Hi All, I'm running Zenoss 1.1.2 with FC6 on a Dell PowerEdge 2650 with 2 - 2.4GHz Intel Xeon Processors and 4Gb of memory. I currently have 247 Cisco devices discovered, consisting of 6500's, 7507's, 3640's, and Cat 5500/5505's. I'm using the standard localhost performance template, collecting standard stats for Network devices and Ethernet interfaces. Currently, there are 47429 RRD files in /opt/zenoss/perf/Devices.
According to 'top', at every performance polling interval zenperfsnmp spikes to 100%. During this time, the server's performance suffers greatly and Zenoss generates heartbeat failure errors. Out of 3895424K memory, 3814424K is consistently used. The zenperfsnmp logfile indicates completion of the performance poll in ~145 seconds. Other than that, there are only a few Bad OID errors, which I am investigating. I'm surprised that there is such a load on the server with this installation. Has anyone else encountered similar effects? I'm considering adding another performance instance, but I wanted to find out if there is any performance tweaking that can be done on the local server, either at the system level or within Zenoss. I still have additional devices to populate into Zenoss, but with the performance impact, I may have to look at restricting where performance polling is done. Any feedback regarding this issue would be greatly appreciated. Thanks, Rob- - -------------------- m2f -------------------- Read this topic online here: http://community.zenoss.com/forums/viewtopic.php?p=7932#7932 -------------------- m2f -------------------- _______________________________________________ zenoss-users mailing list [email protected] http://lists.zenoss.org/mailman/listinfo/zenoss-users
