Forum: CFEngine Help Subject: Tracking down a memory leak Author: sauer Link to topic: https://cfengine.com/forum/read.php?3,23558,23558#msg-23558
I'm seeing a memory leak in cf-monitord on one RHEL 5.7 system (virtualized on VMWare, for what that's worth). It looks like this a few minutes after it's started: PID TTY STAT TIME MAJFL TRS DRS RSS %MEM COMMAND 16912 ? Ss 0:00 0 0 37788 6192 0.9 /var/cfengine/bin/cf-mo And then it steadily grows, eventually looking like this after ~24 hours: PID TTY STAT TIME MAJFL TRS DRS RSS %MEM COMMAND 18187 ? Ss 8:27 185 0 144828 113336 16.9 /var/cfengine/bin/cf-monitord The problem was happening with 3.1.5, and continues to happen with 3.2.0. This has been repeatable every day since I enabled cf-monitord and installed 3.1.5 last week. I haven't opened a bug report yet because this is only occurring on a single system; I have several thousand others which are pretty much the same but are not exhibiting this behavior. Or maybe I'm not noticing it because I don't have many which are down in the 650MB RAM range like this one. :) In any event, I'm not really sure where to start looking for what's causing the ever-increasing memory usage - whether it's a lot of activity on the system causing internal tables to get really big, or a defect. Any suggestions? Having to restart the process daily is a tad inconvenient. _______________________________________________ Help-cfengine mailing list Help-cfengine@cfengine.org https://cfengine.org/mailman/listinfo/help-cfengine
