Uploaded image for project: 'Jenkins'
  1. Jenkins
  2. JENKINS-44193

wfapi/describe execNode empty - I need to query which node a stage runs on via rest call.

XMLWordPrintable

      We use pipeline on a scale of 64+ nodes. We utilize the wfapi/describe endpoint extensively to track and report stage execution to a custom server. When a stage fails or takes too long, I am required to go from our reporting server over to Jenkins then goto the logs of the run then search for a known substring of our server naming standard so I can figure out where I need to ssh / debug issues or what grafana page I need to view to analyze resource utilization.

       

      Please add the value of execNode in the wfapi/describe OR another endpoint. I understand some stages run on master & the default value can remain empty.

       

      I find JENKINS-35634 appears to ask the same, but it is closed as a dup of JENKINS-26138, which seems unrelated.

       

       

      ex:

      {
"_links":
{
"self":
{
"href":"/job/job_name/44/execution/node/16/wfapi/describe"}
}
,"id":"16","name":"Setup for build","execNode":"","status":"SUCCESS","startTimeMillis":1494429943607,"durationMillis":320,"pauseDurationMillis":0}]
}

          [JENKINS-44193] wfapi/describe execNode empty - I need to query which node a stage runs on via rest call.

          James Sandlin created issue -
          James Sandlin made changes -
          Summary Original: wfapi/describe execNode empty New: wfapi/describe execNode empty - I need to query which node a stage runs on via rest call.
          James Sandlin made changes -
          Description Original: We use pipeline on a scale of 64+ nodes. We utilize the wfapi/describe endpoint extensively to track and report stage execution to a custom server. When a stage fails or takes too long, I am required to go from our reporting server over to Jenkins then goto the logs of the run then search for a known substring of our server naming standard so I can figure out where I need to ssh / debug issues or what grafana page I need to view to analyze resource utilization.

           

          Please add the value of execNode in the wfapi/describe OR another endpoint. I understand some stages run on master & the default value can remain empty.

                      		New: We use pipeline on a scale of 64+ nodes. We utilize the wfapi/describe endpoint extensively to track and report stage execution to a custom server. When a stage fails or takes too long, I am required to go from our reporting server over to Jenkins then goto the logs of the run then search for a known substring of our server naming standard so I can figure out where I need to ssh / debug issues or what grafana page I need to view to analyze resource utilization.

           

          Please add the value of execNode in the wfapi/describe OR another endpoint. I understand some stages run on master & the default value can remain empty.

           

          I find JENKINS-35634 appears to ask the same, but it is closed as a dup of JENKINS-26138, which seems unrelated.

           
          James Sandlin made changes -
          Description Original: We use pipeline on a scale of 64+ nodes. We utilize the wfapi/describe endpoint extensively to track and report stage execution to a custom server. When a stage fails or takes too long, I am required to go from our reporting server over to Jenkins then goto the logs of the run then search for a known substring of our server naming standard so I can figure out where I need to ssh / debug issues or what grafana page I need to view to analyze resource utilization.

           

          Please add the value of execNode in the wfapi/describe OR another endpoint. I understand some stages run on master & the default value can remain empty.

           

          I find JENKINS-35634 appears to ask the same, but it is closed as a dup of JENKINS-26138, which seems unrelated.

                      		New: We use pipeline on a scale of 64+ nodes. We utilize the wfapi/describe endpoint extensively to track and report stage execution to a custom server. When a stage fails or takes too long, I am required to go from our reporting server over to Jenkins then goto the logs of the run then search for a known substring of our server naming standard so I can figure out where I need to ssh / debug issues or what grafana page I need to view to analyze resource utilization.

           

          Please add the value of execNode in the wfapi/describe OR another endpoint. I understand some stages run on master & the default value can remain empty.

           

          I find -JENKINS-35634- appears to ask the same, but it is closed as a dup of JENKINS-26138, which seems unrelated.

           

           

          ex:
          {code:java}
          {
          "_links":
          {
          "self":
          {
          "href":"/job/job_name/44/execution/node/16/wfapi/describe"}
          }
          ,"id":"16","name":"Setup for build","execNode":"","status":"SUCCESS","startTimeMillis":1494429943607,"durationMillis":320,"pauseDurationMillis":0}]
          }{code}
          Jesse Glick made changes -
          Component/s New: pipeline-stage-view-plugin [ 21139 ]
          Component/s Original: workflow-api-plugin [ 21711 ]

          Ferruccio Bongianni added a comment -

          Hi,

          I have the very same problem described in this issue, but there hasn't been any activity on this. Has anyone looked at this?

          Thanks

          Ferruccio Bongianni added a comment - Hi, I have the very same problem described in this issue, but there hasn't been any activity on this. Has anyone looked at this? Thanks

          Hannes Kayser added a comment -

          Hi,

          we need the same information since we switch nodes within a build for better load balancing.

          Is there any progress on this issue?

          Thanks

          Hannes

          Hannes Kayser added a comment - Hi, we need the same information since we switch nodes within a build for better load balancing. Is there any progress on this issue? Thanks Hannes

          Connor Newton added a comment -

          I would also like this to be able to create an agent monitor showing what builds/stages are currently executing there.

          Connor Newton added a comment - I would also like this to be able to create an agent monitor showing what builds/stages are currently executing there.

          Connor Newton added a comment - - edited

          It appears that https://github.com/jenkinsci/pipeline-stage-view-plugin/pull/59 has been proposed to fix this issue but a required PR https://github.com/jenkinsci/workflow-api-plugin/pull/77 has been rejected because the maintainer reckons the equivalent functionality is available another way.

          I did a stab in the dark attempt to use that other functionality in the same way and didn't get any results
          https://github.com/jenkinsci/pipeline-stage-view-plugin/pull/59#issuecomment-1627753411

          In fact on the code path #59 proposes an update to unconditionally overriding execName to some constant value was not visible externally on any pipeline node /wfapi/describe. 

          Connor Newton added a comment - - edited It appears that https://github.com/jenkinsci/pipeline-stage-view-plugin/pull/59 has been proposed to fix this issue but a required PR https://github.com/jenkinsci/workflow-api-plugin/pull/77 has been rejected because the maintainer reckons the equivalent functionality is available another way. I did a stab in the dark attempt to use that other functionality in the same way and didn't get any results https://github.com/jenkinsci/pipeline-stage-view-plugin/pull/59#issuecomment-1627753411 In fact on the code path #59 proposes an update to unconditionally overriding execName to some constant value was not visible externally on any pipeline node /wfapi/describe . 

            Unassigned Unassigned
            greenscar James Sandlin
            Votes:
            9 Vote for this issue
            Watchers:
            9 Start watching this issue

              Created:
              Updated: