lkml.org 
[lkml]   [1998]   [Aug]   [17]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [discussion] Swap overcommitment recovery
This is not an endorsement or rejection of your idea, just a comment on
one thing you may have overlooked. One of the things you stated was

> 2) Modify the swap subsystem to set the flag when we reach a configurable
> threshhold. (ex. 90% utilization) Reset the flag when we drop back below
> this threshhold. 2.1) When the flag is set, signal init to run the
> nomemory action for the current runlevel.

I'd recommend some hysteresis factor in setting this limit. Say
something
like trigger at 95% but do not reset until 90%. This will prevent you
from ending up constantly setting and resetting your flag. As you
stated
it originally, if your utilization hovers at around 90%, which it would
tend to on a heavily used system (lots of small processes) you could eat
up
quite a bit of CPU just managing this flag and it's effects on the
system.
If the goal is to preserve enough system so root can still get in and
fix
problems, the last thing you want to do is get into a storm handling
this
flag.


--
Patrick R. Hickel
Senior Computing Systems Administrator
Boeing Information and Support Services

Email: hickel@puffball.ca.boeing.com

-
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.altern.org/andrebalsa/doc/lkml-faq.html

\
 
 \ /
  Last update: 2005-03-22 13:44    [from the cache]
©2003-2011 Jasper Spaans