On Sat, 1 Dec 2007, Phil Stracchino wrote:

> Honestly, nscd has had a reputation for a long time of being horribly
> broken.  The best advice regarding running nscd is "Just don't."  Just
> about every admin I know disables nscd on sight.  There are much better
> tools for the job.

Well, to be true I don't run nscd. It is provided with glibc, but nscd 
isn't running, or isn't started as a daemon. I wasn't aware of it's 
reputation either, anyway. So, I don't know why there is a syscall in the 
log...

To answer your first question, I'm running kernel 2.6.11.10 SMP (on a Xeon 
with hyperthreading).

Something I've got in mind now as well, but I don't know if it can have 
anything to do with it: a couple of weeks ago, I've gotten the idea to use 
the 'make_catalog_backup' script, on another host, to make mysql 
unloads/backups. It's a huge database (around 30 GB), so it takes a while 
to unload. Bacula backups this dumpfile after the dump:

ClientRunBeforeJob = "/usr/local/bacula/etc/make_catalog_backup mydb 
myuser"
ClientRunAfterJob  = "/usr/local/bacula/etc/delete_catalog_backup mydb"

I don't know exactly if the problems started around the time I've 
implemented this, but it was the latest configuration change I made. So I 
disabled this command, and my (differential) backup on thursday and my 
(full) backup on friday already ran fine.

I'll keep you informed if this changes anything, as said, the crash 
happens only every few days/every week or so.

If it doesn't help, I can always try to upgrade my distribution/kernel.

Thanks!



-------------------------------------------------------------------------
SF.Net email is sponsored by: The Future of Linux Business White Paper
from Novell.  From the desktop to the data center, Linux is going
mainstream.  Let it simplify your IT future.
http://altfarm.mediaplex.com/ad/ck/8857-50307-18918-4
_______________________________________________
Bacula-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bacula-devel

Reply via email to