GlusterFS Release 3.10.0 mit neuen Features mehr Performance und Bugfixes

Das Open-Source verteilte Dateisystem, GlusterFS, wurde vor kurzem mit dem Meilenstein 3.10.0 veröffentlicht. Neben über 315 Bugfixes wurden auch die Performance und Usability verbessert.

GlusterFS 3.10.0 Features und Änderungen

Nachfolgend Stichpunktartig ein Überblick über die Änderungen im neuen Release.

• Brick multiplexing (# gluster volume set all cluster.brick-multiplex on)
• Display OP-Version from Clients (# gluster volume status <VOLNAME|all> clients)
• Support get maximum OP-Version (# gluster volume get all cluster.max-op-version)
• Anzeige der verbleibenden Zeit des Rebalance (# gluster volume rebalance <VOLNAME> status)
• Tier nun als eigener Service
• Statedump Support für gfapi basierte Apps (# gluster volume statedump <VOLNAME> client <HOST>:<PID>)
• Trash Verzeichnis wird nicht mehr automatisch erstellt, bis die Funktion manuelle aktiviert wurde
• Parallel ReadDirp mit distribute xlator implementiert (# gluster volume set <VOLNAME> performance.readdir-ahead on und # gluster volume set <VOLNAME> performance.parallel-readdir on)
• md-cache ist optionall – ve cache security.ima xattr (# gluster volume set <VOLNAME> performance.cache-ima-xattrs on)
• CPU Support (SSE und AVX) für disperse Setups (# gluster volume set <VOLNAME> disperse.cpu-extensions <type>)

GlusterFS 3.10.0 Release Notes

Major changes and features

Brick multiplexing

Notes for users: Multiplexing reduces both port and memory usage. It does not improve performance vs. non-multiplexing except when memory is the limiting factor, though there are other related changes that improve performance overall (e.g. compared to 3.9).

Multiplexing is off by default. It can be enabled with

# gluster volume set all cluster.brick-multiplex on

Limitations: There are currently no tuning options for multiplexing – it’s all or nothing. This will change in the near future.

Known Issues: The only feature or combination of features known not to work with multiplexing is USS and SSL. Anyone using that combination should leave multiplexing off.

Support to display op-version information from clients

Notes for users: To get information on what op-version are supported by the clients, users can invoke the gluster volume status command for clients. Along with information on hostname, port, bytes read, bytes written and number of clients connected per brick, we now also get the op-version on which the respective clients operate. Following is the example usage:

# gluster volume status <VOLNAME|all> clients

Limitations:

Known Issues:

Support to get maximum op-version in a heterogeneous cluster

Notes for users: A heterogeneous cluster operates on a common op-version that can be supported across all the nodes in the trusted storage pool. Upon upgrade of the nodes in the cluster, the cluster might support a higher op-version. Users can retrieve the maximum op-version to which the cluster could be bumped up to by invoking the gluster volume get command on the newly introduced global option, cluster.max-op-version. The usage is as follows:

# gluster volume get all cluster.max-op-version

Limitations:

Known Issues:

Support for rebalance time to completion estimation

Notes for users: Users can now see approximately how much time the rebalance operation will take to complete across all nodes.

The estimated time left for rebalance to complete is displayed as part of the rebalance status. Use the command:

# gluster volume rebalance <VOLNAME> status

Limitations: The rebalance process calculates the time left based on the rate at while files are processed on the node and the total number of files on the brick which is determined using statfs. The limitations of this are:

• A single fs partition must host only one brick. Multiple bricks on the same fs partition will cause the statfs results to be invalid.
• The estimates are dynamic and are recalculated every time the rebalance status command is invoked.The estimates become more accurate over time so short running rebalance operations may not benefit.

Known Issues: As glusterfs does not stored the number of files on the brick, we use statfs to guess the number. The .glusterfs directory contents can significantly skew this number and affect the calculated estimates.

Separation of tier as its own service

Notes for users: This change is to move the management of the tier daemon into the gluster service framework, thereby improving it stability and manageability by the service framework.

This has no change to any of the tier commands or user facing interfaces and operations.

Limitations:

Known Issues:

Statedump support for gfapi based applications

Notes for users: gfapi based applications now can dump state information for better trouble shooting of issues. A statedump can be triggered in two ways:

1. by executing the following on one of the Gluster servers,

   # gluster volume statedump <VOLNAME> client <HOST>:<PID>

• <VOLNAME> should be replaced by the name of the volume
• <HOST> should be replaced by the hostname of the system running the gfapi application
• <PID> should be replaced by the PID of the gfapi application
• through calling glfs_sysrq(<FS>, GLFS_SYSRQ_STATEDUMP) within the application
• <FS> should be replaced by a pointer to a glfs_t structure

All statedumps (*.dump.* files) will be located at the usual location, on most distributions this would be /var/run/gluster/.

Limitations: It is not possible to trigger statedumps from the Gluster CLI when the gfapi application has lost its management connection to the GlusterD servers.

GlusterFS 3.10 is the first release that contains support for the new glfs_sysrq() function. Applications that include features for debugging will need to be adapted to call this function. At the time of the release of 3.10, no applications are known to call glfs_sysrq().

Known Issues:

Disabled creation of trash directory by default

Notes for users: From now onwards trash directory, namely .trashcan, will not be be created by default upon creation of new volumes unless and until the feature is turned ON and the restrictions on the same will be applicable as long as features.trash is set for a particular volume.

Limitations: After upgrade for pre-existing volumes, trash directory will be still present at root of the volume. Those who are not interested in this feature may have to manually delete the directory from the mount point.

Known Issues:

Implemented parallel readdirp with distribute xlator

Notes for users: Currently the directory listing gets slower as the number of bricks/nodes increases in a volume, though the file/directory numbers remain unchanged. With this feature, the performance of directory listing is made mostly independent of the number of nodes/bricks in the volume. Thus scale doesn’t exponentially reduce the directory listing performance. (On a 2, 5, 10, 25 brick setup we saw ~5, 100, 400, 450% improvement consecutively)

To enable this feature:

# gluster volume set <VOLNAME> performance.readdir-ahead on
# gluster volume set <VOLNAME> performance.parallel-readdir on

To disable this feature:

# gluster volume set <VOLNAME> performance.parallel-readdir off

If there are more than 50 bricks in the volume it is good to increase the cache size to be more than 10Mb (default value):

# gluster volume set <VOLNAME> performance.rda-cache-limit <CACHE SIZE>

Limitations:

Known Issues:

md-cache can optionally -ve cache security.ima xattr

Notes for users: From kernel version 3.X or greater, creating of a file results in removexattr call on security.ima xattr. This xattr is not set on the file unless IMA feature is active. With this patch, removxattr call returns ENODATA if it is not found in the cache.

The end benefit is faster create operations where IMA is not enabled.

To cache this xattr use,

# gluster volume set <VOLNAME> performance.cache-ima-xattrs on

The above option is on by default.

Limitations:

Known Issues:

Added support for CPU extensions in disperse computations

Notes for users: To improve disperse computations, a new way of generating dynamic code targeting specific CPU extensions like SSE and AVX on Intel processors is implemented. The available extensions are detected on run time. This can roughly double encoding and decoding speeds (or halve CPU usage).

This change is 100% compatible with the old method. No change is needed if an existing volume is upgraded.

You can control which extensions to use or disable them with the following command:

# gluster volume set <VOLNAME> disperse.cpu-extensions <type>

Valid values are:

• none: Completely disable dynamic code generation
• auto: Automatically detect available extensions and use the best one
• x64: Use dynamic code generation using standard 64 bits instructions
• sse: Use dynamic code generation using SSE extensions (128 bits)
• avx: Use dynamic code generation using AVX extensions (256 bits)

The default value is ‚auto‘. If a value is specified that is not detected on run-time, it will automatically fall back to the next available option.

Limitations:

Known Issues: To solve a conflict between the dynamic code generator and SELinux, it has been necessary to create a dynamic file on runtime in the directory /usr/libexec/glusterfs. This directory only exists if the server package is installed. On nodes with only the client package installed, this directory won’t exist and the dynamic code won’t be used.

It also needs root privileges to create the file there, so any gfapi application not running as root won’t be able to use dynamic code generation.

In these cases, disperse volumes will continue working normally but using the old implementation (equivalent to setting disperse.cpu-extensions to none).

More information and a discussion on how to solve this can be found here:

https://bugzilla.redhat.com/1421649

Bugs addressed

Bugs addressed since release-3.9 are listed below.

• #789278: Issues reported by Coverity static analysis tool
• #1198849: Minor improvements and cleanup for the build system
• #1211863: RFE: Support in md-cache to use upcall notifications to invalidate its cache
• #1231224: Misleading error messages on brick logs while creating directory (mkdir) on fuse mount
• #1234054: `gluster volume heal split-brain‘ does not heal if data/metadata/entry self-heal options are turned off
• #1289922: Implement SIMD support on EC
• #1290304: [RFE]Reducing number of network round trips
• #1297182: Mounting with „-o noatime“ or „-o noexec“ causes „nosuid,nodev“ to be set as well
• #1313838: Tiering as separate process and in v status moving tier task to tier process
• #1316873: EC: Set/unset dirty flag for all the update operations
• #1325531: Statedump: Add per xlator ref counting for inode
• #1325792: „gluster vol heal test statistics heal-count replica“ seems doesn’t work
• #1330604: out-of-tree builds generate XDR headers and source files in the original directory
• #1336371: Sequential volume start&stop is failing with SSL enabled setup.
• #1341948: DHT: Rebalance- Misleading log messages from __dht_check_free_space function
• #1344714: removal of file from nfs mount crashs ganesha server
• #1349385: [FEAT]jbr: Add rollbacking of failed fops
• #1355956: RFE : move ganesha related configuration into shared storage
• #1356076: DHT doesn’t evenly balance files on FreeBSD with ZFS
• #1356960: OOM Kill on client when heal is in progress on 1*(2+1) arbiter volume
• #1357753: JSON output for all Events CLI commands
• #1357754: Delayed Events if any one Webhook is slow
• #1358296: tier: breaking down the monolith processing function tier_migrate_using_query_file()
• #1359612: [RFE] Geo-replication Logging Improvements
• #1360670: Add output option --xml to man page of gluster
• #1363595: Node remains in stopped state in pcs status with „/usr/lib/ocf/resource.d/heartbeat/ganesha_mon: line 137: [: too many arguments ]“ messages in logs.
• #1363965: geo-replication *changes.log does not respect the log-level configured
• #1364420: [RFE] History Crawl performance improvement
• #1365395: Support for rc.d and init for Service management
• #1365740: dht: Update stbuf from servers having layout
• #1365791: Geo-rep worker Faulty with OSError: [Errno 21] Is a directory
• #1365822: [RFE] cli command to get max supported cluster.op-version
• #1366494: Rebalance is not considering the brick sizes while fixing the layout
• #1366495: 1 mkdir generates tons of log messages from dht xlator
• #1366648: [GSS] A hot tier brick becomes full, causing the entire volume to have issues and returns stale file handle and input/output error.
• #1366815: spurious heal info as pending heal entries never end on an EC volume while IOs are going on
• #1368012: gluster fails to propagate permissions on the root of a gluster export when adding bricks
• #1368138: Crash of glusterd when using long username with geo-replication
• #1368312: Value of `replica.split-brain-status‘ attribute of a directory in metadata split-brain in a dist-rep volume reads that it is not in split-brain
• #1368336: [RFE] Tier Events
• #1369077: The directories get renamed when data bricks are offline in 4*(2+1) volume
• #1369124: fix unused variable warnings from out-of-tree builds generate XDR headers and source files i…
• #1369397: segment fault in changelog_cleanup_dispatchers
• #1369403: [RFE]: events from protocol server
• #1369523: worm: variable reten_mode is invalid to be free by mem_put in fini()
• #1370410: [granular entry sh] – Provide a CLI to enable/disable the feature that checks that there are no heals pending before allowing the operation
• #1370567: [RFE] Provide snapshot events for the new eventing framework
• #1370931: glfs_realpath() should not return malloc()’d allocated memory
• #1371353: posix: Integrate important events with events framework
• #1371470: disperse: Integrate important events with events framework
• #1371485: [RFE]: AFR events
• #1371539: Quota version not changing in the quota.conf after upgrading to 3.7.1 from 3.6.1
• #1371540: Spurious regression in tests/basic/gfapi/bug1291259.t
• #1371874: [RFE] DHT Events
• #1372193: [geo-rep]: AttributeError: ‚Popen‘ object has no attribute ‚elines‘
• #1372211: write-behind: flush stuck by former failed write
• #1372356: glusterd experiencing repeated connect/disconnect messages when shd is down
• #1372553: „gluster vol status all clients –xml“ doesn’t generate xml if there is a failure in between
• #1372584: Fix the test case http://review.gluster.org/#/c/15385/
• #1373072: Event pushed even if Answer is No in the Volume Stop and Delete prompt
• #1373373: Worker crashes with EINVAL errors
• #1373520: [Bitrot]: Recovery fails of a corrupted hardlink (and the corresponding parent file) in a disperse volume
• #1373741: [geo-replication]: geo-rep Status is not showing bricks from one of the nodes
• #1374093: glusterfs: create a directory with 0464 mode return EIO error
• #1374286: [geo-rep]: defunct tar process while using tar+ssh sync
• #1374584: Detach tier commit is allowed when detach tier start goes into failed state
• #1374587: gf_event python fails with ImportError
• #1374993: bug-963541.t spurious failure
• #1375181: /var/tmp/rpm-tmp.KPCugR: line 2: /bin/systemctl: No such file or directory
• #1375431: [RFE] enable sharding and strict-o-direct with virt profile – /var/lib/glusterd/groups/virt
• #1375526: Kill rpc.statd on Linux machines
• #1375532: Rpm installation fails with conflicts error for eventsconfig.json file
• #1376671: Rebalance fails to start if a brick is down
• #1376693: RFE: Provide a prompt when enabling gluster-NFS
• #1377097: The GlusterFS Callback RPC-calls always use RPC/XID 42
• #1377341: out-of-tree builds generate XDR headers and source files in the original directory
• #1377427: incorrect fuse dumping for WRITE
• #1377556: Files not being opened with o_direct flag during random read operation (Glusterfs 3.8.2)
• #1377584: memory leak problems are found in daemon:glusterd, server:glusterfsd and client:glusterfs
• #1377607: Volume restart couldn’t re-export the volume exported via ganesha.
• #1377864: Creation of files on hot tier volume taking very long time
• #1378057: glusterd fails to start without installing glusterfs-events package
• #1378072: Modifications to AFR Events
• #1378305: DHT: remove unused structure members
• #1378436: build: python-ctypes no longer exists in Fedora Rawhide
• #1378492: warning messages seen in glusterd logs for each ‚gluster volume status‘ command
• #1378684: Poor smallfile read performance on Arbiter volume compared to Replica 3 volume
• #1378778: Add a test script for compound fops changes in AFR
• #1378842: [RFE] ‚gluster volume get‘ should implement the way to retrieve volume options using the volume name ‚all‘
• #1379223: „nfs.disable: on“ is not showing in Vol info by default for the 3.7.x volumes after updating to 3.9.0
• #1379285: gfapi: Fix fd ref leaks
• #1379328: Boolean attributes are published as string
• #1379330: eventsapi/georep: Events are not available for Checkpoint and Status Change
• #1379511: Fix spurious failures in open-behind.t
• #1379655: Recording (ffmpeg) processes on FUSE get hung
• #1379720: errors appear in brick and nfs logs and getting stale files on NFS clients
• #1379769: GlusterFS fails to build on old Linux distros with linux/oom.h missing
• #1380249: Huge memory usage of FUSE client
• #1380275: client ID should logged when SSL connection fails
• #1381115: Polling failure errors getting when volume is started&stopped with SSL enabled setup.
• #1381421: afr fix shd log message error
• #1381830: Regression caused by enabling client-io-threads by default
• #1382236: glusterfind pre session hangs indefinitely
• #1382258: RFE: Support to update NFS-Ganesha export options dynamically
• #1382266: md-cache: Invalidate cache entry in case of OPEN with O_TRUNC
• #1384142: crypt: changes needed for openssl-1.1 (coming in Fedora 26)
• #1384297: glusterfs can’t self heal character dev file for invalid dev_t parameters
• #1384906: arbiter volume write performance is bad with sharding
• #1385104: invalid argument warning messages seen in fuse client logs 2016-09-30 06:34:58.938667] W [dict.c:418ict_set] (–>/usr/lib64/glusterfs/3.8.4/xlator/cluster/replicate.so(+0x58722) 0-dict: !this || !value for key=link-count [Invalid argument]
• #1385575: pmap_signin event fails to update brickinfo->signed_in flag
• #1385593: Fix some spelling mistakes in comments and log messages
• #1385839: Incorrect volume type in the „glusterd_state“ file generated using CLI „gluster get-state“
• #1386088: Memory Leaks in snapshot code path
• #1386097: 4 of 8 bricks (2 dht subvols) crashed on systemic setup
• #1386123: geo-replica slave node goes faulty for non-root user session due to fail to locate gluster binary
• #1386141: Error and warning message getting while removing glusterfs-events package
• #1386188: Asynchronous Unsplit-brain still causes Input/Output Error on system calls
• #1386200: Log all published events
• #1386247: [Eventing]: ‚gluster volume tier start force‘ does not generate a TIER_START event
• #1386450: Continuous warning messages getting when one of the cluster node is down on SSL setup.
• #1386516: [Eventing]: UUID is showing zeros in the event message for the peer probe operation.
• #1386626: fuse mount point not accessible
• #1386766: trashcan max file limit cannot go beyond 1GB
• #1387160: clone creation with older names in a system fails
• #1387207: [Eventing]: Random VOLUME_SET events seen when no operation is done on the gluster cluster
• #1387241: Pass proper permission to acl_permit() in posix_acl_open()
• #1387652: [Eventing]: BRICK_DISCONNECTED events seen when a tier volume is stopped
• #1387864: [Eventing]: ‚gluster vol bitrot scrub ondemand‘ does not produce an event
• #1388010: [Eventing]: ‚VOLUME_REBALANCE‘ event messages have an incorrect volume name
• #1388062: throw warning to show that older tier commands are depricated and will be removed.
• #1388292: performance.read-ahead on results in processes on client stuck in IO wait
• #1388348: glusterd: Display proper error message and fail the command if S32gluster_enable_shared_storage.sh hook script is not present during gluster volume set all cluster.enable-shared-storage command
• #1388401: Labelled geo-rep checkpoints hide geo-replication status
• #1388861: build: python on Debian-based dists use …/lib/python2.7/dist-packages instead of …/site-packages
• #1388862: [Eventing]: Events not seen when command is triggered from one of the peer nodes
• #1388877: Continuous errors getting in the mount log when the volume mount server glusterd is down.
• #1389293: build: incorrect Requires: for portblock resource agent
• #1389481: glusterfind fails to list files from tiered volume
• #1389697: Remove-brick status output is showing status of fix-layout instead of original remove-brick status output
• #1389746: Refresh config fails while exporting subdirectories within a volume
• #1390050: Elasticsearch get CorruptIndexException errors when running with GlusterFS persistent storage
• #1391086: gfapi clients crash while using async calls due to double fd_unref
• #1391387: The FUSE client log is filling up with posix_acl_default and posix_acl_access messages
• #1392167: SMB[md-cache Private Build]:Error messages in brick logs related to upcall_cache_invalidate gf_uuid_is_null
• #1392445: Hosted Engine VM paused post replace-brick operation
• #1392713: inconsistent file permissions b/w write permission and sticky bits(———T ) displayed when IOs are going on with md-cache enabled (and within the invalidation cycle)
• #1392772: [setxattr_cbk] „Permission denied“ warning messages are seen in logs while running pjd-fstest suite
• #1392865: Better logging when reporting failures of the kind “ Failing MKNOD as quorum is not met“
• #1393259: stat of file is hung with possible deadlock
• #1393678: Worker restarts on log-rsync-performance config update
• #1394131: [md-cache]: All bricks crashed while performing symlink and rename from client at the same time
• #1394224: „nfs-grace-monitor“ timed out messages observed
• #1394548: Make debugging EACCES errors easier to debug
• #1394719: libgfapi core dumps
• #1394881: Failed to enable nfs-ganesha after disabling nfs-ganesha cluster
• #1395261: Seeing error messages [snapview-client.c:283:gf_svc_lookup_cbk] and [dht-helper.c:1666ht_inode_ctx_time_update] (–>/usr/lib64/glusterfs/3.8.4/xlator/cluster/replicate.so(+0x5d75c)
• #1395648: ganesha-ha.conf –status should validate if the VIPs are assigned to right nodes
• #1395660: Checkpoint completed event missing master node detail
• #1395687: Client side IObuff leaks at a high pace consumes complete client memory and hence making gluster volume inaccessible
• #1395993: heal info –xml when bricks are down in a systemic environment is not displaying anything even after more than 30minutes
• #1396038: refresh-config fails and crashes ganesha when mdcache is enabled on the volume.
• #1396048: A hard link is lost during rebalance+lookup
• #1396062: [geo-rep]: Worker crashes seen while renaming directories in loop
• #1396081: Wrong value in Last Synced column during Hybrid Crawl
• #1396364: Scheduler : Scheduler should not depend on glusterfs-events package
• #1396793: [Ganesha] : Ganesha crashes intermittently during nfs-ganesha restarts.
• #1396807: capture volume tunables in get-state dump
• #1396952: I/O errors on FUSE mount point when reading and writing from 2 clients
• #1397052: OOM kill of nfs-ganesha on one node while fs-sanity test suite is executed.
• #1397177: memory leak when using libgfapi
• #1397419: glusterfs_ctx_defaults_init is re-initializing ctx->locks
• #1397424: PEER_REJECT, EVENT_BRICKPATH_RESOLVE_FAILED, EVENT_COMPARE_FRIEND_VOLUME_FAILED are not seen
• #1397754: [SAMBA-CIFS] : IO hungs in cifs mount while graph switch on & off
• #1397795: NFS-Ganesha:Volume reset for any option causes reset of ganesha enable option and bring down the ganesha services
• #1398076: SEEK_HOLE/ SEEK_DATA doesn’t return the correct offset
• #1398226: With compound fops on, client process crashes when a replica is brought down while IO is in progress
• #1398566: self-heal info command hangs after triggering self-heal
• #1399031: build: add systemd dependency to glusterfs sub-package
• #1399072: [Disperse] healing should not start if only data bricks are UP
• #1399134: GlusterFS client crashes during remove-brick operation
• #1399154: After ganesha node reboot/shutdown, portblock process goes to FAILED state
• #1399186: [GANESHA] Export ID changed during volume start and stop with message „lookup_export failed with Export id not found“ in ganesha.log
• #1399578: [compound FOPs]: Memory leak while doing FOPs with brick down
• #1399592: Memory leak when self healing daemon queue is full
• #1399780: Use standard refcounting for structures where possible
• #1399995: Dump volume specific options in get-state output in a more parseable manner
• #1400013: [USS,SSL] .snaps directory is not reachable when I/O encryption (SSL) is enabled
• #1400026: Duplicate value assigned to GD_MSG_DAEMON_STATE_REQ_RCVD and GD_MSG_BRICK_CLEANUP_SUCCESS messages
• #1400237: Ganesha services are not stopped when pacemaker quorum is lost
• #1400613: [GANESHA] failed to create directory of hostname of new node in var/lib/nfs/ganesha/ in already existing cluster nodes
• #1400818: possible memory leak on client when writing to a file while another client issues a truncate
• #1401095: log the error when locking the brick directory fails
• #1401218: Fix compound fops memory leaks
• #1401404: [Arbiter] IO’s Halted and heal info command hung
• #1401777: atime becomes zero when truncating file via ganesha (or gluster-NFS)
• #1401801: [RFE] Use Host UUID to find local nodes to spawn workers
• #1401812: RFE: Make readdirp parallel in dht
• #1401822: [GANESHA]Unable to export the ganesha volume after doing volume start and stop
• #1401836: update documentation to readthedocs.io
• #1401921: glusterfsd crashed while taking snapshot using scheduler
• #1402237: Bad spacing in error message in cli
• #1402261: cli: compile warnings (unused var) if building without bd xlator
• #1402369: Getting the warning message while erasing the gluster „glusterfs-server“ package.
• #1402710: ls and move hung on disperse volume
• #1402730: self-heal not happening, as self-heal info lists the same pending shards to be healed
• #1402828: Snapshot: Snapshot create command fails when gluster-shared-storage volume is stopped
• #1402841: Files remain unhealed forever if shd is disabled and re-enabled while healing is in progress.
• #1403130: [GANESHA] Adding a node to cluster failed to allocate resource-agents to new node.
• #1403780: Incorrect incrementation of volinfo refcnt during volume start
• #1404118: Snapshot: After snapshot restore failure , snapshot goes into inconsistent state
• #1404168: Upcall: Possible use after free when log level set to TRACE
• #1404181: [Ganesha+SSL] : Ganesha crashes on all nodes on volume restarts
• #1404410: [Perf] : pcs cluster resources went into stopped state during Multithreaded perf tests on RHGS layered over RHEL 6
• #1404573: tests/bugs/snapshot/bug-1316437.t test is causing spurious failure
• #1404678: [geo-rep]: Config commands fail when the status is ‚Created‘
• #1404905: DHT : file rename operation is successful but log has error ‚key:trusted.glusterfs.dht.linkto error:File exists‘ , ’setting xattrs on failed (File exists)‘
• #1405165: Allow user to disable mem-pool
• #1405301: Fix the failure in tests/basic/gfapi/bug1291259.t
• #1405478: Keepalive should be set for IPv6 & IPv4
• #1405554: Fix spurious failure in bug-1402841.t-mt-dir-scan-race.t
• #1405775: GlusterFS process crashed after add-brick
• #1405902: Fix spurious failure in tests/bugs/replicate/bug-1402730.t
• #1406224: VM pauses due to storage I/O error, when one of the data brick is down with arbiter/replica volume
• #1406249: [GANESHA] Deleting a node from ganesha cluster deletes the volume entry from /etc/ganesha/ganesha.conf file
• #1406252: Free xdr-allocated compound request and response arrays
• #1406348: [Eventing]: POSIX_SAME_GFID event seen for .trashcan folder and .trashcan/internal_op
• #1406410: [GANESHA] Adding node to ganesha cluster is not assigning the correct VIP to the new node
• #1406411: Fail add-brick command if replica count changes
• #1406878: ec prove tests fail in FB build environment.
• #1408115: Remove-brick rebalance failed while rm -rf is in progress
• #1408131: Remove tests/distaf
• #1408395: [Arbiter] After Killing a brick writes drastically slow down
• #1408712: with granular-entry-self-heal enabled i see that there is a gfid mismatch and vm goes to paused state after migrating to another host
• #1408755: Remove tests/basic/rpm.t
• #1408757: Fix failure of split-brain-favorite-child-policy.t in CentOS7
• #1408758: tests/bugs/glusterd/bug-913555.t fails spuriously
• #1409078: RFE: Need a command to check op-version compatibility of clients
• #1409186: Dict_t leak in dht_migration_complete_check_task and dht_rebalance_inprogress_task
• #1409202: Warning messages throwing when EC volume offline brick comes up are difficult to understand for end user.
• #1409206: Extra lookup/fstats are sent over the network when a brick is down.
• #1409727: [ganesha + EC]posix compliance rename tests failed on EC volume with nfs-ganesha mount.
• #1409730: [ganesha+ec]: Contents of original file are not seen when hardlink is created
• #1410071: [Geo-rep] Geo replication status detail without master and slave volume args
• #1410313: brick crashed on systemic setup
• #1410355: Remove-brick rebalance failed while rm -rf is in progress
• #1410375: [Mdcache] clients being served wrong information about a file, can lead to file inconsistency
• #1410777: ganesha service crashed on all nodes of ganesha cluster on disperse volume when doing lookup while copying files remotely using scp
• #1410853: glusterfs-server should depend on firewalld-filesystem
• #1411607: [Geo-rep] If for some reason MKDIR failed to sync, it should not proceed further.
• #1411625: Spurious split-brain error messages are seen in rebalance logs
• #1411999: URL to Fedora distgit no longer uptodate
• #1412002: Examples/getvolfile.py is not pep8 compliant
• #1412069: No rollback of renames on succeeded subvols during failure
• #1412174: Memory leak on mount/fuse when setxattr fails
• #1412467: Remove tests/bugs/distribute/bug-1063230.t
• #1412489: Upcall: Possible memleak if inode_ctx_set fails
• #1412689: [Geo-rep] Slave mount log file is cluttered by logs of multiple active mounts
• #1412917: OOM kill of glusterfsd during continuous add-bricks
• #1412918: fuse: Resource leak in fuse-helper under GF_SOLARIS_HOST_OS
• #1413967: geo-rep session faulty with ChangelogException „No such file or directory“
• #1415226: packaging: python/python2(/python3) cleanup
• #1415245: core: max op version
• #1415279: libgfapi: remove/revert glfs_ipc() changes targeted for 4.0
• #1415581: RFE : Create trash directory only when its is enabled
• #1415915: RFE: An administrator friendly way to determine rebalance completion time
• #1415918: Cache security.ima xattrs as well
• #1416285: EXPECT_WITHIN is taking too much time even if the result matches with expected value
• #1416416: Improve output of „gluster volume status detail“
• #1417027: option performance.parallel-readdir should honor cluster.readdir-optimize
• #1417028: option performance.parallel-readdir can cause OOM in large volumes
• #1417042: glusterd restart is starting the offline shd daemon on other node in the cluster
• #1417135: [Stress] : SHD Logs flooded with „Heal Failed“ messages,filling up „/“ quickly
• #1417521: [SNAPSHOT] With all USS plugin enable .snaps directory is not visible in cifs mount as well as windows mount
• #1417527: glusterfind: After glusterfind pre command execution all temporary files and directories /usr/var/lib/misc/glusterfsd/glusterfind/// should be removed
• #1417804: debug/trace: Print iatts of individual entries in readdirp callback for better debugging experience
• #1418091: [RFE] Support multiple bricks in one process (multiplexing)
• #1418536: Portmap allocates way too much memory (256KB) on stack
• #1418541: [Ganesha+SSL] : Bonnie++ hangs during rewrites.
• #1418623: client process crashed due to write behind translator
• #1418650: Samba crash when mounting a distributed dispersed volume over CIFS
• #1418981: Unable to take Statedump for gfapi applications
• #1419305: disable client.io-threads on replica volume creation
• #1419306: [RFE] Need to have group cli option to set all md-cache options using a single command
• #1419503: [SAMBA-SSL] Volume Share hungs when multiple mount & unmount is performed over a windows client on a SSL enabled cluster
• #1419696: Fix spurious failure of ec-background-heal.t and tests/bitrot/bug-1373520.t
• #1419824: repeated operation failed warnings in gluster mount logs with disperse volume
• #1419825: Sequential and Random Writes are off target by 12% and 22% respectively on EC backed volumes over FUSE
• #1419846: removing warning related to enum, to let the build take place without errors for 3.10
• #1419855: [Remove-brick] Hardlink migration fails with „lookup failed (No such file or directory)“ error messages in rebalance logs
• #1419868: removing old tier commands under the rebalance commands
• #1420606: glusterd is crashed at the time of stop volume
• #1420808: Trash feature improperly disabled
• #1420810: Massive xlator_t leak in graph-switch code
• #1420982: Automatic split brain resolution must check for all the bricks to be up to avoiding serving of inconsistent data(visible on x3 or more)
• #1420987: warning messages seen in glusterd logs while setting the volume option
• #1420989: when server-quorum is enabled, volume get returns 0 value for server-quorum-ratio
• #1420991: Modified volume options not synced once offline nodes comes up.
• #1421017: CLI option „–timeout“ is accepting non numeric and negative values.
• #1421956: Disperse: Fallback to pre-compiled code execution when dynamic code generation fails
• #1422350: glustershd process crashed on systemic setup
• #1422363: [Replicate] „RPC call decoding failed“ leading to IO hang & mount inaccessible
• #1422391: Gluster NFS server crashing in __mnt3svc_umountall
• #1422766: Entry heal messages in glustershd.log while no entries shown in heal info
• #1422777: DHT doesn’t evenly balance files on FreeBSD with ZFS
• #1422819: [Geo-rep] Recreating geo-rep session with same slave after deleting with reset-sync-time fails to sync
• #1422942: Prevent reverse heal from happening
• #1423063: glusterfs-fuse RPM now depends on gfapi
• #1423070: Bricks not coming up when ran with address sanitizer
• #1423385: Crash in index xlator because of race in inode_ctx_set and inode_ref
• #1423406: Need to improve remove-brick failure message when the brick process is down.
• #1423412: Mount of older client fails
• #1423429: unnecessary logging in rda_opendir
• #1424921: dht_setxattr returns EINVAL when a file is deleted during the FOP
• #1424931: [RFE] Include few more options in virt file
• #1424937: multiple glusterfsd process crashed making the complete subvolume unavailable
• #1424973: remove-brick status shows 0 rebalanced files
• #1425556: glusterd log is flooded with stale disconnect rpc messages

Quelle: https://gluster.readthedocs.io/en/latest/release-notes/3.10.0/