RHBA-2016:1937 - Bug Fix Advisory

Topic

An updated Jenkins image that fixes several bugs and adds enhancements is now available for Red Hat OpenShift Enterprise 3.1 and 3.2.

Description

OpenShift Enterprise by Red Hat is the company's cloud computing Platform-as-a-Service (PaaS) solution designed for on-premise or private cloud deployments.

This update fixes the following bugs:

• The OpenShift Pipeline Plug-in for Jenkins did not properly handle permission issues when attempting to access the route for Jenkins. Any of the steps or post-build actions in Jenkins jobs could fail unnecessarily with a Java stack trace citing an inability to access routes. This bug fix adds correct error handling when attempting to list routes. As a result, Jenkins jobs no longer fail unnecessarily if the token used by the OpenShift Pipeline Plug-in for Jenkins to access the OpenShift API Master did not have permission to list routes. (BZ#1369746)
• Some Jenkins plug-in files did not have their ownership changed to non-root during the construction of the OpenShift Jenkins RHEL image because the chown did not traverse symbolic links. The build of the Jenkins master image resulting from calling 'oc new-app' on the sample jenkins master template would fail with a permission problem reported by chgrp. This bug fix Updates the Dockerfile used to construct the OpenShift Jenkins RHEL images to change the ownership of the plug-in files accessed by symbolic links. As a result, running 'oc new-app' on the example jenkins master template succeeds. (BZ#1365767)
• A few plug-ins brought in by Jenkins core were not getting pinned during construction of the OpenShift Jenkins Image for RHEL. Some plug-ins brought in by Jenkins core were not at the version levels expected in the OpenShift Jenkins Image for RHEL. This bug fix adds the pinning of plug-ins in question wihiin the Dockerfile used to construct the OpenShift Jenkins Image for RHEL. As a result, all plug-ins in the OpenShift Jenkins Image for RHEL are at the expected version levels. (BZ#1369680)
• The build step in OpenShift Pipeline Plug-in for Jenkins did not fully check during a certain wait loop for the status of builds running in OpenShift. Jobs in Jenkins with OpenShift Pipeline Plug-in builds steps would wait the full timeout duration unnecessarily before failing a given Jenkins job if the corresponding build in OpenShift failed much sooner. Properly check the OpenShift build status during the wait polling from the Jenkins job leveraging the OpenShift Pipeline Plug-in for the Jenkins build step. Jenkins jobs will fail immediately if a corresponding build in OpenShift fails. (BZ#1369740)

In addition, this update adds the following enhancement:

• Deployments with prehooks and liveness or readiness probes were typically taking more than 1 minute. Therefore, the default deployment timeout was changed from 1 minute to 10 minutes in the OpenShift Pipeline Plug-in for Jenkins. As a result, users of the OpenShift Pipeline Plug-in for Jenkins had to change the deployment timeout setting less often. (BZ#1368372)

This update includes the following image:

openshift3/jenkins-1-rhel7:1.651.2-11

All OpenShift Enterprise 3 users are advised to upgrade to this updated image.

Solution

Before applying this update, make sure all previously released errata relevant to your system have been applied.

The Red Hat Enterprise Linux container images provided by this update can be downloaded from the Red Hat Container Registry at registry.access.redhat.com using the "docker pull" command. Dockerfiles and scripts should be amended either to refer to this new image specifically, or to the latest image generally.

To update the default image stream for this image, see the following documentation after logging into the OpenShift CLI as a cluster administrator:

https://docs.openshift.com/enterprise/3.2/install_config/upgrading/manual_upgrades.html#importing-the-latest-images

To determine whether containers in pods will pull the updated image
automatically, review the following documentation on image pull policies:

https://docs.openshift.com/enterprise/3.2/dev_guide/managing_images.html#image-pull-policy

After applying this update, if you are using persistent storage with the Jenkins image, delete the existing Jenkins configuration from the persistent storage location. Then pull the new image to all OpenShift nodes in your cluster before restarting any pods running the old image version. If you do not delete your existing configuration, the new job configuration included in this image will not be utilized.

Affected Products

• Red Hat OpenShift Container Platform 3.1 x86_64

Fixes

• BZ - 1347849 - Update openshift jenkins image
• BZ - 1365767 - Failed to build Jenkins master
• BZ - 1368372 - [RFE] Wait interval 60000 (60s) is too short for openshiftDeploy stage in jenkins master image
• BZ - 1369680 - Jenkins some plugins version are too old in jenkins image
• BZ - 1369702 - Cannot cancel openshift build when setting post-build in jenkins
• BZ - 1369725 - Build step 'tag openshift image' in webpage is close to right in jenkins webpage
• BZ - 1369740 - Pipeline build should be failed when related build fail
• BZ - 1369746 - Fail to delete resource using `openShiftDeleteResourceByJsonYaml` pipeline dsl step

CVEs

(none)

References

(none)