RHBA-2019:0663 - Bug Fix Advisory

Topic

Updated heketi packages that add an enhancement and fixes several bugs are now available for OpenShift Container Storage 3.11 Batch 2 Update.

Description

Heketi provides the Red Hat Gluster Storage volume lifecycle management. It creates the Red Hat Gluster Storage volumes dynamically and supports multiple Red Hat Gluster Storage clusters.

This advisory adds the following enhancement:

• Previously, Heketi would make an attempt to spread bricks in a replica-set across different zones but there was no way to specify that the bricks must be placed in different zones. Now, Heketi offers new configuration options to specify that the bricks are placed in different zones. If the option is enabled and there is not enough zones or enough space in different zones the volume creation will fail. This behavior can be enabled by setting 'user.heketi.zone-checking strict' in the volume options of the volume create API call or OpenShift storage class (BZ#1635204)

This advisory also fixes the following bugs:

• Previously, the "heketi-cli db check" diagnostic command failed to update the total number of pending operations. With this fix, the total pending operations count is reported correctly. (BZ#1667797)
• Previously, providing an unknown operation identifier to the `heketi-cli server operations info` command or triggering other errors on the server produced an improper error response with "invalid character" from Heketi-cli. With this fix, the Heketi-cli tool reports the correct error message. (BZ#1669935)
• Previously, if the "heketi-cli device delete" sub-command was run for a device that contained no bricks before the device was disabled and removed, the underlying volume group was removed but the device remained in the Heketi database. This led to devices that required the --force-forget flag to be removed cleanly. Now, Heketi does not change the state of the underlying device until the state of the device in Heketi is correct. (BZ#1665386)
• Previously, when the block hosting volume did not exist in gluster, Heketi could not automatically clean up certain failed or stale block volumes. With this fix, Heketi can clean up failed or stale block volumes automatically (BZ#1668181)
• Previously, the Heketi prometheus metrics endpoint did not report the number of block volumes. With this fix, the number of block volumes in the cluster is included in the metrics. (BZ#1589012)

Users of Heketi are advised to upgrade to these updated packages, which fixes these bugs and adds an enhancement.

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

Fixes

• BZ - 1589012 - [RFE] - Block-volume details not listed in heketi metrics
• BZ - 1665386 - heketi device deletion command is deleting vg when we are executing delete command without disabling and removing the device
• BZ - 1667797 - pendingoperations total count is always zero in "heketi-cli db check" output
• BZ - 1668181 - heketi fails to cleanup stale block entries
• BZ - 1669935 - heketi is giving inappropriate Error when id is not present for "heketi-cli server operations info id"
• BZ - 1680468 - Volume creation fails with HEKETI_ZONE_CHECKING set to "strict"

CVEs

(none)

References

(none)