I am attempting to troubleshoot a problem with zencommand timing out. I have created my own test script that outputs data in the format TEST OK;|OneMin=0.00 FiveMin=0.00 FifteenMin=0.00 RunningProcs=1 TotalProcs=138 LastPid=29394
I added a new template to the /Devices/Linux class of servers that calls this command on each of the hosts via SSH. When I run zencommand run -d <host> -v 10, the command completes without issue and the data is recorded properly to the appropriate rrd file(determined by rrdtool info /path/to/file). However, when I run zencommand run --cycle -v 10 without specifying a host, the command times out for most of the hosts that I am attempting to run the command on and the rrd files for the hosts that do appear to complete do not get updated. The symptoms appeared to be similar to the issues described in http://community.zenoss.com/forums/viewtopic.php?t=5467&postdays=0&postorder=asc&start=15 I have checked to ensure that the RRD files are owned by zenoss and have moved the test script to the same location in the filesystem (local and remote) as zenoss is installed. Details: Zenoss 2.2 (recently upgraded from 2.1.1 and I was having the same problems previously) CentOS 5.1 As I just discovered while typing this post the CPU usage is pegged at 99% with a load of over 3(dual-core CPU machine). Memory is also being heavily utilized. Right now, zencommand has 26 checks. I would appreciate any advice with respect to how to go about solving this. -------------------- m2f -------------------- Read this topic online here: http://community.zenoss.com/forums/viewtopic.php?p=20173#20173 -------------------- m2f -------------------- _______________________________________________ zenoss-users mailing list [email protected] http://lists.zenoss.org/mailman/listinfo/zenoss-users
