Is there by any chance any on-going effort on creating a pipeline step that allows passing command line arguments (to prevent command injection) safely to a shell step & externalize the script file? (while typing this, I just realized, that I could do something similar with combination of env-variables and writing the script file from resources to a temp file)
Hello, I'm using jenkins with kubernetes plugin and I'm noticing that no matter what changes I make to the podtemplate, those changes don't get merged with the podtemplate defined in the local Jenkinsfile. Could anyone point me where I can further debug this? Is there a place that plugins log stuff?
Not sure if anyone else is worried about script input escaping, but if you are then you might find https://github.com/samlink-devops/jenkins-pipeline-examples/blob/master/vars/safeSh.groovy useful.
Hi all, has anybody seen jobs being stuck/superslow immediately after the start of execution? Console freezes at second line in the following gist for minutes and then progresses through the next lines for hours before it even starts cloning. This is transient, other identical jobs run fine.
GitHub pull request #8502 of commit fb98cf3585100ea72bffefd47d7a536a8e773401, no merge conflicts.
Running as SYSTEM
Setting status of fb98cf3585100ea72bffefd47d7a536a8e773401 to PENDING with url https://jenkins.galaxyproject.org/job/docker-api/16927/ and message: 'Test started.'
Using context: api test
[EnvInject] - Loading node environment variables.
Building remotely on n2.galaxyproject.org (psu_node doc_builder docker_worker) in workspace /var/jenkins/workspace/docker-api
[WS-CLEANUP] Deleting project workspace...
[WS-CLEANUP] Deferred wipeout is used...
[WS-CLEANUP] Done
pipeline {
agent { label 'windows' }
stages {
stage('Deploy') {
agent { label 'windows' }
steps {
script {
def proceed = true
try {
timeout(time: 15, unit: 'SECONDS') {
input {
message "Is this deployment allowed to continue?"
submitter "Deployment Group"
submitterParameter "APPROVER"
parameters {
choice(
name: 'TARGET_ENV',
choices: "Dev\nTest\nProd",
description: "Please choose an environment target"
)
}
}
}
} catch (err) {
proceed = false
}
if(proceed) {
powershell "Write-Host Deployment to $env.TARGET_ENV just approved by $env.APPROVER"
}
}
}
}
}
}
doesn't seem to work; build just immediately succeeds
bleh
or anyone else got something like this working in a declarative pipeline? :)
oh, this is probably one for the pipeline-authoring-sig
Hi, I have just a quick question about running Jenkins on K8S. I installed via the stable helm chart, and it's working all well except for one aspect. I'm trying to serve Jenkins over port 443 for SSL of course. This works, but Jenkins tries to go to http://domain instead of https://domain on login, or when trying to make pipelines, etc. Any suggestions?
The stable chart is https://github.com/helm/charts/tree/master/stable/jenkins, and I set master.servicePort to 443, and master.jenkinsUrlProtocol to https
Getting InterruptedException in pipeline build especially when the pipeline stage has multiple parallels components?
any idea how to fix this
this happens 1 in every 7 build
[Pipeline] [Build on Linux] deleteDir
[Pipeline] [Build on Linux] echo
[Build on Linux] Build on Linux is done
[Pipeline] [Build on Linux] }
[Pipeline] [Build on Linux] // ws
[Pipeline] [Build on Linux] }
[Pipeline] [Build on Linux] // node
[Pipeline] [Build on Linux] }
[Build on Windows] Cannot contact <windows machine name >: java.lang.InterruptedException
Aborted by xxxxx
[Build on Windows] Sending interrupt signal to process
[Build on Windows] After 10s process did not stop
[Pipeline] [Build on Windows] }
[Pipeline] [Build on Linux] echo
[Build on Linux] Build on Linux is done
[Pipeline] [Build on Linux] }
[Pipeline] [Build on Linux] // ws
[Pipeline] [Build on Linux] }
[Pipeline] [Build on Linux] // node
[Pipeline] [Build on Linux] }
[Build on Windows] Cannot contact <windows machine name >: java.lang.InterruptedException
Aborted by xxxxx
[Build on Windows] Sending interrupt signal to process
[Build on Windows] After 10s process did not stop
[Pipeline] [Build on Windows] }
space onn windows is ample compute power is super high
pipeline got stuck for more than 9 hours
here
until aborted
if you want i can share jenkinsfile
but I wonder if that interruptedexception is not simply a result of you aborting the job
and ultimately you have a problem elsewhere in your pipeline
yes
most of the times , windows give me the problem
slave is running as a service
dm