux Support forums, part of the Unix Operating Systems category; --> Hi, A few days ago we completely reinstalled a server with a tape drive on and now the darn ...
| |||||||
| Register | FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
| ||||
| Hi, A few days ago we completely reinstalled a server with a tape drive on and now the darn tape refuses to make backups anymore. The driver is compiled into the kernel, and all checks run fine: # mt -f /dev/nst0 status SCSI 2 tape drive: File number=0, block number=0, partition=0. Tape block size 0 bytes. Density code 0x41 (DLT 40GB). Soft error count since last status=0 General status bits on (41010000): BOT ONLINE IM_REP_EN But if I want to put some dirs on tape with tar, I get this error: # tar cvf /dev/st0 /some/dir tar: Removing leading `/' from member names tar: /dev/st0: Cannot write: Input/output error tar: Error is not recoverable: exiting now Does anyone knows what might cause this? Maybe the fact that udev is used now (but I think we used it before when we upgraded the kernel)? I'll also see if we can extract something from a tape... Thanks, Wimmy -- +----------+ | PLEASE | | DO NOT | | FEED THE | | TROLLS | +----------+ | | | | .\|.||/.. |
| |||
| do you have the '-' in front of the 'c' option? -Walt Wim Cossement <wcosseme@nospam.bcol.be> wrote: > Hi, > > A few days ago we completely reinstalled a server with a tape drive on > and now the darn tape refuses to make backups anymore. > > The driver is compiled into the kernel, and all checks run fine: > > # mt -f /dev/nst0 status > SCSI 2 tape drive: > File number=0, block number=0, partition=0. > Tape block size 0 bytes. Density code 0x41 (DLT 40GB). > Soft error count since last status=0 > General status bits on (41010000): > BOT ONLINE IM_REP_EN > > But if I want to put some dirs on tape with tar, I get this error: > # tar cvf /dev/st0 /some/dir > tar: Removing leading `/' from member names > tar: /dev/st0: Cannot write: Input/output error > tar: Error is not recoverable: exiting now > > Does anyone knows what might cause this? > > Maybe the fact that udev is used now (but I think we used it before when > we upgraded the kernel)? > > I'll also see if we can extract something from a tape... > > Thanks, > > Wimmy > -- - Please email innkeepATncDOTrrDOTcom if interested or in need of assistance! |
| |||
| So I can still get files back that are on tape, but I cant write any new tapes (and no, the write protection is off)... The file permissions look OK: crw-rw---- 1 root tape 9, 0 Jun 22 09:02 /dev/st0 crw-rw---- 1 root tape 9, 128 Jun 22 09:02 /dev/nst0 lrwxrwxrwx 1 root root 9 Jun 22 09:02 /dev/tape -> /dev/nst0 So what could be wrong? Wimmy |
| |||
| by 'any new tapes' do you mean write entire new volume or diff additions to existing? Sorry for stating the obvious but I want to rule out you have tried to create 'a new tape'. -Walt Wim Cossement <wcosseme@nospam.bcol.be> wrote: > So I can still get files back that are on tape, but I cant write any new > tapes (and no, the write protection is off)... > > The file permissions look OK: > > crw-rw---- 1 root tape 9, 0 Jun 22 09:02 /dev/st0 > crw-rw---- 1 root tape 9, 128 Jun 22 09:02 /dev/nst0 > lrwxrwxrwx 1 root root 9 Jun 22 09:02 /dev/tape -> /dev/nst0 > > So what could be wrong? > > Wimmy -- - Please email innkeepATncDOTrrDOTcom if interested or in need of assistance! |
| |||
| > by 'any new tapes' do you mean write entire new volume > or diff additions to existing? > Sorry for stating the obvious but I want to rule out > you have tried to create 'a new tape'. > -Walt I used a brand new tape, so I just used mt rewind tar -cvvf /dev/nst0 /some/dir > log.txr 2> error.txt as a test. Wim |
| |||
| if the only error output is "tar: /dev/st0: Cannot write: Input/output error" I personally would try some more isolation tactic like additional tapes or drives. (of a known state) Forgive me for pointing to obvious stuff but its real easy to get something on the head or tape. And I personally would want to rule out all hardware before thinking some software problem cropped up. If the drive is external perhaps attach it elsewhere as a proof? again don't shoot the messenger I'm expermenting with using rsync on a disk space as I'm getting tired of my drives floating in and out of ability. -Walt > mt rewind > tar -cvvf /dev/nst0 /some/dir > log.txr 2> error.txt > > as a test. > > Wim |
| |||
| Well, someone told me dmesg gives some info (I was kinda wondering how some people could possibly get those kernel messages during bootin ;-) And it gives me this: st0: Error with sense data: <6>st0: Current: sense key=0xb ASC=0x47 ASCQ=0x0 Info fld=0x2800 Anywhere I can look what those codes mean? That same person suggested the kernel sources, so I'll take a looksee there. And I'm gonna put it on another machine to test as well. Wimmy PenguinsAnonymous@notAchance.org wrote: > if the only error output is > "tar: /dev/st0: Cannot write: Input/output error" > I personally would try some more isolation tactic like > additional tapes or drives. (of a known state) > Forgive me for pointing to obvious stuff but its real > easy to get something on the head or tape. > And I personally would want to rule out all hardware > before thinking some software problem cropped up. > If the drive is external perhaps attach it elsewhere as > a proof? > again don't shoot the messenger > I'm expermenting with using rsync on a disk space as > I'm getting tired of my drives floating in and out of > ability. > > -Walt /.. |
| ||||
| So anyway, it was like, 15:56 CEST Jun 23 2005, you know? Oh, and, yeah, Wim Cossement was all like, "Dude, > st0: Error with sense data: <6>st0: Current: sense key=0xb > ASC=0x47 ASCQ=0x0 > Info fld=0x2800 Checked your cables and proper termination of the chain lately? If that's not an expected message (unhandled timeouts from EOT or whatever), I'd start looking into potential hardware problems. -- Time flies like an arrow, fruit flies like a banana. Perth ---> * 16:13:17 up 32 days, 20:50, 10 users, load average: 3.22, 2.73, 2.40 Linux 2.6.11.10 x86_64 GNU/Linux Registered Linux user #261729 |