All entries for Thursday 05 January 2012

January 05, 2012

Solaris df reported disk usage differently to du

I dropped the hitlogging tablespace whose size was 200+ gb.

run df report

bash-3.00# df -h
la-ora01-db            588G    31K    44G     1%    /oradb
la-ora01-db/data       588G   489G    44G    92%    /oradb/data

bash-3.00# du -s -h /oradb/data
 217G   /oradb/data

-bash-3.00# for pid in `ps -ef | grep ora_ | awk '{print $2}'`; do 
 pfiles  $pid | grep log  ; done | sort | uniq
The files I drooped were hitlogging*.dbf and non of them appear in the list above.
After 3-4 hours waiting to remove the possibility of cache effect, I restarted the database

SQL>shutdown immediate;
Database closed.
Database dismounted.
ORACLE instance shut down.

-bash-3.00# df -h
Filesystem             size   used  avail capacity  Mounted on
/                       39G   8.8G    30G    23%    /
/dev                    39G   8.8G    30G    23%    /dev

la-ora01-db            588G    31K   316G     1%    /oradb
la-ora01-db/data       588G   217G   316G    41%    /oradb/data

Maybe some other user process hold the files? I only inspected the oracle background process and usually the most likely suspect is db file writer (dbwn).

Search this blog

Search Warwick Blogs



Most recent comments

  • Oracle 11g is the clear leader in this area. Oracle has invested heavily into self–tuning capabiliti… by shaane on this entry
  • All the features are very nice.I like Repair Advisors the most.Oracle 10G introduced some of the dat… by lucy on this entry
  • "logical standby failed to re–start, I am also getting the same error. I have search on internet and… by anemia on this entry
  • Thanks for the nice information because from last three days i am stuck on this problem…. by robot vacuum on this entry
  • So what happen City speed dating when you tried this coding once in your PC.. Please tell me so i am… by John Bergeron on this entry
Not signed in
Sign in

Powered by BlogBuilder