Node Step Fails - But Need Other Nodes to Finish

[Russ Robinson]

We have a job setup with 4 steps in it.  If we run it across 10 nodes and 2 of them fail at step 2; is there a setting whereby:

- Those 2 nodes stop/fail with that failed step

- The other 8 nodes continue and run all of the remaining.

Is there an option/feature for this?

[rac...@rundeck.com]

Hi Russ,

I made an example based on 3 nodes, this is the job definition:

- defaultTab: nodes
  description: ''
  executionEnabled: true
  id: 0fcfca07-02f6-4583-a3eb-0002276bdf2d
  loglevel: INFO
  name: HelloWorld
  nodeFilterEditable: false
  nodefilters:
    dispatch:
      excludePrecedence: true
      keepgoing: true
      rankOrder: ascending
      successOnEmptyNodeFilter: false
      threadcount: '3'
    filter: node.*
  nodesSelectedByDefault: true
  plugins:
    ExecutionLifecycle: null
  scheduleEnabled: true
  sequence:
    commands:
    - exec: sleep 5; echo "one"
    - exec: sleep 5; echo "two"
    - exec: sleep 5; echo "three"
    - exec: sleep 5; echo "four"
    keepgoing: false
    strategy: node-first
  uuid: 0fcfca07-02f6-4583-a3eb-0002276bdf2d

Basically setting the Thread Count with the number of nodes and set “Continue running on any remaining nodes before failing the step.” in “If a node fails” (both on the Nodes job tab).

Check the result (in the example fails in the first step but the node can fail on any step and the result is the same).

Hope it helps!

​