Solaris

Greetings.
i have solaris 9.
I had a problem with root file system day before yesterday when it got 100% full, i reduced it to 90% but then after one day it was automatically reduced to 40%!!!
no human intervention was done.
no reboot done.
Last time when i checked in proc FS there were a lot of files with linkcount 0 in proc FS as shown below.
Also can any1 please tell me how files with linkcount 0 are handled in solaris?
where can i find logs of root FS usage in time?
in /var/adm/messages i could only find alarms for FS Full not in percentage.


10993 52 -r--r--r-- 0 root root 52383 Sep 23 02:09 /proc/5062/fd/16
10993 52 -r--r--r-- 0 root root 52383 Sep 23 02:09 /proc/5055/fd/16
10993 52 -r--r--r-- 0 root root 52383 Sep 23 02:09 /proc/5038/fd/16
10993 52 -r--r--r-- 0 root root 52383 Sep 23 02:09 /proc/5057/fd/16
10993 52 -r--r--r-- 0 root root 52383 Sep 23 02:09 /proc/5063/fd/16
10993 52 -r--r--r-- 0 root root 52383 Sep 23 02:09 /proc/5058/fd/16
10993 52 -r--r--r-- 0 root root 52383 Sep 23 02:09 /proc/5053/fd/16
10993 52 -r--r--r-- 0 root root 52383 Sep 23 02:09 /proc/5056/fd/16
10993 52 -r--r--r-- 0 root root 52383 Sep 23 02:09 /proc/5061/fd/16
10993 52 -r--r--r-- 0 root root 52383 Sep 23 02:09 /proc/5064/fd/16
10993 52 -r--r--r-- 0 root root 52383 Sep 23 02:09 /proc/5082/fd/16
756219787 0 -rw------- 0 root dba 0 Sep 23 02:31 /proc/8246/fd/14
14621 0 --w------- 0 root dba 0 Sep 23 02:31 /proc/8246/fd/15
146024 1148608 --w------- 0 rtp99 dba 1175584768 Sep 25 05:02 /proc/13838/fd/6
146026 2947608 --w------- 0 rtp99 dba 3016867840 Sep 25 05:02 /proc/13838/fd/7
146024 1148608 --w------- 0 rtp99 dba 1175584768 Sep 25 05:02 /proc/13838/fd/8
146026 2947608 --w------- 0 rtp99 dba 3016867840 Sep 25 05:02 /proc/13838/fd/9
146024 1148608 --w------- 0 rtp99 dba 1175584768 Sep 25 05:02 /proc/13838/fd/10
146024 1148608 --w------- 0 rtp99 dba 1175584768 Sep 25 05:02 /proc/13838/fd/11
756219787 0 -rw------- 0 root dba 0 Sep 23 02:31 /proc/8244/fd/14
14621 0 --w------- 0 root dba 0 Sep 23 02:31 /proc/8244/fd/15
756219787 0 -rw------- 0 root dba 0 Sep 23 02:31 /proc/8226/fd/14
14621 0 --w------- 0 root dba 0 Sep 23 02:31 /proc/8226/fd/15
53342 5 -rw------- 0 rtp99 dba 4100 Sep 24 17:25 /proc/7375/fd/63
756219787 0 -rw------- 0 root dba 0 Sep 23 02:31 /proc/8247/fd/14
14621 0 --w------- 0 root dba 0 Sep 23 02:31 /proc/8247/fd/15
756219787 0 -rw------- 0 root dba 0 Sep 23 02:31 /proc/8243/fd/14
14621 0 --w------- 0 root dba 0 Sep 23 02:31 /proc/8243/fd/15
756219787 0 -rw------- 0 root dba 0 Sep 23 02:31 /proc/8245/fd/14
14621 0 --w------- 0 root dba 0 Sep 23 02:31 /proc/8245/fd/15
16466 5 -rw------- 0 oracle dba 4698 Sep 24 19:06 /proc/1451/fd/63
403590310 0 -rw------- 0 rtp99 dba 0 Sep 24 17:27 /proc/13832/fd/61
756219787 0 -rw------- 0 root dba 0 Sep 23 02:31 /proc/19837/fd/14
14621 0 --w------- 0 root dba 0 Sep 23 02:31 /proc/19837/fd/15


Thanks in Advance

---------- Post updated at 02:22 AM ---------- Previous update was at 01:41 AM ----------

Any1 please?
I have checked FS full alarm stopped coming exactly on midnight i.e. 00:00:00
Can there be any script for house keeping? how can i check?

Is /var on the same filesystem ? There are many syslog and other log files that are automatically rolled over dependant upon the date. For example /var/log/messages.* are rolled over and compressed.

Hmm, divine intervention ?
Like any other Unix OS. Unlinked files disappear only after they are closed.
There is no such log by default.
Likely.
Have a look at the crontabs.

Thanks for reply