RHBA-2018:1259 - Bug Fix Advisory

Topic

An update is now available for Red Hat Ceph Storage 3.0.

Description

Red Hat Ceph Storage is a scalable, open, software-defined storage platform
that combines the most stable version of the Ceph storage system with a
Ceph management platform, deployment utilities, and support services.

Bug Fixes:

• Previously, an attempt to delete a large RBD image with the "object map" feature enabled could cause the OSD nodes to trigger the "suicide_timeout" and self-terminate. With this update, deleting large RBD images with "object map" no longer causes OSDs to crash. (BZ#1325322)
• Previously, Metadata Server (MDS) daemons could get behind on trimming for large metadata workloads in larger clusters. With this update, MDS no longer gets behind on trimming for large metadata workloads. (BZ#1507629)
• Previously in the Ceph File System, Ceph Metadata Server (MDS) daemons would sometimes crash when scrubbing encountered an inode that had been renamed. With this update, MDS daemons no longer crash and scrubbing continues when encountering renamed inodes.(BZ#1518730)
• Previously, deployment of only one Ceph iSCSI Gateway node was supported because SSL certificates were only generated for one Ceph iSCSI Gateway node when using the "ceph-ansible" utility. This issue has been fixed, the certificates are distributed across all the gateways and more than one iSCSI Gateway node can now be deployed. (BZ#1540845)
• Previously, the Ceph File System (CephFS) client metadata capability trimming would sometimes fail assertions due to logic errors, and the CephFS client would abort. With this update, the logic error has been corrected and the CephFS client completes trimming. (BZ#1541424)
• Previously, when using "ceph-ansible" with the "copy_admin_key" option set to "true", the administrator's keyring would not copy to the other nodes in the Ceph Storage Cluster. With this update, the "copy_admin_key" option works as expected when set to "true". (BZ#1544720)
• Previously, an underlying issue caused the placement group (PG) log to grow without bound in certain situations. This caused OSDs to crash and stalled startup. With this release, the "trim-pg-log" operation has been added to "ceph-objectstore-tool" to allow offline trimming of large PG logs. This restores the PG log within the designated size limits. (BZ#1552094)
• Previously, a placement group (PG) that received only unsuccessful writes—such as deleting a non-existing object—would grow its write operations log indefinitely. This caused the OSDs handling that PG to run out of memory and crash. With this fix, the PG log is now trimmed as expected regardless of whether a write is successful or not, and the OSDs do not run out of memory. (BZ#1554544)

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 Ceph Storage MON 3 x86_64
• Red Hat Ceph Storage OSD 3 x86_64

Fixes

• BZ - 1325322 - OSD hit suicide timeout or OSD goes down and comes back, When deleting large RBD images with features: striping, exclusive, object map
• BZ - 1489866 - ensure that multiple CephContexts in a process image don't crash the program
• BZ - 1493418 - Allow pybind/ceph_volume_client to get, put, and delete RADOS objects
• BZ - 1496544 - client cap recall can cause client caps to go below hard-coded minimum (100)
• BZ - 1499806 - [CephFS]:- Client_metadata not populated in session info after client eviction and reconnect
• BZ - 1501958 - [CephFS]:- Cluster ended up in "damaged" mds when subtree pinning is in progress and tried to do mds failover
• BZ - 1506438 - [RHCeph 3.0 / 12.2.1-26.el7cp] rbd/fsx on bluestore: FAILED assert(p != extents.end())
• BZ - 1507136 - monitor gives constant "is now active in filesystem cephfs as rank" cluster log info messages
• BZ - 1507629 - mds gets significantly behind on trimming while creating millions of files
• BZ - 1509358 - [RHCS 3.y] Change default of rocksdb to no compression
• BZ - 1515337 - [rbd] rbd du on empty pool does not return proper output
• BZ - 1538317 - "rbd snap rm --force" might corrupt snapshots of child images
• BZ - 1540845 - [Ceph-ansible] Failure on TASK [ceph-iscsi-gw : lock ssl file access to root only]
• BZ - 1542459 - nfs-ganesha: ensure FSAL_CEPH can look up inodes by filehandle when Inode is not in cache
• BZ - 1543879 - ceph_volume_client: sets invalid caps for existing ceph auth IDs with no MDS or OSD caps
• BZ - 1544680 - [RADOS]:- corrupting the snapsets from non-primary osds will not be detected as inconsistencies
• BZ - 1544720 - [ceph-ansible] [ceph-container] admin keyring is not copied on to the collocated nodes when copy_admin_key is set to true
• BZ - 1548067 - rebase ceph to 12.2.4
• BZ - 1552094 - ceph-objectstore-tool command to trim the pg log
• BZ - 1552327 - "set_fact docker_exec_cmd" assumes there will be mons, but does not use the external list of mons if provided
• BZ - 1553302 - rebase nfs-ganesha to latest 2.5
• BZ - 1554544 - pg log does not trim when write errors are added, potentially leading to OOM
• BZ - 1564232 - ceph-ansible unable to upgrade ceph-base when ceph-test is installed

CVEs

(none)

References

(none)