RHBA-2019:0405 - Bug Fix Advisory

Topic

Red Hat OpenShift Container Platform release 3.10.119 is now available with
updates to packages and images that fix several bugs.

Description

Red Hat OpenShift Container Platform is Red Hat's cloud computing
Kubernetes application platform solution designed for on-premise or private
cloud deployments.

This advisory contains the RPM packages for Red Hat OpenShift Container
Platform 3.10.119. See the following advisory for the container images for
this release:

https://access.redhat.com/errata/RHBA-2019:0404

This update fixes the following bugs:

• `oc delete --force` was not properly passed to the server. As a result, some resources could not be reclaimed or removed. Now, the `--force` option is properly handled by the server when removing an object. (BZ#1562961)
• Previously, there was no logging cap established by default upon docker installation, causing the host disk to fill completely with logs. A log cap was introduced in the inventory file, reducing the maximum size of logs. (BZ#1594678)
• Images were not pre-pulled when using docker, causing `CRIO` installs to time out. Now, `crictl pull` may be used when `CRIO` is specified as a runtime. (BZ#1639623)
• Playbooks ran a check to see if images existed on disks with version tags, but did not check if the version tag was up-to-date with the repository. If the image existed on a disk, the image pull was skipped. As a result, asynchronous updates would not install properly. Now, the disk check is removed and image pulls succeed to apply asynchronous updates. (BZ#1659600)
• Playbooks would fail with the error "The conditional check 'not openshift_is_atomic | bool' failed". This error is now corrected in the certificate expiry check and the playbooks succeed without errors. (BZ#1662730)
• Due to an API update in `CRIO`, `crictl` 1.10 would not work with older versions of the service. During upgrades from openshift-ansible 3.9, the `cri-tools` package was updated to 1.10 prior to image pre-pull tasks. Now, the pre-pull tasks are ignored, allowing the upgrade to proceed. Images are pulled during the upgrade after the `CRIO` service is upgraded. (BZ#1668649)

All OpenShift Container Platform 3.10 users are advised to upgrade to these
updated packages.

Solution

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

See the following documentation, which will be updated shortly for release 3.10.119, for important instructions on how to upgrade your cluster and fully apply this asynchronous errata update:

https://docs.openshift.com/container-platform/3.10/release_notes/ocp_3_10_release_notes.html

This update is available via the Red Hat Network. Details on how to use the
Red Hat Network to apply this update are available at
https://access.redhat.com/articles/11258.

Affected Products

• Red Hat OpenShift Container Platform 3.10 x86_64
• Red Hat OpenShift Container Platform for Power 3.10 ppc64le

Fixes

• BZ - 1537388 - Global flag --version is not respected
• BZ - 1562961 - Unable to force delete of zombie resources (including projects)
• BZ - 1584696 - oc bash completion not up to date
• BZ - 1594678 - Add default max-size when install docker
• BZ - 1615375 - Failed to create pod, due to CNI issues
• BZ - 1639623 - [3.10] Images are not prepulled when CRIO runtime is used
• BZ - 1659600 - Updating from 3.10.x to 3.10.72 does not pull in 3.10.72 images unless you specify openshift_image_tag=v3.10.72 in inventory
• BZ - 1662730 - While running playbook to check certificate expiry, it is failing with error- "'openshift_is_atomic' is undefined"
• BZ - 1668649 - [3.10]upgrade failed due to crio client and server mismatch
• BZ - 1669311 - A ovs process gets killed when oom-killer is invoked, leaving it in bad state.
• BZ - 1676614 - openshift-ansible fails on openstack without floating IPs
• BZ - 1676671 - Placeholder for OCP 3.10 image refresh
• BZ - 1676673 - Placeholder for OCP 3.11 Image builds
• BZ - 1676905 - [3.10] More consistent modification of journald.conf.
• BZ - 1678096 - Docker service hang
• BZ - 1678442 - While deploying logging in openshift cluster we are getting error while elasticsearch is initiating
• BZ - 1684460 - openshift_logging_curator_run_timezone is undefined

CVEs

(none)

References

(none)