GlancePlus - "Global swap space is nearly full"

[Admin.SIF]

We've got several problens on our development server since 2 days. Fron GlancePlus, I receive thoses messages:

Glocal swap space is nearly full
(and)
File table is nearly full

How can I check what is currently use the Swap space and which FIle table is full ?

[Alan Riggs]

File table refers to nfiles. This can be checked with sar -v.

Swap space can be checked with swapinfo.

Both can be checked within glance, as well. "System Tables" will show you nfiles (and other good stuff). Just hit 't' as a shortcut. "Swap Space" will show you swap allocation. 'w' is the shortcut.

[James R. Ferguson]

Hi:

Choose "t" for "table"; choose "w" for "sWap"; choose "?" to get a Glance menu.

If you look at the tables, you should see limits have been exceeded.

is the number of open files for all processes running on the system. Though each entry is relatively small, there is some kernel overhead in managing this table. Each time a file is opened it will consume an entry in even if the file is already opened by another process. When entries are exhausted, a console and/or syslog error message will appear specifically indicating "File table full".

I would check for orphan processes as one cause. See this thread for more information on finding them:

http://my1.itrc.hp.com/cm/QuestionAnswer/1,1150,0x6f717e990647d4118fee0090279cd0f9,00.html

...JRF...