-
Thu Sep 03 2020 Tiger Yang <tiger.yang@oracle.com> - 6.10-1.el7
- upgrade to 6.10-1
-
Tue Aug 25 2020 Tiger Yang <tiger.yang@oracle.com> - 6.10-1
- Adding release notes for release-6.10
- rpc: Cleanup SSL specific data at the time of freeing rpc object
- socket/ssl: fix crl handling
- locks/fencing: Address hang while lock preemption
- features/shard: Fix crash during shards cleanup in error cases
- cli: duplicate defns of cli_default_conn_timeout and cli_ten_minutes_timeout
- afr: event gen changes
- cluster/afr: Prioritize ENOSPC over other errors
- afr: more quorum checks in lookup and new entry marking
- fuse: degrade logging of write failure to fuse device
- cluster/ec: Return correct error code and log message
- md-cache: fix several NULL dereferences
- tests/bug-844688.t: test bug-844688.t is failing on master
- tests: skip tests on absence of reflink in xfs
- Adding release notes for release-6.9
- features/utime: Don't access frame after stack-wind
- utime: resolve an issue of permission denied logs
- mount/fuse: Wait for 'mount' child to exit before dying
- md-cache: avoid clearing cache when not necessary
- write-behind: fix data corruption
- snap_scheduler: python3 compatibility and new test case
- afr: mark pending xattrs as a part of metadata heal
- cluster/afr: fix race when bricks come up
- open-behind: fix missing fd reference
- gfapi: Suspend synctasks instead of blocking them
-
Mon Aug 10 2020 Tiger Yang <tiger.yang@oracle.com> - 6.8-2
- features/shard: Aggregate file size, block-count before unwinding removexattr
- features/shard: Aggregate size, block-count in iatt before unwinding setxattr
-
Sat Mar 21 2020 Tiger Yang <tiger.yang@oracle.com> - 6.8-1
- glusterd: Brick process fails to come up with brickmux on
- Adding release notes for release-6.8
- events: fix IPv6 memory corruption
- eventsapi: Set IPv4/IPv6 family based on input IP
- core: replace inet_addr with inet_pton
- core: fix memory pool management races
- cluster/ec: Change handling of heal failure to avoid crash
- afr: prevent spurious entry heals leading to gfid split-brain
- lock: check null value of dict to avoid log flooding
- cluster/thin-arbiter: Wait for TA connection before ta-file lookup
- tools/glusterfind: Remove an extra argument
- gf-event: Handle unix volfile-servers
- cluster/ec: skip updating ctx->loc again when ec_fix_open/opendir
- Cluster/afr: Don't treat all bricks having metadata pending as split-brain
- server: Mount fails after reboot 1/3 gluster nodes
- To fix readdir-ahead memory leak
- glusterfind: python3 compatibility
- tools/glusterfind: handle offline bricks
- extras: enable log rotation for USS logs
- cluster/dht: Correct fd processing loop
- doc: Added release 6.7 notes
- rpc: Synchronize slot allocation code
- geo-rep: Fix py2/py3 compatibility in repce
- test: fix non-root test case for geo-rep
- socket: fix error handling
- extras: Cgroup(CPU/Mem) restriction are not working on gluster process
- libgfchangelog : use find_library to locate shared library
- Detach iot_worker to release its resources
- cluster/ec: Update lock->good_mask on parent fop failure
- cluster/ec: Fix reopen flags to avoid misbehavior
- cluster/ec: Always read from good-mask
- cluster/ec: inherit healing from lock when it has info
- cluster/ec: Prevent double pre-op xattrops
-
Wed Nov 20 2019 Tiger Yang <tiger.yang@oracle.com> - 6.6-1
- doc: Added release 6.6 notes
- test: fix suspicous non-root geo-rep test failures
- spec: move libgfchangelog.so from glusterfs-devel to glusterfs-libs package
- geo-rep: Fix Permission denied traceback on non root setup
- cluster/ec: fix EIO error for concurrent writes on sparse files
- features/shard: Send correct size when reads are sent beyond file size
- geo-rep: Fix config upgrade on non-participating node
- tests : test case for non-root geo-rep setup
- geo-rep: Fix the name of changelog archive file
- geo-rep: Fix Config Get Race
- geo-rep: Test case for upgrading config file
- geo-rep : fix gluster command path for non-root session
- geo-rep: Upgrading config file to new version
- dht: Rebalance causing IO Error - File descriptor in bad state
- cluster/afr: Heal entries when there is a source & no healed_sinks
- afr: support split-brain CLI for replica 3
- system/posix-acl: update ctx only if iatt is non-NULL
- packaging: remove leftover bd cruft in rpm .spec
- perf/write-behind: Clear frame->local on conflict error
- fuse: add missing GF_FREE to fuse_interrupt
- ctime/rebalance: Heal ctime xattr on directory during rebalance
- geo-rep : fix mountbroker setup
- gfapi: 'glfs_h_creat_open' - new API to create handle and open fd
- dht: Custom xattrs are not healed in case of add-brick
- storage/posix: set the op_errno to proper errno during gfid set
- core/syncop: Bail out if frame creation fails
- afr/lookup: Pass xattr_req in while doing a selfheal in lookup
- rpc: Update address family if it is not provide in cmd-line arguments
- glusterd: IPV6 hostname address is not parsed correctly
- afr: wake up index healer threads
- afr: support split-brain CLI for replica 3
- system/posix-acl: update ctx only if iatt is non-NULL
- packaging: remove leftover bd cruft in rpm .spec
- perf/write-behind: Clear frame->local on conflict error
- fuse: add missing GF_FREE to fuse_interrupt
- ctime/rebalance: Heal ctime xattr on directory during rebalance
- geo-rep : fix mountbroker setup
- gfapi: 'glfs_h_creat_open' - new API to create handle and open fd
- dht: Custom xattrs are not healed in case of add-brick
- storage/posix: set the op_errno to proper errno during gfid set
- core/syncop: Bail out if frame creation fails
- afr/lookup: Pass xattr_req in while doing a selfheal in lookup
- rpc: Update address family if it is not provide in cmd-line arguments
- glusterd: IPV6 hostname address is not parsed correctly
- afr: wake up index healer threads
- tests: fix spurious failure of bug-1402841.t-mt-dir-scan-race.t
- ctime: Fix ctime issue with utime family of syscalls
- event: rename event_XXX with gf_ prefixed
- ctime: Fix incorrect realtime passed to frame->root->ctime
- performance/md-cache: Do not skip caching of null character xattr values
- afr: restore timestamp of parent dir during entry-heal
- glusterd: ./tests/bugs/glusterd/bug-1595320.t is failing
- rpc: glusterd start is failed and throwing an error Address already in use
-
Thu Oct 24 2019 Tiger Yang <tiger.yang@oracle.com> - 6.5-2
- geo-rep : fix mountbroker setup
- spec: move libgfchangelog.so from glusterfs-devel to glusterfs-libs package
-
Fri Aug 09 2019 Tiger Yang <tiger.yang@oracle.com> - 6.5-1
- doc: Added release 6.5 notes
- features/utime: always update ctime at setattr
- posix/ctime: Fix race during lookup ctime xattr heal
- geo-rep: Fix mount broker setup issue
- features/utime: Fix mem_put crash
- ctime: Set mdata xattr on legacy files
- geo-rep: Fix mount broker setup issue
- gfapi: Fix deadlock while processing upcall
- features/snapview-server: obtain the list of snapshots inside the lock
- cluster/dht: Fix directory perms during selfheal
- glusterd: do not mark skip_locking as true for geo-rep operations
-
Thu Jul 18 2019 Tiger Yang <tiger.yang@oracle.com> - 6.4-1
- samba: Gluster hooks for Samba config without kernel share mode
- doc: Added release notes for 6.4
- core: fix deadlock between statedump and fd_anonymous()
- extras/hooks: Install and package newly added post add-brick hook script
- features/snapview-server: use the same volfile server for gfapi options
- glusterd: add GF_TRANSPORT_BOTH_TCP_RDMA in glusterd_get_gfproxy_client_volfile
- mgmt/glusterd: Fix a memory leak when peer detach fails
- glusterd: conditionally clear txn_opinfo in stage op
- gfapi: fix incorrect initialization of upcall syncop arguments
- upcall: Avoid sending notifications for invalid inodes
- [RFC] change get_real_filename implementation to use ENOATTR instead of ENOENT
- extras/hooks: Add SELinux label on new bricks during add-brick
- features/shard: Fix block-count accounting upon truncate to lower size
- features/shard: Fix integer overflow in block count accounting
- tests: subdir-mount.t is failing for brick_mux regrssion
- posix : add posix_set_ctime() in posix_ftruncate()
- cluster/dht: Fixed a memleak in dht_rename_cbk
- core: fix memory allocation issues
- libglusterfs: Fix compilation when --disable-mempool is used
- mem-pool.{c|h}: minor changes
- core: avoid dynamic TLS allocation when possible
- posix/ctime: Fix ctime upgrade issue
- uss: Fix tar issue with ctime and uss enabled
- Merge tag 'v6.3' into oracle-gluster-6
- core: fix hang issue in __gf_free
- mem-pool: remove dead code.
- tests/utils: Fix py2/py3 util python scripts
-
Mon Jul 01 2019 Tiger Yang <tiger.yang@oracle.com> - 6.3-1
- doc: Added release notes for 6.3
- cluster/ec: honor contention notifications for partially acquired locks
- If bind-address is IPv6 return it successfully
-
Wed May 29 2019 Tiger Yang <tiger.yang@oracle.com> - 6.2-1
- doc: Added release notes for 6.2
- geo-rep: Convert gfid conflict resolutiong logs into debug
- geo-rep: Fix sync hang with tarssh
- tests/geo-rep: Fix arequal checksum comparison
- geo-rep: Fix sync-method config
- geo-rep: Fix rename with existing destination with same gfid
- geo-rep: Fix entries and metadata counters in geo-rep status
- cluster/ec: Reopen shouldn't happen with O_TRUNC
- afr: thin-arbiter lock release fixes
- cluster/afr : TA: Return actual error code in case of failure
- cluster/dht: Refactor dht lookup functions
- cluster/dht: refactor dht lookup functions
- glusterd: define dumpops in the xlator_api of glusterd
- ctime: Fix log repeated logging during open
- extras/hooks: syntactical errors in SELinux hooks, scipt logic improved
- cluster/ec: fix fd reopen
- core: handle memory accounting correctly