[PLUG] Backup Error: Need Advice

[Rich Shepard]

   After running for years without problems I've hit what may be a hardware
(RAM) problem that's affecting my backups. I'd greatly appreciate some help
in figuring out the source of this error.

   When I checked mail this morning (after the first incremental backup
following a successful full backup last Friday night), I had the following
e-mail message from root:

>From root at appl-ecosys.com Tue Mar 21 06:02:46 2006
Date: Tue, 21 Mar 2006 06:02:46 -0800
From: root at appl-ecosys.com
To: rshepard at appl-ecosys.com

Tue Mar 21 06:00:55 PST 2006 - Started BRU Level 1 Backup\n\tcommand = 'bru
-c -vvZ -n /etc/BRULEVEL0 -f /dev/st0 /'
Tue Mar 21 06:02:46 PST 2006 - ERROR IN BRU BACKUP, check log file:
'/tmp/brulog'

   I looked at the log and it just ended at this line:

bru: [I245] "/usr/lib/qt-3.3.4/examples/textedit/textedit.h": skipped file,
excluded by -n flag

   The log file size is 18.3M, so there's no point in mailling it to anyone.
Now the error message in the second e-mail from root was:

/bin/incbru: line 119: 18733 Segmentation fault      $BRUCMD >$LOGFILE 2>&1
incbru: ERROR in BRU Backup

   Line 119 is the end of an IF statement; the instruction to write to the
logfile is actuallly about line 113, but that's immaterial.

   I tried running the backup with a manual start rather than the cron job,
and it halted at the same place.

   Both yesterday morning (when I went to log in after being gone for several
days), and this morning when I went to log in, I saw the results of the seg
fault dump on the console. Here is probably what I saw:

[root at salmo ~]# tail /var/log/messages
Mar 21 06:02:27 salmo kernel: ds: 0018   es: 0018   ss: 0018
Mar 21 06:02:27 salmo kernel: Process bru (pid: 18733, stackpage=e76eb000)
Mar 21 06:02:27 salmo kernel: Stack: f3b54890 c19ac4d0 f6d62900 00000060
00000060 00000008 00000001 00000060 
Mar 21 06:02:27 salmo kernel:        e76ea000 0807fa60 c0108a93 0807fa60
0807ddc0 00000000 0807fa60 0807ff40 
Mar 21 06:02:27 salmo kernel:        bfffea54 000000c4 0000002b 0000002b
000000c4 400d9053 00000023 00000202 
Mar 21 06:02:27 salmo kernel: Call Trace:    [<c0108a93>]
Mar 21 06:02:27 salmo kernel: 
Mar 21 06:02:27 salmo kernel: Code: 8b 40 34 85 c0 75 34 31 c0 85 c0 89 c3 74
14 56 e8 de 26 00 
Mar 21 06:08:05 salmo inetd[11546]: auth/tcp (2): bind: Address already in
use
Mar 21 06:18:05 salmo inetd[11546]: auth/tcp (2): bind: Address already in
use

   Is this a bad memory stick? This is on my main workstation/server so I do
need to find and fix the problem ASAP.

TIA,

Rich

-- 
Richard B. Shepard, Ph.D.             |  Quantifying subjectivity for the
Applied Ecosystem Services, Inc.(TM)  |  benefit of business and society.
<http://www.appl-ecosys.com>     Voice: 503-667-4517      Fax: 503-667-8863