I have what I think is an odd situation, maybe it's my ignorance hanging
out again...
I have a linux guest with 2 vdisk swap disks at different priorities, plus
764MB of real memory (Reserved in VM so it's really real...)
What I don't understand is why I have seen, fairly often, occasions where
all of my swap disks are full, even though there is 370MB of
buffers/cache... I thought linux considered swap evil and wouldn't put
things out there unless it HAD to..
This sample isn't quite "full" but shows the basic idea..
linux2:/var/log # free -m
total used free shared buffers cached
Mem: 755 741 14 0 31 330
-/+ buffers/cache: 378 376
Swap: 253 243 9
So in my naive view of the universe, I should be seeing cached at about
87MB, or so and swap at 0 used, 253 free... Since there shouldn't be any
need to put anything out on swap...I have enough real memory...
I have seen it where swap is totally full (0 free) with similar numbers
for the -/+ buffers/cache: line.
I guess I am trying to find a way to "query" the swap infrastructure to
try and determine what was out on swap at a particular point in time to
try an whittle down this behavior, or at least better understand what is
going on..
If it matters this box is SuSE SLES8, and running db2 udb version 8.
Thanks in advance!
Jeremy
-----------------------------------
Jeremy Warren
Sr. Systems Programmer
KB Toy Stores
mailto:[EMAIL PROTECTED]@kbtoys.com
----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
