Red Hat Bugzilla – Bug 877407
[Cartridge] "Node execution failure" when creating app by --enable-jenkins
Last modified: 2017-03-08 12:34 EST
Created attachment 646320 [details] mcollective.log and production.log Description of problem: If not having jenkins server created and run rhc app create php3 php -pjia --enable-jenkins It will pIrompt the error "Node execution failure". Check the php app again, find jenkins-client is not embedded, jenkins server is created, if embed jenkins-client again, It could be added successfully Version-Release number of selected component (if applicable): http://download.lab.bos.redhat.com/rel-eng/OpenShiftEnterprise/1.0/2012-11-15.2/ How reproducible: always Steps to Reproduce 1.Make sure not having jenkins app. 2.rhc app create php3 php -pjia --enable-jenkins 3.rhc app show -pjia 4.rhc cartridge add jenkins-client -pjia -a php3 Actual results: Step 2: Jenkins client failed to install - Node execution failure (error getting result from node). If the problem persists please contact Red Hat support. Step 3: [jason@jasontest ~]$ rhc app show -pjia php3 php3 @ http://php3-jia.example.com/ =================================== Application Info ================ Gear Size = small Created = 7:08 AM SSH URL = ssh://185dc0f4c6cc44c681604a29846c4c68@php3-jia.example.com UUID = 185dc0f4c6cc44c681604a29846c4c68 Git URL = ssh://185dc0f4c6cc44c681604a29846c4c68@php3-jia.example.com/~/git/php3.git/ Cartridges ========== php-5.3 Step 4: Success Expected results: Step 2: Success, no such error This issue doesn't exist on Online product
Not sure we're seeing quite the same thing, but as far as I can tell, it doesn't work whether there's a Jenkins gear there already or not. In either case, the new app is created, silently failing to add the jenkins-client cart (I don't see an error, the jenkins-client cart simply isn't listed in the output). If you create the Jenkins gear separately you can add the jenkins-client cart to your app just fine after the app is created, so that's a workaround. [demo@broker ~]$ rhc app create phpjenk3 php --enable-jenkins Creating application 'phpjenk3' =============================== Cartridge: php Gear Size: default Scaling: no Namespace: demo [...] Application Info ================ Gear Size = small Created = 2:00 PM UUID = 724a6328e68e41a5bc687c958c14690a SSH URL = ssh://724a6328e68e41a5bc687c958c14690a@phpjenk3-demo.lmeyer1117.example.com Git URL = ssh://724a6328e68e41a5bc687c958c14690a@phpjenk3-demo.lmeyer1117.example.com/~/git/phpjenk3.git/ Cartridges ========== php-5.3 RESULT: Application phpjenk3 was created. [demo@broker ~]$ rhc cartridge add jenkins-client-1.4 -a phpjenk3 Password: ******** Adding 'jenkins-client-1.4' to application 'phpjenk3' Success jenkins-client-1.4 ================== Properties ========== Job URL = https://jenkins-demo.lmeyer1117.example.com/job/phpjenk3-build/
I have tried several times, find that in the bad network or high load environment, the mcollective timeout is short. Then it will reproduce this bug. In the file /etc/openshift/plugins.d/openshift-origin-msg-broker-mcollective.conf, update MCOLLECTIVE_TIMEOUT=100 or bigger, then the bug can't be reproduced. Then I think the bug is duplicated with bug 876465.
This configuration change has been merged upstream. https://github.com/openshift/origin-server/pull/972
Well, now I can't create jenkins server when use "--jenkins-enalbe" option. [root@broker ~]# rhc app create php php --enable-jenkins -pjia Creating application 'php' ========================== Scaling: no Namespace: jia Gear Size: default Cartridge: php Your application's domain name is being propagated worldwide (this might take a minute)... The authenticity of host 'php-jia.example.com (10.4.59.139)' can't be established. RSA key fingerprint is 6c:a5:e5:fa:75:db:5a:7f:dc:a2:44:ed:e4:97:af:3c. Are you sure you want to continue connecting (yes/no)? yes Initialized empty Git repository in /root/php/.git/ done php @ http://php-jia.example.com/ ================================= Application Info ================ SSH URL = ssh://eb7b40d51cd047f989aae08d30112195@php-jia.example.com UUID = eb7b40d51cd047f989aae08d30112195 Gear Size = small Git URL = ssh://eb7b40d51cd047f989aae08d30112195@php-jia.example.com/~/git/php.git/ Created = 10:12 PM Cartridges ========== php-5.3 RESULT: Application php was created. [root@broker ~]# I think it's same with bug 878407. Could you check the code please? Thanks. Then I will verify this bug.
Addition information: Version: http://buildvm-devops.usersys.redhat.com/puddle/build/OpenShiftEnterprise/1.1.x/2012-12-09.1 rhc 1.0.6
Hm, In comment (In reply to comment #5), I mentioned bug 878407, it's for rhc-client, and has fixed in upstream. But this bug is for node. So let the bug be "ON_QA" again...
Verify: [root@broker ~]# gem install commander -v 4.1.2 Fetching: commander-4.1.2.gem (100%) Successfully installed commander-4.1.2 1 gem installed Installing ri documentation for commander-4.1.2... Installing RDoc documentation for commander-4.1.2... [root@broker ~]# [root@broker ~]# [root@broker ~]# [root@broker ~]# [root@broker ~]# [root@broker ~]# rhc app create php1 php --enable-jenkins -pjia Creating application 'php1' =========================== Cartridge: php Scaling: no Namespace: jia1 Gear Size: default Jenkins domain name is being propagated worldwide (this might take a minute)... Your application's domain name is being propagated worldwide (this might take a minute)... The authenticity of host 'php1-jia1.example.com (10.4.59.139)' can't be established. RSA key fingerprint is 6c:a5:e5:fa:75:db:5a:7f:dc:a2:44:ed:e4:97:af:3c. Are you sure you want to continue connecting (yes/no)? yes Initialized empty Git repository in /root/php1/.git/ done php1 @ http://php1-jia1.example.com/ ==================================== Application Info ================ UUID = 0c594fc9c5444d2b9c379dd6d87cd1fc SSH URL = ssh://0c594fc9c5444d2b9c379dd6d87cd1fc@php1-jia1.example.com Git URL = ssh://0c594fc9c5444d2b9c379dd6d87cd1fc@php1-jia1.example.com/~/git/php1.git/ Gear Size = small Created = 2:03 AM Cartridges ========== php-5.3 jenkins-client-1.4 = https://jenkins-jia1.example.com/job/php1-build/ RESULT: Application php1 was created. Application jenkins was created. Jenkins created successfully. Please make note of these credentials: User: admin Password: mSgbVkHRWbeW Note: You can change your password at: https://jenkins-jia1.example.com/me/configure Version: http://buildvm-devops.usersys.redhat.com/puddle/build/OpenShiftEnterprise/1.1.x/2012-12-09.1
Do we need to ship a new version of rubygem-commander-4.0.3-4.el6op.noarch in order for this fix to work?
No need to ship a new version as far as I know, the fix make it compatible with both old commander and new commander, rebase should be ok. This is the bug cloned from upstream: https://bugzilla.redhat.com/show_bug.cgi?id=885587 This is the upstream fix: https://github.com/openshift/rhc/pull/233
Ahh, thanks for the clarification!
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. http://rhn.redhat.com/errata/RHSA-2013-0220.html