No commit activity in last 3 years
No release in over 3 years
Sensu + Slurm/SchedMD
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
 Dependencies

Development

~> 1.7
~> 0.10
~> 10.0
~> 3.2
~> 3.1
~> 0.49.0
~> 0.9.11

Runtime

 Project Readme

umarcts-sensu-plugins-slurm

Overview

This package/handler is meant for use in HPC setups that use SchedMD/Slurm. The main purpose of this handler is the notify the scheduler when a node is down so jobs can be drained and no new jobs land on that node. The mechanism for this is scontrol.

To use handler-scontrol in Sensu

Define handler

{
  "handlers": {
    "slurm": {
      "command": "/opt/sensu/embedded/bin/handler-scontrol.rb",
      "type": "pipe",
      "severities": [ "critical" ]
    }
  }
}

Set handled_by_scontrol: true in check definition(s)

{
  "checks": {
    "check_slurmd_process": {
      "command": "check-process.rb -p slurmd",
      "interval": 30,
      "occurrences": 2,
      "handled_by_scontrol": true,
      "subscribers": [ "compute" ],
      "handlers": [ "slurm" ]
    }
  }
}