Re: Runaway process and oom-killer

2007-06-14 Thread Helge Hafting
John Sigler wrote: My guess: Something needs memory but finds there is none to be had oom-killer is invoked and targets myapp. myapp takes some time to die. Particularly, the memory it uses isn't freed up instantly. In the meantime something else needs memory and find none. (Another packet recei

Re: Runaway process and oom-killer

2007-06-14 Thread John Sigler
Chris Friesen wrote: Helge Hafting wrote: My guess: Something needs memory but finds there is none to be had oom-killer is invoked and targets myapp. myapp takes some time to die. Particularly, the memory it uses isn't freed up instantly. Has anyone considered actually bumping up the priorit

Re: Runaway process and oom-killer

2007-06-14 Thread John Sigler
Helge Hafting wrote: John Sigler wrote: Andrea Arcangeli wrote: On Wed, Jun 13, 2007 at 10:49:29AM +0200, John Sigler wrote: Question 2: how can I tell which process or kernel thread was hogging most of the RAM when the oom-killer kicked in? Theoretically the one that was killed first bu

Re: Runaway process and oom-killer

2007-06-13 Thread Andrea Arcangeli
On Wed, Jun 13, 2007 at 09:10:33AM -0600, Chris Friesen wrote: > Helge Hafting wrote: > > >My guess: > >Something needs memory but finds there is none to be had > >oom-killer is invoked and targets myapp. > >myapp takes some time to die. Particularly, the memory it uses > >isn't freed up instantly

Re: Runaway process and oom-killer

2007-06-13 Thread Chris Friesen
Helge Hafting wrote: My guess: Something needs memory but finds there is none to be had oom-killer is invoked and targets myapp. myapp takes some time to die. Particularly, the memory it uses isn't freed up instantly. Has anyone considered actually bumping up the priority of the task being ki

Re: Runaway process and oom-killer

2007-06-13 Thread Helge Hafting
John Sigler wrote: Andrea Arcangeli wrote: On Wed, Jun 13, 2007 at 10:49:29AM +0200, John Sigler wrote: Question 2: how can I tell which process or kernel thread was hogging most of the RAM when the oom-killer kicked in? Theoretically the one that was killed first but not for sure in curren

Re: Runaway process and oom-killer

2007-06-13 Thread John Sigler
Andrea Arcangeli wrote: On Wed, Jun 13, 2007 at 10:49:29AM +0200, John Sigler wrote: Question 2: how can I tell which process or kernel thread was hogging most of the RAM when the oom-killer kicked in? Theoretically the one that was killed first but not for sure in current mainline hence see

Re: Runaway process and oom-killer

2007-06-13 Thread Andrea Arcangeli
On Wed, Jun 13, 2007 at 10:49:29AM +0200, John Sigler wrote: > Question 1: why isn't this output picked up by klogd and sent to the > appropriate file (kern.log in my case)? I guess not a question for lkml, most distros uses syslog-ng anyway. > Question 2: how can I tell which process or kernel

Runaway process and oom-killer

2007-06-13 Thread John Sigler
Hello, I have a problem with an app I wrote: after it has run for a long time (sometimes 10-15 hours, sometimes several days) the oom-killer kicks in and snipes most processes in the system. (My process runs in SCHED_RR.) There is a good probability that there is a memory leak in my app, but