On Tue, Feb 23, 2010 at 10:04 PM, Paul Querna <p...@querna.org> wrote: > On Fri, Feb 19, 2010 at 1:08 PM, Niklas Gustavsson <nik...@protocol7.com> > wrote: >> Would it be okay to install the Slave status >> (http://wiki.hudson-ci.org/display/HUDSON/slave-status) in Hudson? >> That way, slaves would be possible to monitor from Nagios. I'm seeing >> some crashes of Hudson on the Windows slave, and would like to be >> notified when this happens. > > how do you intend to actually monitor it? NPRE?
I was hoping we could use check_http from the main Nagios server, similar to for example JIRA. The major difference would be that in this case we also need to check for a specific reply using the -r argument. Here's the example from the slave-status plugin docs: check_http -H localhost -p 3141 -r '<status>Running</status>' Of course, in our case with not using NRPE, localhost would be replace by the real host name. Do you think this would work? /niklas
