Hi folks,
while our Bacula setup worked flawlessly for the last few months, last night's
backup (starting at 4:20am, to an HP LTO-2 drive, with the volume Sunday-0001
in the drive) didn't exhibit the usual behavior. Our system console shows an
entry
Oct 30 04:20:57 erebor st0: Error with sense data: <6>st0: Current: sense
key=0x3
Oct 30 04:20:57 erebor ASC=0x14 ASCQ=0x0
Oct 30 04:20:57 erebor Info fld=0x7fffff
Oct 30 04:22:47 erebor scsi0:0:4:0: Attempting to abort cmd ffff810042013d00:
0x12 0x0 0x0 0x0 0xff 0x0
Oct 30 04:22:47 erebor scsi0:0:4:0: Command not found
Oct 30 04:22:47 erebor st0: Error with sense data: <6>st0: Current: sense
key=0x3
Oct 30 04:22:47 erebor ASC=0x14 ASCQ=0x0
Oct 30 04:23:57 erebor scsi0:0:4:0: Attempting to abort cmd ffff810042013d00:
0x12 0x0 0x0 0x0 0xff 0x0
...
The corresponding Bacula job then began sending mails with the text
30-Oct 04:23 erebor-sd: Please mount Volume "Sunday-0002" on Storage Device
"LTO-2" for Job DefaultBackup.2005-10-30_04.05.00
Note that `Sunday-0001' at this point contains roughly 56GB of data, thus
there should be ample space on the media. I manually inserted Sunday-0002,
which allowed the backup to continue, leading to a final report mail
beginning with
30-Oct 04:05 erebor-dir: Start Backup JobId 38,
Job=DefaultBackup.2005-10-30_04.05.00
30-Oct 04:05 erebor-sd: Volume "Sunday-0001" previously written, moving to end
of data.
30-Oct 04:22 erebor-sd: DefaultBackup.2005-10-30_04.05.00 Error: Unable to
position to end of data on device "/dev/nst0". ERR=dev.c:494 ioctl MTEOM
error on /dev/nst0. ERR=Input/output error.
30-Oct 04:22 erebor-sd: Marking Volume "Sunday-0001" in Error in Catalog.
30-Oct 04:23 erebor-sd: Please mount Volume "Sunday-0002" on Storage Device
"LTO-2" for Job DefaultBackup.2005-10-30_04.05.00
30-Oct 05:23 erebor-sd: Please mount Volume "Sunday-0002" on Storage Device
"LTO-2" for Job DefaultBackup.2005-10-30_04.05.00
30-Oct 07:23 erebor-sd: Please mount Volume "Sunday-0002" on Storage Device
"LTO-2" for Job DefaultBackup.2005-10-30_04.05.00
30-Oct 11:23 erebor-sd: Please mount Volume "Sunday-0002" on Storage Device
"LTO-2" for Job DefaultBackup.2005-10-30_04.05.00
30-Oct 19:23 erebor-sd: Please mount Volume "Sunday-0002" on Storage Device
"LTO-2" for Job DefaultBackup.2005-10-30_04.05.00
31-Oct 10:06 erebor-sd: DefaultBackup.2005-10-30_04.05.00 Warning: Director
wanted Volume "Sunday-0002".
Current Volume "Sunday-0001" not acceptable because:
1998 Volume "Sunday-0001" status is Error, but should be Append, Purged or
Recycle.
31-Oct 10:06 erebor-sd: Please mount Volume "Sunday-0002" on Storage Device
"LTO-2" for Job
31-Oct 10:37 erebor-sd: Wrote label to prelabeled Volume "Sunday-0002" on
device "/dev/nst0"
31-Oct 10:37 erebor-sd: Spooling data ...
...
So, it appears we encountered our first actual tape error on volume
Sunday-0001. Now, I have a couple of questions:
1) From the Bacula mails, I gather that the error occured before data was
actually written to the tape. So we shouldn't have lost any data, right?
2) What do I do with the Sunday-0001 volume, now that it is marked with an
Error status? Should I remove it from the Sunday pool, and add a fresh volume
instead? Should I test the volume outside of Bacula (dd write/read?)
3) In general, how common are such media errors on LTO-2 tapes? This is the
first time we are using such a drive, so we don't have any experience with
it. In contrast, with our DLT drives, we never had a single media error over
the last 4 years.
Any help is appreciated,
Andreas Koch
-------------------------------------------------------
This SF.Net email is sponsored by the JBoss Inc.
Get Certified Today * Register for a JBoss Training Course
Free Certification Exam for All Training Attendees Through End of 2005
Visit http://www.jboss.com/services/certification for more information
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
