-
Bug
-
Resolution: Fixed
-
Major
-
None
When executing a curried closure in parallel the output appears to be cached (see the output for the spam branch below).
Example pipeline:
def example_c = { input -> node { echo "$input" } } def map = [:] map['spam'] = example_c.curry('spam') map['eggs'] = example_c.curry('eggs') parallel map
Output:
[Pipeline] parallel [Pipeline] [spam] { (Branch: spam) [Pipeline] [eggs] { (Branch: eggs) [Pipeline] [spam] node [spam] Running on java8-jenkins-ec2-slave (i-a541b5be) [Pipeline] [eggs] node [Pipeline] [spam] { [eggs] Running on java8-jenkins-ec2-slave (i-a541b5be) [Pipeline] [spam] echo [spam] eggs [Pipeline] } [Pipeline] [eggs] { [Pipeline] // node [Pipeline] } [Pipeline] [eggs] echo [eggs] eggs [Pipeline] } [Pipeline] // node [Pipeline] } [Pipeline] // parallel [Pipeline] End of Pipeline
- duplicates
-
JENKINS-26481 Mishandling of binary methods accepting Closure
- Resolved
- relates to
-
JENKINS-38268 Parallel step and closure scope
- Resolved
- links to