Skip to content

Latest commit

 

History

History
29 lines (26 loc) · 1.72 KB

JobFiltering.md

File metadata and controls

29 lines (26 loc) · 1.72 KB
Raw

Moneo Job Level Filtering

Description

Moneo provides a way to designate a job name/Id for nodes participating in a job. The Job Id can be used to filter on nodes that belong to a specific job.

Steps

Moneo must be already deployed and running on the cluster where job grouping will be designated.(Please note that this feature is not supported if Moneo-worker is deployed inside docker, will support it in the future.)

  1. To set a job Id to a subset of nodes a config.ini file with the desired hosts must be passed in:
  • python3 moneo.py -j 4 -c hostfile
  • The 'hostfile' file will be the same format as the config file used to deploy Moneo with the exception that the worker list will only include the hosts for the job group.
  • i.e. 
    192.168.0.100
192.168.0.101
192.168.0.110
    Note: The master node can also be work node as well. The master node will have the Grafana and Prometheus docker containers deployed to it.
  • The hosts listed under the worker label will be part of the specified job group.
  1. The python3 moneo.py -j 4 -c hostfile command can be used in a job scheduling script such as a PBS scipt, Sbatch script or called manually from command line.
  2. At the completion of the of the job python3 moneo.py -j None -c hostfile command can be called again with either "None" or some default Job ID. AGain this can be put at the end of a job scheduling script.

Portal View

  1. Once the Job Id has been specified using the moneo CLI command the Job Ids should populate the Grafana portal: image Note: This might require a browser refresh for recent Job ID updates.