SUMMARY: vdump errors

From: Jim Zimmerman (jzimmerman@malyswest.com)
Date: Fri May 07 2004 - 18:35:19 EDT


Thanks to Dr. Thomas Blinn. Moving tmp has been on the agenda. I was
just surprised to have the errors appear when it had not been an issue
before. I have included my original question below, and a summarization
below that.

Thank you,

Jim

-----------------------------------

Original question:

I have an ES40 running 5.1A PK3. I am getting these kinds of errors
every once in while when running vdump on the root partition. I was
under the impression that this was not supposed to happen anymore
with vdump and temporary files. I never used to get these types of
errors. They seem to be occurring more frequently. HP told me I need
to run verify against the root domain. However, I would like to resolve
this without having to bring the system down.

Thank you,

Jim Zimmerman
jzimmerman@malyswest.com

Backing up / ... [Fri May 7 04:06:10 PDT 2004]
  Usage Stats (KB): ADVFS:113249 On Tape:0 Needed:113249
  Tape file position: 3
path : /
dev/fset : root_domain#root
type : advfs
advfs id : 0x3e34c1d6.000e5b40.1
Mon May 3 04:08:16 2004
vdump: Date of last level 7 dump: Mon May 3 04:08:16 2004
vdump: Dumping directories
vdump: Dumping 4700040 bytes, 162 directories, 69 files
vdump: Dumping regular files
vdump: unable to get info for file
<./cluster/members/member0/tmp/croutZEA0iVbJb>; [2] No such file or
directory
vdump: unable to get info for file
<./cluster/members/member0/tmp/ckhosts.lock>; [2] No such file or
directory

vdump: Status at Fri May 7 04:06:17 2004
vdump: Dumped 4704166 of 4700040 bytes; 100.1% completed
vdump: Dumped 162 of 162 directories; 100.0% completed
vdump: Dumped 67 of 69 files; 97.1% completed
vdump: Dump completed at Fri May 7 04:06:17 2004
  Dump of / to /dev/ntape/tape0_d1 failed (RC=1)
        UnixFsDump failed
  Attempting to restart failed filesystem with next tape
Loading tape from slot 4

-----------------------------------

Summarization of response:

Dr. Blinn stated that it is extremely unlikely that there is any file
corruption based the error I am getting. He suggested migrating
/cluster/members/member0/tmp to a different ADVfs file domain, and
mounting it in the fstab to /cluster/members/member0/tmp. In addition,
create a different backup strategy for /tmp, if a backup is necessary,
and removing /tmp from the regular backup. He also suggested migrating
the /var/tmp directory off of the var_domain.



This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 10:49:58 EDT