lkml.org 
[lkml]   [2000]   [Jul]   [8]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Kernel 2.2.14 OOM killer strikes.


On Sat, 8 Jul 2000, Mike A. Harris wrote:

>
> >Now, whether Netscape should register such a handler is up for debate -
> >it may very well be the culprit for causing the OOM situation in the
> >first place. However, it _could_ also free a lot of memory from its
> >internal cache as well.
>
> Yes indeed. Netscape is definitely an OOMish app. It would be
> nice to be able to say to netscape "hey, give me back 50Mb of RAM
> you potato!" and have netscape actually do it, assuming it isn't
> memleaking (yeah right... ;o)
>


Hey! That's the real problem! SIGDANGER is a good ideia, but if you're
going to install a sig handler for every application running, then I don't
know if it will work at all.

OK, Netscape usually has tons of memory that could freed, but it is
_not_ an application that you can rely on to do that for you. Netscape
is extremely buggy! You will be in luck if it even reacts when it receives
SIGDANGER. Sometimes it even ignores SIGTERM.

Actually Netscape is one probable reason why your system could be
OOM'ing, since it frequently gets out of control. If it installs a handler
for SIGDANGER, when your system runs out of VM it surely won't kill
Netscape. From the last discussions I understood that the purpose of
SIGDANGER was not to kill any process with a handler for it, for they were
important. As for other processes with no handler, some will get killed to
free memory.

regards

Claudio Martins




-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:57    [W:0.152 / U:0.176 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site