Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
    Artem Stasiuk
    @terma
    hi
    Artem Stasiuk
    @terma
    marekhanzlik
    @marekhanzlik
    Hi, does the tagging of ec2 instances support adding label as a tag?
    Arindom01
    @Arindom01
    Hi, adding labels to Fleet agent in an ASG does not really work. I assigned a fleet agent "nonprodslave2" label but when I actually launch the job I get this "‘Jenkins’ doesn’t have label ‘nonprodslave2’; ‘user2’ lacks permission to run on ‘nonprodslave2 i-07*6e’
    Somehow the plugin is adding the instance_id at the end of the label which fails in "Role-Strategy" plugin
    Logged an issue in the support board
    Artem Stasiuk
    @terma
    @marekhanzlik sorry, about delay, but answer nope, it doesn't, just cloud name, however it possible, could you share use case and create issue about that https://github.com/jenkinsci/ec2-fleet-plugin/issues/new thx
    Tianheng Zhou
    @tianheng-ayla
    I'm seeing the plugin keeps changing my autoscaling group setting to protect from scale in
    Tianheng Zhou
    @tianheng-ayla
             // without scale in protection auto scaling group could terminate random ec2 instances
                        // in case of scale in, instead we need to enable so plugin could decide
                        // which empty instance should be terminated
                        .withNewInstancesProtectedFromScaleIn(true));
    I see you have this in the code, is this meaning I should disable the asg auto scaling policy and let your plugin to determine when to scale down?
    Artem Stasiuk
    @terma
    Yep, that's idea of that plugin, to auto manager Jenkins capacity based on amount of tasks, instead of indirect CPU or mem usage.
    Tianheng Zhou
    @tianheng-ayla
    I see, good to know.
    Narz Marbeth David
    @narzdavid

    I'm testing out the EC2 spot fleet behavior currently. So, I configured my fleet to downscale after work hours to 0. The scheduled action will modify the min and max capacity to 0 at 9pm.

    That works fine if Jenkins does not modify the spot request. However as I increased the build queue, the plugin seems to send the following:

    image.png
    I noticed that the fleet will no longer downscale unless I manually reset it back to excessCapacityTerminationPolicy: Default using the aws cli
    Is there something I could configure in the plugin so it doesn't override my scheduled scaling action?
    Stefan Goethals
    @zipkid
    Hi, i am testing this module but am running into jenkinsci/ec2-fleet-plugin#265. Is there any work happening to fix this?
    Serhii Matsak
    @sergeymatsak87
    Hi all. Did anybody use ec2-fleet jenkins plugin to spin up ec2 spot instances in aws? We are trying to find out the root cause related to interruption (JenkinsFleet …eced was marked offline: Node is being removed). On AWS side spots are terminated with status instance-terminated-by-user . Would appreciate for the quick feedback - raised an issue to github jenkinsci/ec2-fleet-plugin#337