RHBA-2015:0682 - Bug Fix Advisory

Topic

Updated glusterfs, glusterfs-geo-replication, glusterfs-rdma, and glusterfs-server packages that fix numerous bugs and add various enhancements are now available for Red Hat Storage 3.

Description

Red Hat Storage is software-only, scale-out storage that provides flexible
and affordable unstructured data storage for an enterprise. GlusterFS, a
key building block of Red Hat Storage, is based on a stackable user-space
design and can deliver exceptional performance for diverse workloads.
GlusterFS aggregates various storage servers over network interconnections
into one large, parallel network file system.

This update fixes numerous bugs and adds various enhancements in RDMA,
Snapshot, NFS, Quota, and Geo-replication. Space precludes documenting all of these changes in this advisory. Users are directed to the
Red Hat Storage 3 Technical Notes, linked to in the References
section, for information on the most significant of these changes.

This advisory adds the following enhancements:

• Red Hat Storage supports Just a Bunch Of Disks (JBOD) on servers with up to 24 disks.Complexity of RAID is avoided by using 3-way replication with JBOD.
• Synchronous three-way replication is now fully supported in Red Hat Storage volumes. 3-way replicated volumes do not require hardware or software RAID. Three-way replication yields best results when used in conjunction with JBODs. Quorum functionality on three-way replicated volumes helps to prevent split-brain scenarios. 3-way replicated volumes provide additional resiliency in public cloud deployments as well.(BZ#1184075)
• Multi-threaded epoll is introduced which enables events to be distributed and improves the performance due the parallel processing of requests/responses received. (BZ#1104459)
• With this fix, the CLI command logs are now dumped into the 'cmd_history.log' file. Now, this file is a non-hidden file.(BZ#1181051)

All users of Red Hat Storage are advised to install these updated packages.

Solution

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

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/site/articles/11258

Affected Products

• Red Hat Enterprise Linux Server 6 x86_64
• Red Hat Enterprise Linux Server 5 x86_64
• Red Hat Gluster Storage Server for On-premise 3 for RHEL 6 x86_64
• Red Hat Gluster Storage Nagios Server 3 for RHEL 6 x86_64

Fixes

• BZ - 987511 - gluster-cli: Format the output of 'gluster pool list' properly
• BZ - 1035540 - Typo in the error message reported when creating the volume with bricks on "/"
• BZ - 1035647 - Migration of files to the newly added bricks were skipped from the bricks which were 100% filled.
• BZ - 1090910 - xml output needed for geo-rep CLI commands
• BZ - 1104459 - RFC: make epoll multithreaded
• BZ - 1110730 - brick process crashed when rebalance and rename was in progress
• BZ - 1128156 - ssh AuthorizedKeyFiles are assumed to be in $HOME/.ssh/authorized_keys ignoring any configuration from /etc/ssh/sshd_config
• BZ - 1132920 - "gluster volume set help" for server.statedump-path has wrong description
• BZ - 1136714 - DHT + rebalance :- DATA LOSS - while file is in migration, creation of Hard-link and unlink of original file ends in data loss(both files are missing from mount and backend
• BZ - 1139104 - quota: rename of "dir" fails in case of quota space availability is around 1GB
• BZ - 1144015 - gluster accepts invalid values when changing cluster.min-free-disk option
• BZ - 1144117 - dist-geo-rep: zero byte files created in slave even before creating any data in master by processing stale changelogs in working dir
• BZ - 1161111 - gluster vol set testvol features.uss command accepts invalid random values
• BZ - 1162306 - Directories not visible anymore after add-brick, new brick dirs not part of old bricks
• BZ - 1162573 - DHT + rebalance :- file permission got changed after file migration
• BZ - 1164906 - Geo-replication changelog crawl causes cpu spike - creates lot of stale changelog files
• BZ - 1167593 - quota: EDQUOT when trying to rename a file back to original name
• BZ - 1177134 - glusterd: when there is loss in server quorum then it should block all operation
• BZ - 1180560 - [SNAPSHOT]: Snapshot restore fails after adding a node to master with geo-replication involved
• BZ - 1182554 - [SNAPSHOT]: In a n-way replica volume, snapshot should not be taken, even if one brick is down.
• BZ - 1183920 - cli crashes when listing quota limits with xml output
• BZ - 1184075 - [rfe] 3-Way Replication support in AFR with Quorum enabled
• BZ - 1186127 - Performance enhancement for RDMA
• BZ - 1186657 - Inode infinite loop leads to glusterfsd segfault
• BZ - 1188522 - entry, metadata self-heal in 3.0 and 3.1 are not compatible
• BZ - 1188685 - Current Volume status command doesn't show RDMA port for tcp,rdma type volume
• BZ - 1189871 - remove samba-glusterfs dependency
• BZ - 1193995 - [RHEV-RHS] Fuse mount process crashed, while using gluster volume as storage domain in RHEV
• BZ - 1194138 - When RDMA init fails, the resources are not cleaned up properly
• BZ - 1194166 - epoll: configuration default for {client,server}.event-threads option not honoured
• BZ - 1194275 - [Snapshot]: Glusterd crashed during snap restore after adding a new node without hostname resolution
• BZ - 1194525 - Perf: Crash seen during IOZone performance regression runs.
• BZ - 1194574 - [geo-rep]: In a mountbroker setup, set_geo_rep_pem_keys.sh fails to copy keys to its slave peers.
• BZ - 1194605 - DHT + epoll : client crashed
• BZ - 1195378 - Perf: DHT errors filling logs when perf tests are run.
• BZ - 1196029 - [dht]: Failed to rebalance files when a replica-brick-set was removed
• BZ - 1196098 - CTDB: Dependency errors while upgrading to ctdb2.5.4 and ISO spin failed for RHS3.0.4
• BZ - 1196607 - Any op on files in the root directory of the volume fails unless absolute path is specified.
• BZ - 1199441 - BVT: Samba crashed while running fs-sanity tests on gluster cifs mount
• BZ - 1200246 - libglusterfs event library does not compile in RHEL5
• BZ - 1200660 - CIFS:[USS]: .snaps is not visible or accessible from the Windows Client with Samba 4.1 even if the "show-snapshot-directory" option is set on the volume
• BZ - 1200981 - RHSS 3.0u3 to RHSS 3.0u4 upgrade using ISO source tree fails due to samba related dependency issues
• BZ - 1201613 - RDMA: Crash seen during smallfile read test.
• BZ - 1201628 - Tune dirty_ratio and dirty_background_ratio for RHS
• BZ - 1201633 - [epoll+Snapshot] : Snapd crashed while trying to list snaps under .snaps folder
• BZ - 1202541 - Perf: Metadata operation performance on replicated volumes regressed on 3.0.4 builds.
• BZ - 1203172 - SMB: Iozone running for too long and hung after 13 hours on dis-rep volume accessed as samba share.

CVEs

(none)

References

• http://www.redhat.com/security/updates/classification/#normal