>>> On Fri, Jul 6, 2007 at 1:59 PM, in message <[EMAIL PROTECTED]>, "Peter E. Abresch Jr. - at Pepco" <[EMAIL PROTECTED]> wrote: > This is occurring on fully patched SLES9x 2.6.5-7.286-s390x. It is running > as a z/VM Guest on z/VM 5.1 plus selected maintenance. > > Free shows the following: > > total used free shared buffers cached > Mem: 763536 758684 4852 0 126760 49640 > -/+ buffers/cache: 582284 181252 > Swap: 899896 560492 339404
That's an awful lot of page space in use, and very little free memory. As Ursula Braun stated in the thread Brad pointed out, if it only happens rarely, this might not be a big deal, but if it starts happening regularly, you probably need to increase your virtual storage allocation. I would be curious to see what your paging _rate_ is for that system. If it's very low, you may be able to avoid the problem in the future by adjusting the kernel's "swappiness" to keep more free memory around. > This has not occurred in the past and has not occurred since yesterday. It > seems to have been trigged by cclclnt which is CICS Transaction Server > Client which was using the HSI interface. Eventually cclclnt died and had > to be restarted. I am completely unfamiliar with this middleware, so I don't have anything to base suggestions on. Does the installation documentation recommend any particular virtual storage size? -snip- > 768M has been more than enough in the past. I am going to try to trace > back the cclclnt task and see if that was the catalyst. It has always been > running in the past without problems. How long had the system been up? It's possible that there's a memory leak in the application (or in something else running on the same system) that might cause things to get critical after some time has passed. Mark Post ---------------------------------------------------------------------- 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
