Marián Labuda <mlabuda> updated the status of jira JBIDE-18454 to Closed
Failed QA, maybe OSE-2.0 and OSE-1.2 needn't the new jenkins packages. old jdk: java-1.7.0-openjdk-1.7.0.65-2.5.1.2 new jdk: java-1.7.0-openjdk-1.7.0.71-2.5.3.1 old jenkins packages: jenkins-1.509.1-1.el6op.noarch,jenkins-plugin-openshift-0.6.25-1.el6op.x86_64,openshift-origin-cartridge-jenkins-1.16.1-2.el6op.noarch New jenkins packages: jenkins-1.565.3-1.el6op.noarch.rpm,jenkins-plugin-openshift-0.6.25.1-0.el6op.x86_64.rpm , openshift-origin-cartridge-jenkins-1.16.2.1-1.el6op.noarch.rpm 1） OSE 6.6，old jdk，old jenkins packages Result: php build success using jenkins 2） OSE 6.6，new jdk，old jenkins packages Result: php build success using jenkins 3） OSE 6.6，old jdk，New jenkins packages Result: php build fails with the following error message WARNING: Cancelling build due to earlier exceptions com.openshift.client.InvalidCredentialsOpenShiftException: Your credentials are not authorized to access "https://broker.ose20-host.com.cn/broker/rest/user" 4） OSE 6.6，new jdk，New jenkins packages Result: php build fails with the following error message WARNING: Cancelling build due to earlier exceptions com.openshift.client.InvalidCredentialsOpenShiftException: Your credentials are not authorized to access "https://broker.ose20-host.com.cn/broker/rest/user"
Created attachment 945468 [details] jenkins log The jenkins log after 'build php app using jenkins'
Reading the above results there's a pattern of authorization problems when using the new jenkins packages. Are you sure that those are configured right? If so, we have a "new" problem either in jenkins packages or underlying openshift-java-client.
I also encounter this issue, when I downgrade all the jenkins and jenkins plugin to previous released version, it is working well. Obviously this is a regression bug introduced by new packages.
I cannot replicate this failure on RHEL 6.6. Here's my relevant packages: [root@broker ~]# rpm -qa | grep 'jenkins\|openjdk-1.7' openshift-origin-cartridge-jenkins-1.16.2.1-1.el6op.noarch jenkins-plugin-openshift-0.6.25.1-0.el6op.x86_64 openshift-origin-cartridge-jenkins-client-1.17.1-2.el6op.noarch java-1.7.0-openjdk-1.7.0.65-2.5.1.2.el6_5.x86_64 jenkins-1.565.3-1.el6op.noarch [root@broker pt1]# git push Counting objects: 5, done. Delta compression using up to 2 threads. Compressing objects: 100% (3/3), done. Writing objects: 100% (3/3), 284 bytes, done. Total 3 (delta 2), reused 0 (delta 0) remote: Executing Jenkins build. remote: remote: You can track your build at https://jenkins-jlent20.ose20zscl11.example.com/job/pt1-build remote: remote: ERROR - Couldn't schedule job remote: !!!!!!!! remote: Deployment Halted! remote: If the build failed before the deploy step, your previous remote: build is still running. Otherwise, your application may be remote: partially deployed or inaccessible. remote: Fix the build and try again. remote: !!!!!!!! remote: An error occurred executing 'gear postreceive' (exit code: 1) remote: Error message: Failed to execute: 'control post-receive' for /var/lib/openshift/543c2213c4ba086554000001/jenkins-client remote: remote: For more details about the problem, try running the command again with the '--trace' option. To ssh://543c2213c4ba086554000001.example.com/~/git/pt1.git/ 27bd3be..96b923e master -> master It looks like the jenkins job wasn't scheduled, but it in fact was. The reason the jenkins-client thinks it failed is because jenkins returned a "201 created" HTTP response instead of the expected 200 or 302: curl -s -w %{http_code} --output /dev/null -X POST --insecure https://${JENKINS_USERNAME}:${JENKINS_PASSWORD}@jenkins-jlent20.ose20zscl11.example.com/job/pt1-build/build 201 In the jenkins log, we can see that the build did in fact run: ... Oct 13, 2014 3:15:38 PM hudson.slaves.NodeProvisioner update INFO: pt1-build provisioningE successfully completed. We have now 2 computer(s) Oct 13, 2014 3:15:45 PM hudson.plugins.openshift.OpenShiftComputerLauncher launch INFO: Slave connected. Oct 13, 2014 3:16:04 PM hudson.model.Run execute INFO: pt1-build #1 main build action completed: SUCCESS I'll attach the full jenkins log in a moment.
Created attachment 946583 [details] jenkins log from jolamb test instance showing successful run
Yeah, I mentioned to Jason the other PR needed to fix the response code handling. (Don't have it handy but hopefully he does now)
Using the updated environment( was updated from old puddle), The issue reported in comment 7 still exists. Using new installation of ose-2.0 puddle-2014-10-13-3, I got same result as Comment 11. "ERROR - Couldn't schedule job ***" is so obvious error. I am waiting the other fixed PR. I will dig more about What's the difference between update Env and new installation Env.
Andre Dietisheim <adietish> updated the status of jira JBIDE-18454 to Reopened
Andre Dietisheim <adietish> updated the status of jira JBIDE-18454 to Resolved
Andre Dietisheim <adietish> updated the status of jira JBIDE-18454 to Closed
Just an update, I still can't figure out why the different 2.0 envs are seeing different behavior. The relevant packages are all the same, etc. Running some more tests today, since even though the issue this fix is supposed to addressed has probably passed for most users, this still needs to be sorted out for future jenkins updates.
Anping, can you tell me what puddle version of OSE 2.0 your test env was upgraded from? Without knowing that, I'm not sure I can reproduce your issue.
The prior version is ose-2.0-2014-10-06.2.
Today, I also can't reproduced the issue in the build which is upgraded from the GA build to puddle 2014-10-16.4; unlucky the Environment hit the issue have been removed. I suggest close it or downgrade it if no one can reproduced.
According to comment 23, this bug is not reproduced now, but still need update Cartridge-Version field in the manifest.yml file just like what we did in 2.1 (BZ#1153317)