Michael Rothwell wrote: > Killing, for instance, kswapd. On my machine last week, the kernel > started killing processes. Its first choice was kswapd, and then it went > on to my window manager and about 10 other programs. > > It was pretty retarded. I agree, killing init, kernel threads, etc, should be the -last- possible choice of kills. For some reason the OOM killer never seems to get the process that's currently sucking in memory, it -always- gets the innocent guys and on X machines invariably kills it thus rendering the machine unusable as the video isn't recoverable. > I know Linux overcommits memory and has no way of knowing who the real > overallocator culprit is, but wouldn't some resource accounting be good? > Or perhaps a tunable or compile-time option to turn off overcommits? Yes, it would be good. -d -- "The difference between 'involvement' and 'commitment' is like an eggs-and-ham breakfast: the chicken was 'involved' - the pig was 'committed'."
begin:vcard n:Ford;David x-mozilla-html:TRUE org:<img src="http://www.kalifornia.com/images/paradise.jpg"> adr:;;;;;; version:2.1 email;internet:david _at_ kalifornia.com title:Blue Labs Developer x-mozilla-cpt:;-12480 fn:David Ford end:vcard
Follow-Ups:
- Re: Out of memory?Jesse Pollard <pollard _at_ cats-chateau.net>
- Out of memory?Alexander Valys <avalys _at_ optonline.net>
- Re: Out of memory?David Ford <david _at_ kalifornia.com>
- Re: Out of memory?Michael Rothwell <rothwell _at_ flyingbuttmonkeys.com>
- Prev by Date: Re: Updated Linux 2.4 issues page
- Next by Date: BUG: 2.4 Whom can I discuss the initrd boot sequences with??
- Prev by thread: Re: Out of memory?
- Next by thread: Re: Out of memory?
- Indexes:[Main][Thread]