Re: pipeline groovy - can somebody please explain why variable value passing is working like this?
I'm just interested in this thread, so I'm subscribing by answering. Sorry for the confusion :) Am 14. November 2016 12:15:16 MEZ, schrieb Jonathan Hodgson <j.r.hodg...@gmail.com>: >I'm sorry, but am I supposed to know what that means? > >On Monday, November 14, 2016 at 10:25:12 AM UTC, Thomas Keller wrote: >> >> /sub > >-- >You received this message because you are subscribed to a topic in the >Google Groups "Jenkins Users" group. >To unsubscribe from this topic, visit >https://groups.google.com/d/topic/jenkinsci-users/j7lvrd4UqO8/unsubscribe. >To unsubscribe from this group and all its topics, send an email to >jenkinsci-users+unsubscr...@googlegroups.com. >To view this discussion on the web visit >https://groups.google.com/d/msgid/jenkinsci-users/d1b04553-3549-4312-b71f-680a8183e312%40googlegroups.com. >For more options, visit https://groups.google.com/d/optout. -- Sent by my Droid -- You received this message because you are subscribed to the Google Groups "Jenkins Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to jenkinsci-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/jenkinsci-users/D6DE232E-EE9E-4355-BAEB-F867E20E27FC%40thomaskeller.biz. For more options, visit https://groups.google.com/d/optout.
Re: Multibranch Pipeline trigger
Ok, I figured it out myself. Several issues on my side: 1) In order to get the upstream trigger set up, the dependent project needs to be run at least once, otherwise Jenkins obviously does not know anything about the trigger in first place. 2) Absolute project references of course need to start with a slash, i.e. /repos/barrepo/master 3) If relative project references are used, you need to remember that - by default - the trigger code resolves these references relative to the _parent_ project of the build. So if you have foorepo/master and barrepo/master, then you need to use "../barrepo/master", and not "../../barrepo/master" or anything alike. -- You received this message because you are subscribed to the Google Groups "Jenkins Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to jenkinsci-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/jenkinsci-users/521d11dd-1075-4965-834c-b077b70933a5%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: Pipeline: Multiproject repository
Multibranch pipelines are really only suitable for single-project
multi-branch repos. The farest I got with a multi-project setup was using
the Jenkins Repo Plugin (for the Google Repo Tool), but this would reverse
the use case you're looking for: It pings all subprojects / subrepos and
kicks off a build if any of those changed; you'd then still have to figure
out manually what exactly would need to be build once your build has been
kicked off.
I however suspect you could achieve this by remembering the last revision a
specific subtree / repo was changed yourself (git log -1 path/to/subtree).
Am Montag, 14. November 2016 12:18:06 UTC+1 schrieb Sverre Moe:
>
> I need some help brainstorming a solution for multiproject repositories.
> Pipeline building a repository that contains several delivery projects.
>
> The root of the repository contains several delivery projects.
> - delivery-project
> - delivery1
> - CMakeList.txt
> - delivery-project-delivery1.spec
> - src
> - delivery2
> - CMakeList.txt
> - delivery-project-delivery2.spec
> - delivery3
> - CMakeList.txt
> - delivery-project-delivery3.spec
> - src
>
> I could keep one Jenkinsfile on the root. Then iterate through all
> directories and build each delivery.
> However when there is an SCM change in only one of the deliveries I only
> want that one to be built.
> This could be solved if I could determine which directory has been
> changed. Something like this (ignore closure which doesn't work in
> Pipeline):
> changedDirectories.each { directory ->
> dir(directory) {
> doBuild()
> }
> }
> Though I'm not sure how to get the SCM changes to determine this.
>
> Considering this delivery git repository will only have a master branch, a
> Multibranch Pipeline project is not necessary, and a standard Pipeline job
> would suffice.
>
> I could also create a Folder job for delivery-project, and one Pipeline
> job for each delivery. Each subproject could contain its own Jenkinsfile
> which would be specified in the configuration in Script Path as
> delivery1/Jenkinsfile. However when a new delivery project is added to the
> repository it would require me to manually add a new Pipeline job for it.
>
--
You received this message because you are subscribed to the Google Groups
"Jenkins Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to jenkinsci-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/jenkinsci-users/d5abe033-2f3b-4b79-a55c-d97f59e78260%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Re: pipeline groovy - can somebody please explain why variable value passing is working like this?
/sub -- You received this message because you are subscribed to the Google Groups "Jenkins Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to jenkinsci-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/jenkinsci-users/1c095dc2-2586-43aa-a006-e9991f85c044%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: Getting "java.io.NotSerializableException" exception in Jenkins pipeline, groovy.
IIRC objects used in pipelines need to be able to get serialized, because the build should be pausable and resumable at any time. Try to refactor the AntBuilder object creation and call into a groovy method and annotate this method with @NonCPS -- You received this message because you are subscribed to the Google Groups "Jenkins Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to jenkinsci-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/jenkinsci-users/470eba97-6898-49e8-b273-214bb3cfc1aa%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: Multibranch Pipeline trigger
Has anybody an idea? -- You received this message because you are subscribed to the Google Groups "Jenkins Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to jenkinsci-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/jenkinsci-users/9fee361d-574f-4eda-b117-9a7d21c82d6d%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: Multibranch Pipeline trigger
*bump* -- You received this message because you are subscribed to the Google Groups "Jenkins Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to jenkinsci-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/jenkinsci-users/32f57b52-f8cb-4d72-8bd4-840603e73f02%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Multibranch Pipeline trigger
Hey all!
I have multiple library projects that I build as multibranch pipelines.
Naturally these library projects have dependencies and usually fetch these
dependencies as a SNAPSHOT artifact from a previous build of the other
library.
For the case where a such a dependency gets a change, I want to trigger a
new build for all dependent library builds. These are my Jenkinsfile's for
the two repos:
* in foo.git/Jenkinsfile
#!/usr/bin/env groovy
node {
stage('Test') {
checkout scm
sh "( sleep 5 && echo foo )"
}
}
* in bar.git/Jenkinsfile:
#!/usr/bin/env groovy
properties([
pipelineTriggers([
upstream(
threshold: hudson.model.Result.SUCCESS,
upstreamProjects: 'repos/foo/master'
)
])
])
node {
stage('Test') {
checkout scm
sh "( sleep 5 && echo bar )"
}
}
Both Jenkins projects are saved underneath the directory "repos". Now I
tried several versions how to reference "upstreamProjects", with relative
paths "../../foo/master", "../foo/master" or simply "foo/master", or with
the absolute path like above, but all without success: the "bar.git"
project is just never triggered.
What am I doing wrong? Is this even a supported use case and if not, what
are my alternatives given the above setup?
Thanks in advance,
Thomas.
--
You received this message because you are subscribed to the Google Groups
"Jenkins Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to jenkinsci-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/jenkinsci-users/6fc22699-13bc-4814-a49d-f8fef30fd206%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Re: Automated install of JDK doesn't seem to work on a Mac
Hi, Following the game-of-life tutorial setup description I was confused. I finally figured out you don't want Jenkins to install Maven or JDK if you have the OS X developer tools installed. Just check java --version and mvn --version for the home directory and put that into the appropriate HOME field in your configuration. cheers, pdx-tom On Friday, December 30, 2011 2:48:52 AM UTC-8, sweeney wrote: Hi, we recently added a couple of new nodes to our Jenkins cluster; a 64- bit W2K8 box and a Mac mini server, in addition to the CentOS master and four W2K8 32-bit blades. Up to now we have been managing the development build stack manually on the existing nodes, but as our environment is becoming more heterogenous, I thought I'd try the automatic tool install feature to see if that would simplify administration of the cluster. I've had mixed results. Ant downloads and installs fine everywhere, so the basic mechanism seems to be sound. However, I've not had so much luck with the JDK. It downloads and installs fine on CentOS and W2K8 32-bit with UAC disabled. On the 64-bit node, UAC is still enabled, and the install fails: Building remotely on hudson3 Installing JDK jdk-6u29-oth-JPR Downloading JDK from http://download.oracle.com/otn/java/jdk/6u29-b11/jdk-6u29-windows-x64.exe Downloading 70687000bytes Installing S:\Hudson\tools\jdk1.6.0_29_auto_\jdk.exe [jdk1.6.0_29_auto_] $ S:\Hudson\tools\jdk1.6.0_29_auto_\jdk.exe /s /v / qn /L \S:\Hudson\tools\jdk1.6.0_29_auto_\jdk.exe.install.log\ REBOOT=ReallySuppress INSTALLDIR=\S:\Hudson\tools\jdk1.6.0_29_auto_\ java.io.IOException: Cannot run program S:\Hudson\tools \jdk1.6.0_29_auto_\jdk.exe (in directory S:\Hudson\tools \jdk1.6.0_29_auto_): CreateProcess error=740, The requested operation requires elevation This wasn't unexpected, and since it works with UAC disabled on the 32- bit nodes I have an obvious workround. On the mac, however, there seems to be a more fundamental problem: Building remotely on 10.5.0.200 JDK installation skipped: Unknown CPU name: mac os x Updating svn://luke/configuration-management/autotools U pom.xml At revision 82488 JDK installation skipped: Unknown CPU name: mac os x [autotools] $ /Users/jenkins/slave/tools/Ant_1.8.2/bin/ant -version Error: JAVA_HOME is not defined correctly. We cannot execute /Users/jenkins/slave/tools/jdk1.6.0_29_auto_/bin/ java Build step 'Invoke Ant' marked build as failure Finished: FAILURE 'uname -a' reports Darwin iOS-Build-Server.local 11.2.0 Darwin Kernel Version 11.2.0: Tue Aug 9 20:54:00 PDT 2011; root:xnu-1699.24.8~1/ RELEASE_X86_64 x86_64 on this node. The machine has Xcode installed (and not GCC), if that makes any difference. -- You received this message because you are subscribed to the Google Groups Jenkins Users group. To unsubscribe from this group and stop receiving emails from it, send an email to jenkinsci-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.
Jenkins Job list empty after JENKINS_HOME was moved
Hi all! I moved our Jenkins instance from a java -jar-based installation to one using Tomcat as servlet container. Jenkins comes up as it should and according to catalina.out it finds its Home directory at the correct place (and all files in the home directory are owned by the tomcat user), still, the jobs are not loaded and apparently other things like installed plugins neither. Is there any way to debug this issue? Thanks in advance, Thomas. -- You received this message because you are subscribed to the Google Groups Jenkins Users group. To unsubscribe from this group and stop receiving emails from it, send an email to jenkinsci-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.
Re: Jenkins Job list empty after JENKINS_HOME was moved
Sorry, I forgot, but this is Jenkins 1.524. Am Montag, 29. Juli 2013 18:01:08 UTC+2 schrieb Thomas Keller: Hi all! I moved our Jenkins instance from a java -jar-based installation to one using Tomcat as servlet container. Jenkins comes up as it should and according to catalina.out it finds its Home directory at the correct place (and all files in the home directory are owned by the tomcat user), still, the jobs are not loaded and apparently other things like installed plugins neither. Is there any way to debug this issue? Thanks in advance, Thomas. -- You received this message because you are subscribed to the Google Groups Jenkins Users group. To unsubscribe from this group and stop receiving emails from it, send an email to jenkinsci-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.
