RHBA-2021:3729 - Bug Fix Advisory

Topic

Updated glusterfs packages that add an enhancement and fix several bugs are now available for Red Hat Gluster Storage 3.5 on Red Hat Enterprise Linux 8.

Description

Red Hat Gluster Storage is a software only scale-out storage solution that provides flexible and affordable unstructured data storage. It unifies data storage and infrastructure, increases performance, and improves availability and manageability to meet enterprise-level storage challenges.

This advisory adds the following enhancement:

• Previously, when a client gets disconnected and reconnected, it reopens the saved FDs with subsequent operations on that. Hence, causes problems in applications where POSIX lock on FD is mandatory to write to that file. With this update, a new option "client.strict-locks" is added, when enabled it prevents reopening of saved FDs after reconnecting if POSIX locks are held on them. As a result, subsequent writes on these FDs failed as this is necessary for stricter lock compliance as bricks cleanup any granted locks when a client disconnects. All the servers and clients must upgrade to RHGS-3.5.5 for this to be in effect and the "cluster.op-version" must set as "70200". This option sets OFF by default. (BZ#1689375)

Bug Fix(es):

• Previously, nfs-ganesha silently failed with no indication to the user if the gluster_shared_storage volume does not mount while attempting to start nfs-ganesha. With this update, additional logic is added to the ganesha HA setup script to check the gluster_shared_storage volume is mounted, and to fail with an error if it is not mounted. As a result, the gluster CLI must report that the startup of nfs-ganesah has failed and the user can take steps to resolve the issue. (BZ#1918018)
• Previously in AFR, the fsync FOP does not have the logic of releasing the active client's eager-lock on the inode when there is a conflicting request from another client. As a result, no call-bailing of FOPS due to starved locks on RHHI-V workloads which are fsync heavy, the blocked inodelks can potentially lead to VM pauses. With this update, it is fixed by making AFR fsync aware of conflicting locks so that it releases the active lock if there is a contention. (BZ#1943467)

Users of glusterfs with Red Hat Gluster Storage are advised to upgrade to these updated packages.

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 8 x86_64
• Red Hat Enterprise Linux for x86_64 8 x86_64

Fixes

• BZ - 1600379 - Inconsistent custom xattr on backend directories after bringing back the down sub-volumes
• BZ - 1782428 - Unlinking the file with open fd, is leading to unexpected behavior with sharding enabled
• BZ - 1815462 - [RHEL 8.1] Migration error may appear during remove-brick
• BZ - 1889966 - Volume status shows rebalance as not started on doing a replace-brick on dist-arbiter volume
• BZ - 1891403 - [GSS] gluster block hosting volume free space doesn't matches between heketi and gluster.
• BZ - 1903911 - schedule_georep.py resulting in failure towards the end with python3
• BZ - 1919132 - md-cache incorrectly caches some xattrs as missing when they are not
• BZ - 1925425 - Tracker bug for important sharding patches to move into downstream
• BZ - 1927411 - [Ganesha] "ls" returning "ls: reading directory .: Invalid argument" on 2 clients when creates,deletes and ll were running in parallel
• BZ - 1928676 - rpcsvc: Add latency tracking for rpc programs
• BZ - 1942816 - Avoid dict_del logs in non distributed environment
• BZ - 1943467 - AFR does not release eager locks for fsync when there is contention
• BZ - 1946171 - Change self-heal-window-size to 1MB by default
• BZ - 1954028 - [RHGS Based RHEL-8.4] yum install redhat-storage-server will fail with redhat-logos package conflict
• BZ - 1955461 - RHVH 4.4.6: There are gluster related AVC denied errors in audit.log after upgrade
• BZ - 1957191 - Geo-replication gets delayed when it processes a lot of renames
• BZ - 1957641 - Rebalance doesn't migrate some sparse files
• BZ - 1992987 - [RHEL8.4] Upgrade to RHEL8.5 fails due to redhat-logos package conflict

CVEs

(none)

References

(none)