On Sat, Jul 28, 2007 at 09:54:01AM -0400, Greg Troxel wrote:
> During last night's backups, I got tons of errors in /var/log/messages,
> enough to run /var out of space.  The following is a brief sample; there
> are about 400 messages (5 lines each) per second.
> 
> Has anyone seen this?  It seems like a combination of a hardware error
> and amanda being overly aggressive.
> 
> 
> Jul 28 09:00:01 fnord newsyslog[8358]: log file turned over
> Jul 28 09:00:01 fnord syslogd: restart
> Jul 28 09:00:01 fnord /netbsd: st0(ahc2:0:5:0):  Check Condition on CDB: 0x08 00 00 80 00 00
> Jul 28 09:00:01 fnord /netbsd: SENSE KEY:  Blank Check
> Jul 28 09:00:01 fnord /netbsd: INFO FIELD:  32768
> Jul 28 09:00:01 fnord /netbsd: ASC/ASCQ:  End-Of-Data Detected

Sense "Blank Check" means "End of data encountered on a read operation".
Amanda tries to read (CDB 08) 32768 bytes from either the end of a tape file
or a blank tape.
If there is a hardware error it must be before these messages.

-- 
Juergen Hannken-Illjes - hannken@eis.cs.tu-bs.de - TU Braunschweig (Germany)