RHEA-2018:2697 - Product Enhancement Advisory

Topic

Updated Red Hat OpenShift Container Storage packages that fix several bugs and add various enhancements are now available for Red Hat Gluster Storage 3.4 and OpenShift Container Platform 3.10.

Description

The Red Hat OpenShift Container Storage solution provides persistent storage service for OpenShift Containers and OpenShift Infrastructure services. It can be deployed on OpenShift cluster eliminating the need for an external storage cluster or an appliance. Optionally, it can be deployed as stand-alone in an independent mode. This solution is based on Red Hat Enterprise Linux 7.5 and Red Hat Gluster Storage 3.4

This advisory adds the following enhancements:

• With the release of OpenShift Container Platform 3.10, we are rebranding Container-Native Storage as Red Hat OpenShift Container Storage. Along with this rebranding, the terms describing the method of deployment have also been changed.
• Converged mode (formerly referred to as ‘Container Native Storage’ or ‘CNS’) refers to running storage services in a container, like any other service.
• Independent mode (formerly referred to as ‘Container Ready Storage’ or ‘CRS’) refers to running storage services outside the container platform as a standalone subsystem which gets consumed by OpenShift Container Platform.
• Enhancements and features available in the Red Hat Openshift Container Storage are now available in two separate guides - one for deployment and upgrade and another for administrative tasks
• Ansible is now the preferred mode of installation as it automates the process of installing Red Hat Openshift Container Storage in both converged and independent mode. Ansible is less error-prone when compared to manually executing numerous commands.
• Enhanced OCS monitoring and configuration visibility using the Prometheus framework has been included in this release. You can now obtain the statistics of GlusterFS PVs by setting up Prometheus and enabling volume metrics for glusterfs plugin.
• You can now set the ‘persistentVolumeReclaimPolicy’ in the storage class to reclaim a persistent volume. When this parameter is set to "Retain" the underlying persistent volume is retained even after the corresponding persistent volume claim is deleted. This ensures that data is not lost if you accidentally delete the PVCs.
• With this release, you can replace a block from a node that is out of resource or is faulty, to a new node.
• Arbiter volume supports all persistent volume types with better consistency and less disk space requirements. The arbitrated volume uses client quorum to compare this metadata with the metadata of the other nodes to ensure consistency in the volume and prevent split-brain conditions.
• With this release, you can do an Ansible-based uninstall (cleanup) of all resources and artifacts from the cluster. By using Ansible-based uninstall, any existing GlusterFS-related resources and configurations from a Red Hat OpenShift Container Storage deployment is removed.
• With this release, you can now generate a block configuration and then load the configuration to the block hosting volume.

This update also fixes the following bugs:

• Previously, if any glusterd to glusterd connection reconnected during a volume delete operation it could have led to a stale volume being present in the gluster pool. This stale volume led to the 'brick part of another brick' error during subsequent volume create operations. With this fix, subsequent volume create operations don't fail. (BZ#1484412)
• Even though glusterfs fuse client log level has been set to ERROR, due to an issue in glusterfs client, it logs Info|Warn|Error messages regardless of the log level value set in the client process. This issue has been fixed with latest glusterfs client release. (BZ#1501439)

Users of OpenShift Container Storage are advised to upgrade to these updated packages, which add these enhancements and fixes these bugs.

Solution

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

For details on how to apply this update, refer to:

https://access.redhat.com/articles/11258

Affected Products

• Red Hat Gluster Storage Server for On-premise 3 for RHEL 7 x86_64

Fixes

• BZ - 1426259 - [GSS] [RFE] Customer need the ability to set the persistentVolumeReclaimPolicy per storage class
• BZ - 1484412 - [GSS] Error while creating new volume in CNS "Brick may be containing or be contained by an existing brick"
• BZ - 1501439 - [GSS][Tracker] glusterfs plugin logs Warning|Info|Error messages even when --log-level=ERROR
• BZ - 1569492 - Red Hat OpenShift Container Storage 3.10 Release
• BZ - 1585911 - CNS- Advanced Installation-Multipathd service not started/enabled by default on glusterfs nodes

CVEs

(none)

References

(none)