Pipeline from SCM configured to do a sparse checkout and ignore commits from an automated user
Pipeline DSL operates within a docker container, with changelog and poll both set to false.
Using strategy: Default
[poll] Last Built Revision: Revision 57399c846b68250e81dff6f34feeb292664bdba5 (origin/master)
> git rev-parse --is-inside-work-tree # timeout=10
Fetching changes from the remote Git repositories
> git config remote.origin.url g...@github.com/kallan357/thing.git # timeout=10
Fetching upstream changes from g...@github.com/kallan357/thing.git
> git --version # timeout=10
using GIT_SSH to set credentials kallan357
> git fetch --tags --progress g...@github.com/kallan357/thing.git +refs/heads/*:refs/remotes/origin/*
Polling for changes in
> git rev-parse origin/master^{commit} # timeout=10
> git log --full-history --no-abbrev --format=raw -M -m --raw 57399c846b68250e81dff6f34feeb292664bdba5..31621a03c0eb1d58be8e8b0d0eebed6bae6ccb23 # timeout=10
Ignored commit 31621a03c0eb1d58be8e8b0d0eebed6bae6ccb23: Found excluded author: jenkins-guy
Using strategy: Default
[poll] Last Built Revision: Revision 57399c846b68250e81dff6f34feeb292664bdba5 (origin/master)
using GIT_SSH to set credentials kallan357
> git --version # timeout=10
> git ls-remote -h g...@github.com/kallan357/thing.git # timeout=10
Found 45 remote heads on g...@github.com/kallan357/thing.git
[poll] Latest remote head revision on refs/heads/master is: 31621a03c0eb1d58be8e8b0d0eebed6bae6ccb23
Done. Took 0.43 sec
Changes found
Here is my DSL that lives inside SCM alongside my project:
node('java-docker-node') {
timeout(15) {
withEnv(envArray) {
stage 'Checkout'
git branch: 'master', credentialsId: 'abc123', changelog: false, poll: false, url: 'g...@github.com/kallan357/thing.git'
stage 'Build'
sh '''
mvn clean install
'''
stage 'Bump versions'
sh '''./bump_version.sh'''
}
}
}
My bump_version script will bump my pom and commit a new version to SCM. Because of the double sets of polling my job gets stuck in an infinite loop.
Am I missing something? I've executed my build, which I suspect should modify the Jenkins polling behavior to only have a single attempt.
Thanks,
Kyle Allan