Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
    Peter Brown
    @TherealPoots
    Hi folks, I'm new to jenkins and have a non-technical question. Would you say it's overkill If i aim to become a groovy expert in order to be able to understand any complex pipelines that I may come across in the workplace? I've started a groovy course, and the tutor has recommended we read Groovy in Action, which is almost a thousand pages long
    Mark Waite
    @MarkEWaite
    If groovy scripting will help you in other areas (like automating tasks using free-standing Groovy scripts), then learning groovy is a good investment. If your only reason to learn Groovy is for the Jenkins Pipeline Domain Specific Language, that seems like a lot of learning for a language that you would generally prefer to use as little as possible. I prefer to keep my scripting in a script file that any developer can call. Allows easier diagnosis and investigation if things are not behaving as expected
    Peter Brown
    @TherealPoots
    @MarkEWaite: thanks for the response. A friend of mine was on a contract working for a large IT company where they were using shared libraries for their pipelines. He says they used a lot of groovy which put him in a state of panic as he couldnt make heads or tails of it. I'm trying to enter the devops world and want to give myself as much of head start as possible.
    Mark Waite
    @MarkEWaite
    Scripting languages like Groovy are great tools to solve DevOps problems in general. They can be used to solve them from separately called scripts (preferred) or from inside the Jenkins Pipeline. Learning a scripting language like Groovy opens all sorts of possibilities for ways to help people solve problems faster and better, independent of the Jenkins Pipeline
    Peter Brown
    @TherealPoots
    Are you familiar with groovy at all? I want to know whether the following areas of groovy are relevant to jenkins shared libraries or pipelines in general: metaprogramming, builders, and concurrent groovy.
    I'm want to avoid learning things I don't need to learn
    Mark Waite
    @MarkEWaite
    I know just enough groovy to adapt to Jenkins Pipelines when I can't use declarative syntax.
    Peter Brown
    @TherealPoots
    Gotcha. Thank you.
    Eliza Leong
    @LizLeong
    Hi Everyone: I am attempting to use the migration plugin that helps one move from freestyle to declarative pipeline. I installed the Declarative Pipeline Migration Assistant and the Pipeline plugins as instructed in https://docs.cloudbees.com/docs/admin-resources/latest/pipelines/declarative-pipeline-migration-assistant . After that I restarted Jenkins. The problem is this: I do not see the "To Declarative" menu option after these steps. Has anyone successfully used this tool to generate a Jenkinsfile? (This is on Windows).
    Jerry Wiltse
    @solvingj
    Are there any new or advanced strategies around creating Jenkins Shared Library functions which are callable from either a CPS context or a NonCPS context? In a worst case scenario, I could see creating duplicate functions: one with @NonCPS and one without. But, I would prefer not to do that.
    Jerry Wiltse
    @solvingj
    I just found this slideshow from Jesse Glick from what looks like 2016; In it, I just discovered a small detail which I'm hoping is no longer valid. On slide 28, it says "exotic constructs not yet translated in CPS". One of the examples is "obj as Interface". We're doing casting to this as IJenkinsBuild (an interface we define) in some of my most recent global vars. It seems to be working but, maybe there's issues we're not aware of. Does anyone know if support for casting has been added to the CPS translation in recent years?
    http://events17.linuxfoundation.org/sites/events/files/slides/Building%20an%20extensible%20DSL%20in%20Apache%20Groovy.pdf
    sjothili
    @sjothili
    Hi,
    i am novice to Jenkins Shared library. I am looking out for basic shared lib usage documents or links with more examples. If anybody have it handy, please do share. Appreciate this help!
    dduportal
    @dduportal:matrix.org
    [m]
    Hello there, the latest version 2.40 of the plugin "Pipeline: Nodes and Processes" (https://github.com/jenkinsci/workflow-durable-task-step-plugin/releases/tag/workflow-durable-task-step-2.40) seems to have a metadata issue: it is marked as incompatible with the core version for LTS (2.303.1). The release note mentions 2.307+ for the test harness of this plugin, but how can a user know more about this?
    (and side question: why is it present in the update center? I thought it was filtered per Jenkins core version)
    Mark Waite
    @MarkEWaite
    I'm not seeing any issue with the 2.40 release of "Pipeline: Nodes and Processes" in my checks @dduportal:matrix.org . The plugins site https://plugins.jenkins.io/workflow-durable-task-step/ shows that it requires Jenkins 2.249.1. It is installed on my Jenkins 2.303.1 and running. Plugin installation manager tool 2.11.0 did not complain when I ran it to check for plugin updates on my 2.303.1 installation. Are you still seeing that failure?
    Steven Foster
    @sfosterdw
    Hi team, I was thinking of adding some extra metadata to the ChangeRequestSCMHead interface to centrally support recent features of a few different SCMs as per https://issues.jenkins.io/browse/JENKINS-66782 if anyone has any thoughts or other suggestions that'd be neat.
    Jerry Wiltse
    @solvingj
    Does anyone have any insight as to why nested parallel stages are still "intentionally unsupported"? Is there some CPS limitation making it impossible, or is it prohibitively complex to implement, etc? There are several tickets about it. If someone would explain "why" it's that way, I feel like it would alleviate a lot of confusion and frustration for people.
    https://issues.jenkins.io/browse/JENKINS-56586
    Steven Foster
    @sfosterdw
    something something difficult to visualize properly is my recollection
    Jerry Wiltse
    @solvingj
    Yes the blue ocean support is a related discussion, but specifically it would be great if we could find some link to public discussion or explanation about it.
    And, it seems like the lack of visualization support in a third-party plugin like blue ocean should not block the core capabilities of pipeline from doing something so extremely common and reasonable.
    Jerry Wiltse
    @solvingj
    @bitwiseman can you comment on how things have gone with SCRIPT_SPLITTING_TRANSFORMATION in https://issues.jenkins.io/browse/JENKINS-37984
    Do you think it will ever be something like a first class global option in GUI, or a per-job setting, or even something that could become the default?
    Also, is there a way to see how big the bytecode is?
    Jerry Wiltse
    @solvingj
    I'm trying to figure out how to refactor my code so that it doesn't blow up the bytecode size, but I don't know how to figure out what's contributing the most.
    Also, is there a way to use the when{}block with scripted pipeline? The reason is that in declarative pipeline, when when{} returns false, blue ocean shows the stage, but as NOT EXECUTED status. I want this same thing in scripted.
    I guess it's here: https://issues.jenkins.io/browse/JENKINS-47286
    Postponed in 2019?
    Jerry Wiltse
    @solvingj
    everywhere I turn on this issue is a dead end. can't make this pipeline fit as declarative because it's too big. Can't get the equivalent visualization in scripted pipeline because the logic of when{} block and marking stage status as skipped is only possible via declarative.
    Cant' refactor the biggest things out of declarative because stages in declarative have to be in-lined, can't reference named closures.
    Jerry Wiltse
    @solvingj
    I found this, but it's a bit sad IMO: https://comquent.de/en/skipped-stages-in-jenkins-scripted-pipeline/
    It definitely feels like something I would say "do not do that"
    Jerry Wiltse
    @solvingj
    Is it possible the pipeline-as-Yaml plugin would suffer the same Bytecode size limitation , or would it be likely to avoid this problem?
    Steven Foster
    @sfosterdw
    I designed a pipeline using multiple jobs as a means for extra parallelism, and that probably helped avoid this bytecode problem
    Jerry Wiltse
    @solvingj
    Thank you @sfosterdw : I don't have that option at this time, but I appreciate the response.
    cruizji
    @cruizji
    Hello, I am Junior with Jenkins, I am working with a Pipeline to deploy terraform but I can´t pass jenkins variables to terraform, Could you help me, please?
    1 reply
    Girdhar Singh Rathore
    @girdhar-singh-rathore

    Hi,

    I'm using wirteYaml pipeline utility ref: https://www.jenkins.io/doc/pipeline/steps/pipeline-utility-steps/#writeyaml-write-a-yaml-from-an-object-or-objects

    I'm facing very strange issue, when i write the object it put the aliases and anchors

    charts:

    • &id001
      chartName: kuku1
      releaseName: kuku-release
    • &id002
      chartName: kuku2
      releaseName: mupi-nf-sanity
      charts:
    • *id001
    • *id002
    • chartName: kuku-3

    how to avoid this anchors and aliases
    Any help would be greatly appreciated

    Liam Newman
    @bitwiseman
    @solvingj > everywhere I turn on this issue is a dead end.
    Believe me I know how you feel.
    Liam Newman
    @bitwiseman
    In order for script splitting to become a first class citizen will require a commitment of time and effort that I have not had and am unlikely to have any time soon. Script splitting also has some known limitations which cannot be worked around (def variables not allowed) . It might be possible to make it a per pipeline feature enabled by a directive. I'd be happy to help someone else do that but I do not have the bandwidth myself.
    @solvingj As to nested Parallel blocks - again, this might be something that could be implemented now. There's no functional limitation that I can think of. It would just need - again - someone to commit the resources to raising it to the community, socializing the idea, getting buy in, making the change, adding sufficient tests, updating the docs, etc, etc.
    Srees100
    @Srees100
    Hi all
    How to create a pipeline to move the code from bitbucket to s3. On top of that s3 code how can i build glue job. One more thing the files present in bitbucket needs to be as it is in the s3 bucket...
    kannanap
    @kannanap
    Hi Team
    I need urgent help i am getting the below timeout error while running Multi-pipeline job in jenkins
    2022-02-02 14:44:25 :aem-frontend:digi-leap:test started...
    2022-02-02 14:44:25 :aem-frontend:digi-leap:test finished, took 0ms
    Cancelling nested steps due to timeout
    Sending interrupt signal to process
    After 20s process did not stop
    Console output :
    ==============
    2022-02-02 14:44:25 :aem-frontend:digi-leap:test started...
    2022-02-02 14:44:25 :aem-frontend:digi-leap:test finished, took 0ms
    Cancelling nested steps due to timeout
    Sending interrupt signal to process
    After 20s process did not stop
    some one can help me to resolve this issue pls
    kannanap
    @kannanap
    xxf
    farimah
    @fanaei_gitlab
    Hi Team ,
    I have a quick question about having 2 code review application in 1 jenkins pipeline. There is a request to leverage both SonarQube and AWS CodeGuru on code review part of a pipeline.
    I was wondering if it is even possible in Jenkins pipeline ? or if there can be any better solution rather than that?
    Thanks
    Praveen v
    @praveen216:matrix.org
    [m]
    Hi Team, I'm trying to load a list of accounts in Jenkins job parameters which will come from txt/csv file. Is there any way to achieve this? Thank you!!