After a lot of time and diagnostic work, it appears that my problem was
hardware-based. Since the drive was still under warranty, Quantum sent
me a new one, and everything seems to be working nicely since I
installed it. Between btape, taperx.lx (quantum utility), and just plain
tar, we were able to show that the drive was failing, regardless of tape
or application.
As always, I appreciate the help I see on this list.
Mike Morgan
Michael Morgan wrote:
> I seem to be having trouble writing large backups to tape from bacula.
> It has been working just fine for months, and within the last 2-3 weeks,
> I have started seeing errors and the backups will not complete.
>
> What happens it that it spools to disk, then begins to write to tape.
> After 3-9GB of data is written to tape, it seems to sense that the tape
> is full and give me the following error:
>
> Data spooling: 1 active jobs, 20,200,011,658 bytes; 1 total jobs,
> 20,200,011,658 max bytes/job.
> Attr spooling: 1 active jobs, 0 bytes; 1 total jobs, 4,245,075 max bytes.
> You have messages.
> *messages
> 19-Feb 11:01 longstreet-sd: User specified spool size reached.
> 19-Feb 11:01 longstreet-sd: Writing spooled data to Volume. Despooling
> 20,200,011,658 bytes ...
> 19-Feb 11:06 longstreet-sd: pickett.2007-02-19_10.21.59 Error:
> block.c:569 Write error at 9:3794 on device "LTO-2" (/dev/nst0).
> ERR=Input/output error.
> 19-Feb 11:06 longstreet-sd: pickett.2007-02-19_10.21.59 Error: Error
> writing final EOF to tape. This Volume may not be readable.
> dev.c:1687 ioctl MTWEOF error on "LTO-2" (/dev/nst0). ERR=Input/output
> error.
> 19-Feb 11:06 longstreet-sd: End of medium on Volume "weekly3-LogAn"
> Bytes=9,244,182,528 Blocks=143,293 at 19-Feb-2007 11:06.
> 19-Feb 11:09 longstreet-sd: Please mount Volume "Weekly2-Logan" on
> Storage Device "LTO-2" (/dev/nst0) for Job pickett.2007-02-19_10.21.59
>
> I am running bacula on a Fedora Core 6 system, Athlon 64 CPU, 1GB RAM,
> Certance LTO-2 1/2-height tape drive (single drive, no changer). I have
> searched back through the archives, but didn't see anything that seemed
> to apply to my situation or configuration. I did see one topic on
> permissions to the tape drive, but adjusting that didn't make any differece.
>
> I have run external diagnostics on the drive, using the btape test, and
> Quantum's diagnostic utility. All external tests pass with flying
> colors. I have done 180GB r/w tests without a hitch. I have tried
> different tapes, new tapes, old tapes, but whatever I do, I keep getting
> the same errors.
>
> I can, however, write small jobs to the tape without a problem. I can
> run a 1.2GB test backup without a hitch. I can even restore from
> multiple small backups on a tape. I only get into trouble when I try to
> run one of the main server backups (80-120GB/server).
>
> I've checked the server logs for SCSI errors or timeouts, and don't see
> anything significant. I did see one permissions error, and searching on
> that in the logs, found a potential solution. It didn't fix the problem.
>
> So, I have tried, and I am now stumped!
>
> SD config follows:
>
> Storage { # definition of myself
> Name = longstreet-sd
> SDPort = 9103 # Director's port
> WorkingDirectory = "/var/bacula"
> Pid Directory = "/var/run"
> Maximum Concurrent Jobs = 20
> }
>
> #
> # List Directors who are permitted to contact Storage daemon
> #
> Director {
> Name = longstreet-dir
> Password = "magic5"
> .
> .
> .
> Device {
> Name = LTO-2 #
> Media Type = LTO-2
> Archive Device = /dev/nst0
> AutomaticMount = yes; # when device opened, read it
> AlwaysOpen = yes;
> RemovableMedia = yes;
> RandomAccess = no;
> Maximum Spool Size = 20200000000
> Spool Directory = /export/backup2
> # Changer Command = "/etc/bacula/mtx-changer %c %o %S %a %d"
> # Changer Device = /dev/sg0
> # AutoChanger = yes
> # Enable the Alert command only if you have the mtx package loaded
> # Alert Command = "sh -c 'tapeinfo -f %c |grep TapeAlert|cat'"
> }
> .
> .
> .
>
>
> As I mentioned, this used to all work flawlessly, so I am at a loss as
> to what I can try next. We did upgrade the box from FC3 to FC6 within
> the last few months, but the backups were still running fine after the
> upgrade, this trouble only cropped up in the last few weeks.
>
> Any ideas?
>
> Thank you.
>
> Mike
--
Michael L. Morgan
Director of Operations
Iodynamics, LLC
(435) 760-1046
-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
