Re: heavy activity problems

Ali Akcaagac wrote:

On Sun, 2002-08-18 at 20:48, hobbit aloss ukuu org uk wrote:
if it goes in critic conditions and swap around (on worst case for many
minutes) but should not fail. only if the ram really exceeds then it
should crash or exit correctly.
This isn't the Linux kernel messing around, is it? Are there
any messages in dmesg about "out of memory"?
Just a shot in the dark, having met it once.

yes there are a couple of messages.... BUT....

let's say i start an application in an environment that has enough
memory, that app usually get setarted and fills it's startaddress and
endaddress in the memorylist. it also allocates other memory for it's
workstuff and datastuff..

with other words everything of this has been done already. so all this
memory is already addressed as 'beeing in use' so the kernel should have
no reason to kill the app only because of beeing out of memory (where i
recall that there is still 100mb free spread over physical and swap
memory). so the only reason for the kernel to kill the app is that
nautilus or the other apps are allocating memory or doing some strange
background activity for no reasons.

an app that runs, runns until it crashes or the machine fails for other
reasons. but it should not crash after it has allocated all memory and
is doing nothing in the background.
You haven't really given enough information to diagnose this problem. It would help to know what messages were printed to stderr when nautilus crashed (on Red Hat, the default session script redirects stderr to ~/.xsession-errors file). As Telsa mentioned, it would also be good to check the syslog to see if the kernel is killing processes to free memory. Without this information, all we can do is guess at the cause of problem.


Email: james daa com au              |
WWW: | Jan 22-25 Perth, Western Australia.

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]