For a good reason, systems tend to kill the process which has the
largest VM. In some cases it's a process that just went stray. In the
more common cases it would be that process that's working on
an important simulation for the last month or just your X server.
> I'm quite sure other OSes have similar funcitonality, such as AIX
AIX 3.x had this feature and IMHO it was causing too much problems.
IIRC an option to change the default bahavior came only in 4.x.
AIX also has two signals that are sent to all processes when the
swap space is full above certain thresholds, thus warning processes
about getting close to OOM condition. IIRC, a process that catches
SIGWARN is considered "well bahaved" by the system so it does not
get the SIGKILL that follows.
-- Itai
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/