[collectd] cpu wait time on collectd server
siiludus at gmail.com
Wed Sep 12 15:44:07 CEST 2007
Yes it was indeed disk wait.
I haven't had time yet to try the rrd tuning suggestions, but i now
activated caching in the rrdtool plugin.
The results are dramatic. I will attach a graph to illustrate it.
In the graph begginnig we are testing collectd with few hosts sending data.
Then we add clients until about "wed 12.00" when we had reached 30+
clients sending data and ~1500 rrd files.
And then i activate rrd plugin cache.
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 48473 bytes
Desc: not available
Url : http://mailman.verplant.org/pipermail/collectd/attachments/20070912/d8ce09f7/attachment-0001.png
More information about the collectd