Solaris

Hi there,
Root filesystem is above threshold, I have search and cleared unwanted files which are filling up space. But the root fs is still above threshold.
I don't know about veritas volume management. Can anyone show me how to solve this. Du shows /proc is occupying a lot of space. Most of the process ids are oracle processes. Can anyone provide assistance?

# df -h
Filesystem size used avail capacity Mounted on
/dev/vx/dsk/bootdg/rootvol
7.9G 7.0G 795M 91% /
/platform/sun4u-us3/lib/libc_psr/libc_psr_hwcap1.so.1
7.9G 7.0G 795M 91% /platform/sun4u-us3/lib/libc_psr.so.1
/platform/sun4u-us3/lib/sparcv9/libc_psr/libc_psr_hwcap1.so.1
7.9G 7.0G 795M 91% /platform/sun4u-us3/lib/sparcv9/libc_psr.so.1
fd 0K 0K 0K 0% /dev/fd
/dev/vx/dsk/bootdg/var
4.9G 808M 4.1G 17% /var
...
/dev/vx/dsk/bootdg/u01
4.9G 3.8G 1.1G 77% /u01
/dev/vx/dsk/bootdg/IN
40G 3.2G 37G 9% /IN
/dev/vx/dsk/ora-dg/ora_data
60G 25G 35G 42% /data

# du -sk * | sort -nr | head
219843482 proc
25790376 data
6235549 usr
3928171 u01
3346203 IN
829981 opt
822167 var
113685 etc
80160 tmp
72401 kernel

# cd /proc
# du -sk * | sort -nr | head
27066677 1486
25421117 1484
25382765 1488
25373110 1490
23289381 12493
14171990 1494
9867208 6770
9867208 6769
5975724 7408
4129901 1768

# for prcs in `du -sk * | sort -nr | head | grep -v grep | awk '{print $2}'`
> do ps -ef| grep $prcs | grep -v grep
> sleep 3
> done
oracle 1486 1 0 Apr 04 ? 76:45 ora_lgwr_SCP
oracle 1484 1 0 Apr 04 ? 103:35 ora_dbw0_SCP
oracle 1488 1 0 Apr 04 ? 685:17 ora_ckpt_SCP
oracle 1490 1 0 Apr 04 ? 36:26 ora_smon_SCP
oracle 12493 12463 0 Sep 30 ? 77:19 oracleSCP (DESCRIPTION=(LOCAL=YES)(ADDRESS=(PROTOCOL=beq)))
oracle 1494 1 0 Apr 04 ? 58:27 ora_mmon_SCP
oracle 6770 6739 0 Jan 28 ? 56:47 oracleSCP (DESCRIPTION=(LOCAL=YES)(ADDRESS=(PROTOCOL=beq)))
oracle 6769 6740 0 Jan 28 ? 14:59 oracleSCP (DESCRIPTION=(LOCAL=YES)(ADDRESS=(PROTOCOL=beq)))
oracle 7408 1 0 21:30:49 ? 0:17 ora_cjq0_SCP
oracle 1768 1605 0 Apr 04 ? 1:26 oracleSCP (DESCRIPTION=(LOCAL=YES)(ADDRESS=(PROTOCOL=beq)))

You are exploring the wrong area. The fact /proc is reporting it uses 219 GB while your root partition size is 7.9 GB should have given a clue to you. Forget about /proc which uses no space on disk anyway.
I suspect you cleared files which were open by running processes. The size they use on disk won't be recovered as long as the processes are alive. The simplest way would be to either restart the services using the cleared files or to simply reboot the server.
However, /usr is using 6.2 out of 7.9 GB. If you still need to recover space after rebooting the server, I'm afraid you'll have to either remove unused packages or relocate them. Beware that both options are risky.

Thank you for your assistance. You are right I was looking at the wrong place. Space is now reclaimed after clearing some files from application directory.