This nagios/icinga plugin simply connects to a ceph-dash instance, gathers the information about the overall cluster health and outputs a nagios compatible message including several perfdata metrics.
dependencies
Just plain python. Uses urllib2 to gather data from the ceph-dash instance.
./check-ceph-dash.py --url 'http://url.to.ceph.dash/'
OK: ceph cluster operates with no problems|bytes_used=244455534592 bytes_total=359702620323840 bytes_avail=359458164789248 data_bytes=110961570327 num_pgs=5120 op_per_sec=1 read_bytes_sec=0 write_bytes_sec=1913
By using the generated performance data and sending them to a graphite backend, you can create quite fancy metric dashboards. See the following screenshot for an example (the frontend is Graphana). I have included the configuration file in the contrib
folder. You just have to edit the field with <YOUR_GRAPHITE_PREFIX>
to your graphite data sources.