Is multiple triggers an option?
We have a huge pipeline that is currently set with a webhook trigger defined in a pipelineTriggers section in the properties section. However, sometimes we would like some of these jobs to ONLY run nightly CI builds. I'm trying to avoid having a second version of this huge pipeline script where the only difference is the piplineTrigger section. In a perfect world I'd have a Boolean parm for CI that when checked, the job used a cron type trigger and when unchecked, it used the web hook trigger... but I have seen anything like that. any help would be appreciated. Thanks, Rusty -- 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/CACZq5i5bJb_Cjwx-iP3wHWiqQTvNGbjU2xXxahLKkf0HsxcgcA%40mail.gmail.com.
Can't create New Item
I am running Jenkins version 2.303.3. we recently moved the jenkins master from jdk 8 to jdk 11. However, since that move, we are unable to add any new projects or folders thru the GUI. when the "New Item" link is clicked, Jenkins just produces a blank page. Anyone else experienced this behavior? any help would be appreciated. Thanks, Rusty Sapper -- 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/CACZq5i40uN-Ds9zzMMkJoeMzoTVJ1F38FPf2p-Uh%3DGdfSaxShg%40mail.gmail.com.
Node preference not working
I have several cloud templates defined, each pointing to a different physical node that can run our build containers. I have several templates defined under each cloud(that spin up the build containers). I have a common label defined on each template across the clouds. I have defined a node preference score on each template. However, now matter what I set the preference score to, Jenkins always uses the first cloud and template defined. it will use that template until it reaches the instance cap. then moves to the next one. This is not good utilization of the resources. Should the node preference score be working? is there any other way to control how Jenkins utilizes the container templates? Linux installation. jenkins version 2.303.1 docker plugin version 1.2.3 thanks for any help. -Rusty -- 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/CACZq5i56eTf%2BCzfUHUVV6WCJ4Wf%2B9tvX%2B4aFAZ8bVcP0ymQUMg%40mail.gmail.com.
Remote client doesn't restart after boot
I have a situation where Jenkins starts a job on a remote test machine, and as part of the test, the system must boot into an operating system that cannot have networking, well, working. The test machine runs some stuff (15 minutes to hours or days (or even weeks), then boots back to Linux. However, Jenkins has a tendency to not restart the Java client after the test host boots Linux. (And you can't start the agent yourself, you must let Jenkins do that, since it talks to that agent via stdin/out.) Settings for the host: # executors: 1 Launch via ssh Max # retries 0 Seconds to wait between retries 30 I'm using pipeline. First it uses master node to do setup, the switches to test host node to do some more setup. Finally it switches back to master, registers a web hook, then switches back to test host to do the stage-specific stuff and then reboot. Then switch back to master, wait for web hook. So, that's the pipeline script in english-ish. Once the test finishes and the test host is in Linux, Jenkins doesn't seem to notice. So I go to the node status page in Jenkins and hit "relaunch agent" - it starts just fine and everything continues like designed. But that's a manual step which isn't acceptable. So, either I need to get Jenkins to relaunch the agent itself, or I need a URL on Jenkins server that I can curl to to get it relaunched. Or something. Any ideas? Thanks!!! -- 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/8d48e3c0-efe8-46a5-a754-0e3b49c4335f%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
I'm trying to learn Jenkins Pipeline - pointers to good programmer documentation? Or is Jenkins World the place I should go????
I've read the "getting started with pipeline" pages (https://jenkins.io/doc/book/pipeline/getting-started/). I've spent literally hours looking for something really helpful, in one place, for someone who wants to write CODE in the pipeline script. I did just find https://jenkins.io/doc/book/pipeline/syntax/, which is a great step in the direction I was looking for. However, an update. I've been searching around and gotten a lot further, but I want more, quicker. I should mention that I've been a software developer for many years (oh, what, 40 or so? C, C++ primarily, but a little Java), I teach a class called "Linux Operating System" at a community college (have been for 3 years or so), I write scripts (bash, python, perl, etc) all the time, so I hope I have more than a small clue about 'things'. I note that Jenkins World is coming soon to SF. Perhaps that's the best place to learn/etc... Here's the deal. I'm trying to Jenkins-ize our test infrastructure. Right now it is all done via shell scripts and ssh. However, it's a monolith. You set your configuration for each test host (in a pretty big config file), start the run, and wait (up to more than a day! (sometimes multiple days)) for the test run to complete. When the test run completes, a script runs which generates a report, including of what tests ran, which of those that ran failed, and so forth (NOTE that just because a single test fails does NOT mean we want to kill the entire test run). Oh, yes - I should mention that the test run involves MULTIPLE test hosts, each running basically independently of each other. Once a test host finishes its test run, it is available for another test run. Every day we build our system and that triggers a full test run, as defined by a config file (which defines which hosts to run, and the test config file to use for that host - each host can have a different test configuration. Indeed, it can have a completely different product being tested). To do a test run on a host, we first run some scripts on the server which sets up the global environment (we monitor serial port activity and store the output in a log file in the test report directory, for example), then we set up stuff on the test host (which starts out running Linux) for the test to come. Right now, MOST of our tests are run under FreeDOS (don't ask - we currently have no choice in that matter), then when those tests finish the host reboots into Linux, runs any Linux-based tests, and then runs 'finish', which copies the logs to the server, creates the test report, and emails it to a specified group. For one thing, we'd like to have better visibility into the progress of the test (even possibly with partial results), and the ability to stop the testing at any stage in the process fairly easily. So, I'm working on re-doing it all with Jenkins driving things. Right now I've got one pipeline which will use a config file to specify which test hosts (and configuration) to start running - its more or less done (reads config file, parses, does what it says). I've got another pipeline which runs the 'setup global environment' script, and which I've just proven works fine, as far as it goes. No scripts AFTER that have been implemented... I'm working on implementing the rest of the thing (both the 'actually start the run' part that happens after the global environment setup, and all the different stages. Right now I've got a pipeline which parses another config file and runs the scripts specified (and their command-line args). It kind of works too, but I'm thinking that the command-line args are way too unfriendly for 'normal' users. So, I'm halfway competent (which means at least 50% INcompetent!), but feel like if I knew a bit more I'd not rush down a path that I'll have to redo later. Are there any good documents to help me move more into the competent phase? As I asked above, is Jenkins World the best place (bang for buck)? Thanks! Rusty -- 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/BY2PR0401MB09661B1CC7B181FFBF46F0129DB30%40BY2PR0401MB0966.namprd04.prod.outlook.com. For more options, visit https://groups.google.com/d/optout.
Any pointers on docs for using Jenkins to control remote jobs on various machines, non-java.
t take over as above it probably isn't much use, as the system works (mostly) as we need right now). *Can Jenkins do this? Is Jenkins the completely wrong tool for this? Should I be looking for something else?Any pointers greatly appreciated!* Thanks! Rusty -- 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/a1481c89-ac67-42ca-a84c-2dd04b2f9768%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
PATH being ignored on SLES 10 with Jenkins ver. 1.466.2
I am running into an interesting problem with jenkins on SLES 10.4. I have
two different python versions installed because I need python2.6 and not
2.4 for portability reasons that are not important.
Obviosly, I set up path variables on the machine so that jenkins would
build using python2.6 and not 2.4 I have set a node variable
PATH
/usr/local/bin/python:/usr/local/bin/:$PATH:/usr/local/bin/python:/usr/local/bin/
I have set the .bashrc file
I have set the /etc/profile
yet the path returned, even after a machine restart,disconnect-reconnect or
both, jenkins always returns the path...
PATH /usr/bin:/bin:/usr/sbin:/sbin
But running a simple job of :
import os
import sys
print sys.executable
print os.getenv("PATH", None)
print sys.path
import automate
Returns
*23:11:38* [sysn0t4] $ python /tmp/hudson3083762089535439523.py*23:11:38*
/usr/local/bin/python*23:11:38*
/usr/local/bin/python:/usr/local/bin/:/usr/local/bin/python:/usr/local/bin/:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin/python:/usr/local/bin/:/usr/local/bin/python:/usr/local/bin/*23:11:38*
['/tmp', '/usr/lib/python24.zip', '/usr/lib64/python2.4',
'/usr/lib64/python2.4/plat-linux2', '/usr/lib64/python2.4/lib-tk',
'/usr/lib64/python2.4/lib-dynload', '/usr/lib64/python2.4/site-packages',
What on earth is going on acording to the path returned.
/usr/local/bin/python (python2.6) should have been found first, but it
isn't. How do I insert python 2.6 onto the path correctly?
--
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: Build with for loop like finesse
Unfortunatly, after attempting the said method.. jenkins still ended up attempting both in parallel. It looks like that method has a race condition where it only checks at the beginning of the build if any downstream projects are building; which is a problem if any builds next to the build end up executing anything in the downstream. Adding a wait period to one of the builds will fix the race condition, but that is super hacky in my opinion. On Wednesday, August 15, 2012 5:15:53 AM UTC-6, cjo wrote: > > Yes, apart from the issue I mentioned, haven't had a chance to look > through existing issues yet to see if it is reported. > > > On Tuesday, August 14, 2012 9:21:21 PM UTC+1, rusty wrote: >> >> so. f I understand this completly >> If I "Block build when downstream project is building" Jenkins will >> resolve all of the downstream builds recursively allowing all children as >> well as grandchildren projects to be built before building project 1 again >> with different parameters? >> >> On Tuesday, August 14, 2012 11:31:36 AM UTC-6, cjo wrote: >>> >>> >>> If all of the builds are triggered by project 1, and you only want to >>> run one set of parameters at a time through the chain at a time. >>> You should be able to use the "Block build when downstream project is >>> building" and "Block build when upstream project is building" advanced >>> options on all of the projects which will enforce this behaviour. >>> >>> Doing this will allow you to add Project1 to the queue multiple times >>> with different parameters but only have one run for each set of parameters. >>> Warning: We found an issue with this if a downstream project is blocked >>> (waiting for a resource/slave to come online) which will allow Project 1 to >>> run again. >>> (Due to fix for https://issues.jenkins-ci.org/browse/JENKINS-8929) >>> >>> We ran this for several months without problem and in some cases had >>> 5-10 instances of Project 1 queued with different parameters. >>> This was needed as we had a dependency between multiple projects needing >>> to share workspaces, due to the SCM and size of the project we were using >>> along with the 30 different build configurations. >>> >>> Chris >>> >>> On Tuesday, August 14, 2012 5:54:20 PM UTC+1, rusty wrote: >>>> >>>> The problem is that I can have no concurrency in between the two top >>>> projects nor any child projects that they have. I have looked at throttle >>>> builds plugin, but it doesn't provide what I need which is full atomicity >>>> on the top level builds and all recursive underlying builds. >>>> >>>> On Tuesday, August 14, 2012 10:37:05 AM UTC-6, sti wrote: >>>>> >>>>> All the builds will enter the queue at the same time, but how many >>>>> will actually run in parallel depends on how many executors you have or >>>>> if >>>>> you have used other means of concurrency control like Throttle Concurrent >>>>> Builds plugin. >>>>> >>>>> -- Sami >>>>> >>>>> rusty kirjoitti 14.8.2012 >>>>> kello 19.26: >>>>> >>>>> >>>>> I am already using a multi-configuration project, I think that the >>>>> disconnect is stemming from how do I use multiple axis to accompolish >>>>> this. >>>>> >>>>> Won't adding another axis start the two projects in parallel? I can't >>>>> have two projects in paralelll... >>>>> >>>>> On Tuesday, August 14, 2012 3:06:38 AM UTC-6, sti wrote: >>>>>> >>>>>> Use a multi-configuration project with one, two or more axes. >>>>>> >>>>>> -- Sami >>>>>> >>>>>> rusty kirjoitti 14.8.2012 >>>>>> kello 5.39: >>>>>> >>>>>>
Re: Build with for loop like finesse
so. f I understand this completly If I "Block build when downstream project is building" Jenkins will resolve all of the downstream builds recursively allowing all children as well as grandchildren projects to be built before building project 1 again with different parameters? On Tuesday, August 14, 2012 11:31:36 AM UTC-6, cjo wrote: > > > If all of the builds are triggered by project 1, and you only want to run > one set of parameters at a time through the chain at a time. > You should be able to use the "Block build when downstream project is > building" and "Block build when upstream project is building" advanced > options on all of the projects which will enforce this behaviour. > > Doing this will allow you to add Project1 to the queue multiple times with > different parameters but only have one run for each set of parameters. > Warning: We found an issue with this if a downstream project is blocked > (waiting for a resource/slave to come online) which will allow Project 1 to > run again. > (Due to fix for https://issues.jenkins-ci.org/browse/JENKINS-8929) > > We ran this for several months without problem and in some cases had 5-10 > instances of Project 1 queued with different parameters. > This was needed as we had a dependency between multiple projects needing > to share workspaces, due to the SCM and size of the project we were using > along with the 30 different build configurations. > > Chris > > On Tuesday, August 14, 2012 5:54:20 PM UTC+1, rusty wrote: >> >> The problem is that I can have no concurrency in between the two top >> projects nor any child projects that they have. I have looked at throttle >> builds plugin, but it doesn't provide what I need which is full atomicity >> on the top level builds and all recursive underlying builds. >> >> On Tuesday, August 14, 2012 10:37:05 AM UTC-6, sti wrote: >>> >>> All the builds will enter the queue at the same time, but how many will >>> actually run in parallel depends on how many executors you have or if you >>> have used other means of concurrency control like Throttle Concurrent >>> Builds plugin. >>> >>> -- Sami >>> >>> rusty kirjoitti 14.8.2012 >>> kello 19.26: >>> >>> >>> I am already using a multi-configuration project, I think that the >>> disconnect is stemming from how do I use multiple axis to accompolish this. >>> >>> Won't adding another axis start the two projects in parallel? I can't >>> have two projects in paralelll... >>> >>> On Tuesday, August 14, 2012 3:06:38 AM UTC-6, sti wrote: >>>> >>>> Use a multi-configuration project with one, two or more axes. >>>> >>>> -- Sami >>>> >>>> rusty kirjoitti 14.8.2012 >>>> kello 5.39: >>>> >>>>
Re: Build with for loop like finesse
The problem is that I can have no concurrency in between the two top projects nor any child projects that they have. I have looked at throttle builds plugin, but it doesn't provide what I need which is full atomicity on the top level builds and all recursive underlying builds. On Tuesday, August 14, 2012 10:37:05 AM UTC-6, sti wrote: > > All the builds will enter the queue at the same time, but how many will > actually run in parallel depends on how many executors you have or if you > have used other means of concurrency control like Throttle Concurrent > Builds plugin. > > -- Sami > > rusty > kirjoitti > 14.8.2012 kello 19.26: > > > I am already using a multi-configuration project, I think that the > disconnect is stemming from how do I use multiple axis to accompolish this. > > Won't adding another axis start the two projects in parallel? I can't > have two projects in paralelll... > > On Tuesday, August 14, 2012 3:06:38 AM UTC-6, sti wrote: >> >> Use a multi-configuration project with one, two or more axes. >> >> -- Sami >> >> rusty kirjoitti 14.8.2012 >> kello 5.39: >> >>
Re: Build with for loop like finesse
I am already using a multi-configuration project, I think that the disconnect is stemming from how do I use multiple axis to accompolish this. Won't adding another axis start the two projects in parallel? I can't have two projects in paralelll... On Tuesday, August 14, 2012 3:06:38 AM UTC-6, sti wrote: > > Use a multi-configuration project with one, two or more axes. > > -- Sami > > rusty > kirjoitti > 14.8.2012 kello 5.39: > >
Build with for loop like finesse
PROBLEM: I need to build a project with certain parameters which then kicks of various other projects. I then need to build the same first project with a different set of parameters as soon as the full build pipeline finishes from the original build. eg: project1 with first set of parameters: project2: project3: project4: project5: project6: project7: project1 with second set of parameters: project2: project3: project4: project5: project6: project7: Note that all children projects depend on parameters passed down from project1 and that due to lack of hardware, I cannot build the two different project1's in parallel. QUESTION: How do I accomplish top level builds that spawn of a plethora of children projects and then be able to spawn another top level build when the current top level build, and all children builds including project6 from my example, has finished?
