I'd start by not using ubuntu in production.

If you insist:
You need to run your kernel at 1000hz and not tickless.

We recommend CentOS because it's been the most stable and reliable distro we have found.


On Thu, Jun 3, 2010 at 2:03 PM, Matthew Fong <mattdfong@gmail.com> wrote:
I'm running 1.0.6 and I noticed that right after upgrading from 15135 that the load (15 min avg) on my system seems to have increased about double running the same amount of traffic. I've been running top all day for the past few days and I noticed that for the most part I always have 75% idle cpu, but maybe once ever 45 minutes or an hour there is an spike the 1 minute load averages (maybe up to 9). This is only a dual core processor so I believe 9 usually indicates that are some processes that are waiting. The idle % does not change, and I have free memory, but for whatever reason the linux load algorithm spikes (maybe very high for like 1-5 seconds).

At first I thought this might be related to a tickless timer, so I disable it from the kernel, but today I am still getting the same spikes. So my question is, are these spikes something to worry about, or should I just monitor the % of idle cpu available on my box for guesstimating how much further I can push this system? Thanks.

--matt

_______________________________________________
FreeSWITCH-users mailing list
FreeSWITCH-users@lists.freeswitch.org
http://lists.freeswitch.org/mailman/listinfo/freeswitch-users
UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users
http://www.freeswitch.org




--
Anthony Minessale II

FreeSWITCH http://www.freeswitch.org/
ClueCon http://www.cluecon.com/
Twitter: http://twitter.com/FreeSWITCH_wire

AIM: anthm
MSN:anthony_minessale@hotmail.com
GTALK/JABBER/PAYPAL:anthony.minessale@gmail.com
IRC: irc.freenode.net #freeswitch

FreeSWITCH Developer Conference
sip:888@conference.freeswitch.org
googletalk:conf+888@conference.freeswitch.org
pstn:+19193869900