Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations Chris Miller on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Weird FS size problem 1

Status
Not open for further replies.

KenCunningham

Technical User
Mar 20, 2001
8,475
GB
Hi folks. I've a wierd problem with my p650 AIX 5.2 box. It houses a live and a test environment for an Oracle database. These should be almost exactly the same, give or take minor differences in the size of the control file, as the test environment is regularly refreshed using the live datafiles themselves. I've verified this in every way I know of, but the test filesystem stubbornly remains 0.5 Gb larger than the live if I do a df -g on both.

Has anyone any experience of this and how can I further verify how this extra space is being used in the test filesystem? TIA.
 
Perhaps try comparing output of a df -k of each databases' filesystem, to narrow down whether you have a subdirectory which is bigger than the other. I'm thinking maybe a datafile was dropped from the database but might still be on disk.
 
Hi Zahier, I've done that thanks, even to the extent of exporting the output of an ls -la to Excel in case I missed something when adding up the file sizes manually. Still that darn .5 Gb plagues me!! Thanks for the suggestion though.
 
Hi Ken,

The only times I've seen discrepancies between ls and df output have been time where sparse files are in use.

You may be looking at your ORACLE TEMP dbf files being sparse, and their actual utilisation profile being different.

Or perhaps one systems are sparse, while the others are not.

When sparse files are used, df reports actual (real) disk blocks allocated, while ls displays the files size as shown within its inode.
As I'm sure you already know ;-)

Worth a look anyway :)

All the Best.

____________________
Sometimes it pays to stay in bed on Monday, rather than spending the rest of the week debuging Mondays code.
 
Many thanks, I hadn't thought of that, but it's certainly a possibility. I'll investigate further and get back to you. Have a good weekend!
 
d3vzero. Short weekend that wasn't it? You hit the nail on the head - have a Friday star!!!
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top