This package provides check_ekg
, a generic plugin for Nagios (and
compatible) monitoring systems which translates metrics exposed by
ekg into perfdata suitable
for use with pnp4nagios and similar tools.
There is currently no support for threshold checks; the plugin will
always return an OK
result if it can obtain and parse output from
the EKG endpoint. Its primary function is to allow gathering
performance data over time from EKG-enabled Haskell applications;
there are no current plans to add support for thresholds/alerting,
though I'd consider it in the future if there was demand.
The plugin can either be executed directly on the Nagios host or invoked via NRPE.
Example invocation:
$ check_ekg -e http://haskell-app.example.com:8000
OK: perfdata only | iterations=60614c;;;; rts_gc_bytes_allocated=0c;;;; rts_gc_bytes_copied=0c;;;; rts_gc_cpu_ms=0c;;;; rts_gc_cumulative_bytes_used=0c;;;; rts_gc_current_bytes_slop=0.0;;;; rts_gc_current_bytes_used=0.0;;;; rts_gc_gc_cpu_ms=0c;;;; rts_gc_gc_wall_ms=0c;;;; rts_gc_max_bytes_slop=0.0;;;; rts_gc_max_bytes_used=0.0;;;; rts_gc_mutator_cpu_ms=0c;;;; rts_gc_mutator_wall_ms=0c;;;; rts_gc_num_bytes_usage_samples=0c;;;; rts_gc_num_gcs=0c;;;; rts_gc_par_avg_bytes_copied=0.0;;;; rts_gc_par_max_bytes_copied=0.0;;;; rts_gc_par_tot_bytes_copied=0.0;;;; rts_gc_peak_megabytes_allocated=0.0;;;; rts_gc_wall_ms=0c;;;; ekg_server_timestamp_ms=1435214338224c;;;;
$ echo $?
0
nagios-plugin-ekg
has been tested with GHC 7.8.4 and GHC 7.10.1.