On 9/7/10 9:29 AM, "no-re...@cfengine.com" <no-re...@cfengine.com> wrote: > Forum: Cfengine Help > Subject: Re: cf-monitord memory usage > Author: Igor > Link to topic: https://cfengine.com/forum/read.php?3,18104,18105#msg-18105 > > I have cd-monitord started about a month. Same results: > > 129M 49880K nanslp 1 4:42 0.00% cf-monitord > 129M 125M nanslp 1 4:44 0.00% cf-monitord > 61680K 57252K nanslp 1:21 0.00% cf-monitord > 61680K 57304K nanslp 0:58 0.00% cf-monitord > > This is not good...
Does it ever level off, or just continue to climb until the daemon becomes unresponsive? What platform? I ask, because I have observed and dug into the same issue on FreeBSD 6.1/i386 and 6.3/amd64 with 2.x. After setting up cacti and watching the trend, it looked like an obvious leak. I tried to be graceful and let it run, but it would eventually stop responding after consuming enough memory (usually at 3AM, paging people). My only way to get it stable enough for production was a combination of daemon monitoring and event handlers. It would be nice to know this has been officially diagnosed/resolved. _______________________________________________ Help-cfengine mailing list Help-cfengine@cfengine.org https://cfengine.org/mailman/listinfo/help-cfengine
