Some of the jenkins jobs are stuck with the following message
"[EnvInject] - Loading node environment variables".
I tried restarting the slave and that fixed the problem and every thing was working fine for few builds but then it stopped working again.
Any help would be appreciated.
Try renaming the project.
I don't know why this works - I just read that this had worked for someone else, and it worked for me!
In my case helped to restart Jenkins agent on which the previous "normal" build has executed (probably Jenkins controller assigned builds to the same agent, but the agent had some problems).
Related
For my build via Groovy script (without using Docker or maven or k8s, just call another sh file), when triggered by timer, sometimes, it cannot finish with issue:
process apparently never started in jenkinsWs/myjob#tmp/durable-d6c09021
(running Jenkins temporarily with -Dorg.jenkinsci.plugins.durabletask.BourneShellScript.LAUNCH_DIAGNOSTICS=true might make the problem clearer)
Cannot contact Slave29: java.io.FileNotFoundException: File 'jenkinsWs/myjob#tmp/durable-d6c09021/output.txt' does not exist
But it always works when I run it manually. I've used monitor tools for my slave machine to track what happened, but nothing strange.
I've tried all solution from this but nohope.
Thanks in advance for any idea.
I've installed Docker plugin and was able to setup a test job to run on a slave container.
The test job is a basic loop with echo that runs for 15 minutes.
My problem is that the job never finishes. It runs for a few minutes and then the connection to the slave docker is lost and job gets stuck and fails.
After reading about this issue, I tried several suggested solutions, but it didn't help.
One of the things that can affect my issue is disabling DockerContainerWatchdog from running.
My problem is that I can't figure out how to disable DockerContainerWatchdog from running.
Any help will be greatly appreciated.
Thanks,
Boris.
Got it!
Run the following code in Jenkins' Script Console:
System.setProperty("com.nirima.jenkins.plugins.docker.DockerContainerWatchdog.enabled", "false")
Please note: This will be active only until Jenkins' next restart.
Over the weekend, my Jenkins slave was down. Now it is up again, I can SSH from and to the master, so I expect Jenkins should be able to restart the agent on that slave. It is configured to be started via SSH.
What happens instead is that the master seemingly can't connect to the slave, the log shows exactly nothing.
What can I do to get this unstuck?
I tried already temporarily disabling and reenabling the node.
Restarting the master helped. My problem sounded similar to this old Jenkins bug, the workaround worked for me.
I am using docker to run my Jenkins instance.
The version which we are using was 2.121. Since last Saturday after the container restart, all my pipeline got vanished. After several hit and trial, when I updgraded to version 1.121.1. All my job got visible again. But when I tried to run any jobs. It is not getting started, after clicking on the build button nothing is happening.
What should I do? any suggestion would be helpful. Thanks
I'm trying not (not hard enough it seems) to get our jenkins server to provision a jenkins-slave using docker.
I have installed the Docker-plugin and configured it according to the description on the page. I have also tested the connectivity and at least this part works.
I have also configured 1 label in the plugin and in my job. I even get a nice page showing me the connected jobs for this slave.
When I then try to start a build nothing really happens. A build is scheduled, but never started - (pending—Waiting for next available executor).
From the message it would seem like jenkins is not able to start the slave via docker....
I'm using docker 1.6.2 and the plugin is 0.10.1.
Any clue to what is going on would be much appreciated!
It seems the problem was that I had added the docker version in the plugin config. That is apparently a no-go according to this post