invalid arguments "read error" from dump(8)

[Vivek Khera]

Near the end of a rather long level 0 dump to a DLT drive today, I got
these errors spit at me.  /dev/aacd0s2f is a RAID10 config of 4 36Gb
drives for a 69,577,710 K-byte file system.  It is an active partition
storing a postgresql database that is being updated 24x7.

Is this just due to the activity on the FS or am I potentially looking
at some hardware issues arising?

The tape host is running 4.8-RELEASE and the server being dumped is
4.7-RELEASE-p3.

  DUMP: read error from /dev/aacd0s2f: Invalid argument: [sector -2126184086]: count=512
  DUMP: read error from /dev/aacd0s2f: Invalid argument: [sector -2126184085]: count=512
  DUMP: read error from /dev/aacd0s2f: Invalid argument: [sector -2126184084]: count=512
  DUMP: read error from /dev/aacd0s2f: Invalid argument: [sector -2126184083]: count=512
  DUMP: read error from /dev/aacd0s2f: Invalid argument: [sector -2126184082]: count=512
  DUMP: read error from /dev/aacd0s2f: Invalid argument: [sector -2126184081]: count=512


and in /var/log/messages I got these corresponding:


May 13 16:12:36 d01 /kernel: dscheck(#aacd/0x30005): negative b_blkno -2126184086
May 13 16:12:36 d01 /kernel: dscheck(#aacd/0x30005): negative b_blkno -2126184085
May 13 16:12:36 d01 /kernel: dscheck(#aacd/0x30005): negative b_blkno -2126184084
May 13 16:12:36 d01 /kernel: dscheck(#aacd/0x30005): negative b_blkno -2126184083
May 13 16:12:36 d01 /kernel: dscheck(#aacd/0x30005): negative b_blkno -2126184082
May 13 16:12:36 d01 /kernel: dscheck(#aacd/0x30005): negative b_blkno -2126184081


there were a bunch before these last few, not all contiguous, but in
small groups of contiguous b_blkno.

-- 
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Vivek Khera, Ph.D.                Khera Communications, Inc.
Internet: khera at kciLink.com       Rockville, MD       +1-240-453-8497
AIM: vivekkhera Y!: vivek_khera   http://www.khera.org/~vivek/