On Fri, Apr 05, 2002 at 03:05:28AM +0200, Bernd Eckenfels wrote:
> Perhaps it is just a dcache problem. Perhaps you can see anything in
> /proc/slabinfo or you could try to run find first? Could be that memory is
What do the columns in slabinfo mean?
> filled up and therefore the cache for the directory entries does not grow as
> needed. Are you sure switching off ldap helps in all cases? do you run nscd?
[504] [snoopy:unstable:bam] ~ >free
total used free shared buffers
cached
Mem: 126340 122432 3908 0 37632
34996
-/+ buffers/cache: 49804 76536
Swap: 390456 83860 306596
snoopy:unstable:~# /etc/init.d/nscd stop
Stopping Name Service Cache Daemon: nscd.
snoopy:unstable:~# /etc/init.d/nscd start
Starting Name Service Cache Daemon: nscd.
snoopy:unstable:~# logout
[506] [snoopy:unstable:bam] ~ >free
total used free shared buffers
cached
Mem: 126340 121544 4796 0 31288
36328
-/+ buffers/cache: 53928 72412
Swap: 390456 83860 306596
now the problem has gone away... Interesting.
Might be something to do with nscd not caching properly, maybe?
I will check the memory usage of nscd tommorrow, for now:
[507] [snoopy:unstable:bam] ~ >ps auwx | grep nscd
root 16854 0.0 1.4 14720 1776 ? S 12:13 0:00 /usr/sbin/nscd
root 16855 0.0 1.4 14720 1776 ? S 12:13 0:00 /usr/sbin/nscd
root 16856 0.0 1.4 14720 1776 ? S 12:13 0:00 /usr/sbin/nscd
root 16857 0.0 1.4 14720 1776 ? S 12:13 0:00 /usr/sbin/nscd
root 16858 0.0 1.4 14720 1776 ? S 12:13 0:00 /usr/sbin/nscd
root 16859 0.0 1.4 14720 1776 ? S 12:13 0:00 /usr/sbin/nscd
root 16860 0.0 1.4 14720 1776 ? S 12:13 0:00 /usr/sbin/nscd
bam 16923 0.0 0.4 1616 584 pts/1 S 12:16 0:00 grep nscd
--
Brian May <[EMAIL PROTECTED]>
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
