My guess is that it is caused by an erroneous (custom) gmetric causing an XML
error.
So far I have been unable to find what is causing this or how to reproduce
this.
It only seems to be triggered occasionally under certain conditions that are
not clear to me. Dumping the XML at the time of th
Based on your graphs this happens randomly ? It would be interesting to
see if you cannot connect to gmetad during those times. Stracing gmetad
and doing
netstat -an | grep 865
may be helpful.
BTW there is a gmetad health checker someone wrote which may alert you to
this situation early.
htt
We detect it when the website stops responding (as described on
ganglia-developers list).
Then it is 'fixed' by indeed simply restarting gmetad.
As of January 2013, SARA has a new name: SURFsara.
ing. Ramon Bastiaans - Senior Systems Programmer - Cluster Computing
| Operations, Support & Devel
3 matches
Mail list logo
