From: Tom <tom...@gm...> - 2008-05-30 16:26:14
|
Any ideas? On Wed, May 28, 2008 at 1:26 PM, Tom <tom...@gm...> wrote: > I read through the thread posted a while back about zombie processes but I > don't see any evidence of any zombies on my system. However, after a few > hours after restarting MH (sometimes longer, sometimes shorter), mh takes up > around 90% of the cpu. it goes back down after a few seconds then goes back > up. > > Is there a way to see a breakdown of what is running and causing the high > utilization? > > I'm running on CentOS using the latest version from SVN. > > -------------------- > As I was typing this email, I noticed something. > I opened up 2 sessions over SSH and while watching TOP, I was also watching > the MH output > > When MH is eating up the CPU, I see the following behavior: > The spikes happens every minute when the object states are getting saved. > The Saving object states is always followed by a Paused for x seconds > > 05/28/08 01:10:00 PM: Saving object states ... done > 05/28/08 01:10:00 PM Warning, memory leak detected: 109.0 hours: 1.0 MB in > 1.3 hours. 78.2 -> 79.1 at 0.72 MB/hour. Total: 0.37 MB/hour > 05/28/08 01:10:00 PM Paused for 27 seconds > > 05/28/08 01:14:00 PM: Saving object states ... done > 05/28/08 01:14:00 PM Paused for 27 seconds > > When I restarted MH, the CPU doesn't spike every minutes and object state > saves happen without the "Paused for x seconds" > > ----------------------- > > Here's a top when mh is taking up 94.8% of the CPU. > > Tasks: 64 total, 2 running, 62 sleeping, 0 stopped, 0 zombie > Cpu(s): 94.6% us, 5.4% sy, 0.0% ni, 0.0% id, 0.0% wa, 0.0% hi, 0.0% si > Mem: 515544k total, 428484k used, 87060k free, 111816k buffers > Swap: 1048568k total, 236k used, 1048332k free, 148052k cached > > PID USER PR NI %CPU TIME+ %MEM VIRT RES SHR S COMMAND > 26917 root 25 0 > 94.8 2064:47 13.9 78256 70m 2896 R mh > 9218 root 15 0 2.6 0:00.33 0.5 > 8832 2612 2100 S sshd > 9254 root 16 0 2.6 0:00.83 0.2 3836 992 780 R top > 1 root > 16 0 0.0 0:17.17 0.1 1688 544 468 S init > 2 root 34 19 0.0 0:00.02 > 0.0 0 0 0 S ksoftirqd/0 > 3 root 5 -10 0.0 0:00.02 0.0 0 0 0 S > events/0 4 > root 6 -10 0.0 0:00.03 0.0 0 0 0 S khelper > 5 root 5 -10 0.0 > 0:00.00 0.0 0 0 0 S kblockd/0 > 6 root 15 0 0.0 0:00.00 0.0 0 0 > 0 S khubd > 35 root 15 0 0.0 0:29.98 0.0 0 0 0 S kapmd > 38 root 20 0 > 0.0 0:00.00 0.0 0 0 0 S pdflush > 39 root 15 0 0.0 0:16.99 0.0 > 0 0 0 S pdflush > > |