Description of problem: After restarting master api server, new triggered builds will become to complete/failed/cancelled directly from new status, no running staus in openshift client. Works well before restart master. Version-Release number of selected component (if applicable): ci.dev.../openshift/jenkins-slave-nodejs-rhel7 965eb5e5766a ci.dev.../openshift/jenkins-slave-maven-rhel7 7c6bbfb9656e ci.dev.../openshift/jenkins-1-rhel7 bd3ab5d39d48 openshift v3.3.0.16 kubernetes v1.3.0+507d3a7 How reproducible: always Steps to Reproduce: 1.Create a jenkins pipeline app. 2.Trigger some builds from openshift client and jenkins webconsole 3.Restart master api server 4.Trigger some builds from openshift client and jenkins webconsole again. 5.Check builds from openshift client and jenkins webconsole Actual results: jenkinspipeline builds skip running status in openshift client Expected results: Should sync all build status between openshift client and jenkins webconsole. Additional info:
also tracking as: https://github.com/fabric8io/openshift-jenkins-sync-plugin/issues/99
This is a problem if the API server host is rebooted. If the API server process is stopped/started then it reconnects fine. This is a TCP timeout issue that affects oc client too (try it with `oc get pods -w` & then force reboot the API server - notice no timeout). I'm looking at implementing WS pings but not sure if API server supports pings (it should if it's spec compliant).
This is a fix in kubernetes-client - PR https://github.com/fabric8io/kubernetes-client/pull/517. Will release client after merged & update jenkins plugin with new version after that.
I've released openshift-jenkins-sync-plugin 0.0.13 (https://repo.jenkins-ci.org/releases/io/fabric8/jenkins/plugins/openshift-sync/0.0.13/) which should fix this issue. @bparees: Forgotten how to get the Jenkins image updated with this new version?
centos image is available w/ the updated plugin. rhel has not been rebuilt yet.
New jenkins-1 and jenkins-2 images have been built with the updated openshift-sync plugin.
Verified brew-pulp.../openshift3/jenkins-1-rhel7 6f476676a4a1 openshift-sync 0.0.14 openshift v3.4.0.19+346a31d kubernetes v1.4.0+776c994 etcd 3.1.0-rc.0 Actual result: After restart master api,new triggered pipeline builds could be running status, then change to complete.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2017:0066