-
Fri Nov 03 2017 Tiger Yang <tiger.yang@oracle.com> - 3.10.7-1.el6
- BUILDINFO: commit=6a3b4a6a712c2bae2375d760346b208101abc3be
- Upgrade to 3.10.7-1
-
Fri Nov 03 2017 Tiger Yang <tiger.yang@oracle.com> - 3.10.7-1
- glusterd: use sys_lstat instead of lstat
- glusterd: delete source brick only once in reset-brick commit force
- glusterd: clean up portmap on brick disconnect
- glusterd: persist brickinfo's port change into glusterd's store
- doc: Added release notes for 3.10.7 release
- md-cache: Use correct xattr keynames for virtual glusterfs ACLs.
- libgfchangelog: Fix possible null pointer dereference
- glusterd: Gluster should keep PID file in correct location
- features/worm: Adding check to newloc when doing rename
- packaging: manual systemctl daemon reload required after install
- glusterfsd: Dereferencing the null pointer
- md-cache: avoid checking the xattr value buffer with string functions.
- gfapi: set lkowner in glfd
-
Mon Oct 09 2017 Joe Jin <joe.jin@oracle.com> - 3.10.6-1
- afr: don't check for file size in afr_mark_source_sinks_if_file_empty
- doc: Added release notes for 3.10.6 release
- cluster/dht: EBADF handling for fremovexattr and fsetxattr
- mount/fuse: Make event-history feature configurable
- afr: auto-resolve split-brains for zero-byte files
- dht: add FOP check to dht_file_setattr_cbk
- ganesha: fix create-export-ganesha.sh script error
- Fix incorrect parameters on gluterd.sysconfig
- ganesha-ha.sh: fix realpath does not existed on OL6
- posix: add sanity checks for removing the gfid symlink for directories
- afr: Prevent null gfids in self-heal entry re-creation
- afr: heal metadata in discover code path
- glusterd: fix invalid memory reference returned
- rpc: TLSv1_2_method() is deprecated in OpenSSL-1.1
- posix: add null gfid checks
- afr: check validity of afr_reply
- features/shard: Return aggregated size in stbuf of LINK fop
- features/shard: Increment counts in locks
- glusterd : glusterd fails to start when peer's network interface is down
- cluster/dht: Check for open fd only on EBADF
- ganesha-ha: don't set SELinux booleans if SELinux is disabled
- packaging: /run and /var/run; prefer /run
- packaging: use rdma-core(-devel) instead of ibverbs, rdmacm
- cluster/ec: Node uuid xattr support update for EC
- xlators/ganesha : remove ganesha xlator code from 3.10
- cluster/ec: return all node uuids from all subvolumes
- spec: enable glusterfs-regression-tests build
-
Wed Aug 30 2017 Joe Jin <joe.jin@oracle.com> - 3.10.5-1
- api: memory leak in glfs_h_acl_get(), missing dict unref
- doc: Added release notes for 3.10.5 release
- nfs: add NULL check for call state in nfs3_call_state_wipe
- cluster/rebalance: Fix hardlink migration failures
- cluster/dht: fix on demand migration files from client
- cluster/dht: initialize throttle option "normal" to same in init and reconfigure
- cluster/dht: Make rebalance throttle option tuned by number
- cluster/dht: rebalance perf enhancement
- cluster/dht: correct space check for rebalance
- cluster/dht: Skip file migration if the subvol that meets min-free-disk
- cluster/dht: Make rebalance honor min-free-disk
- nfs/nlm: keep track of the call-state and frame for notifications
- dht/rebalance: Crawler performance improvement
- nfs/nlm: use refcounting for nfs3_call_state_t
- refcount: typecast function for calling on free
- performance/io-cache: update inode contexts of each entry in readdirplus
- scripts: invalid test(1) in extras/S32gluster_enable_shared_storage.sh
- packaging: /var/lib/glusterd/options should be %config(noreplace)
- nfs/nlm: handle reconnect for non-NLM4_LOCK requests
- nfs/nlm: unref fds in nlm_client_free()
- nfs: make nfs3_call_state_t refcounted
- geo-rep: Fix syncing of self healed hardlinks
- packaging: own files in (/var)/run/gluster/shared_storage/nfs-ganesha
- common-ha: enable and disable selinux gluster_use_execmem
- packaging: glusterfs-ganesha update sometimes fails semanage
- cluster/dht: Fix fd check race
- cluster/dht: Check if fd is opened on dst subvol
- fuse: memory leak fixes
- afr: mark non sources as sinks in metadata heal
- cli/xml: fix return handling
- geo-rep: Fix entry failure because parent dir doesn't exist
- cluster/ec: correctly handle end of file for seek
- doc: release notes for 3.10.4
- glfsheal: prevent background self-heals
- Revert "Revert "glusterd: disallow rebalance & remove-brick on a sharded volume""
- cluster/dht: Additional checks for rebalance estimates
- core: assorted typos and spelling mistakes from Debian lintian
- cluster/afr: Returning single and list of node uuids from AFR
- cluster:dht Fix crash in dht_rename_lock_cbk
- feature/changelog: Fix buffer overflow crash
- common-ha: enable and disable selinux ganesha_use_fusefs
- rpc: add options to manage socket keepalive lifespan
- cluster/dht: Include dirs in rebalance estimates
- posix-acl: Whitelist virtual ACL xattrs
- libglusterfs : Fix crash in glusterd while peer probing
- Revert "glusterd: disallow rebalance & remove-brick on a sharded volume"
- feature/bitrot: Fix ondemand scrub
- posix: Revert modifying op_errno in __posix_fd_ctx_get
- afr: add errno to afr_inode_refresh_done()
- posix: use the correct op_errno
- common-ha: surviving ganesha.nfsd not put in grace on fail-over
- doc: release notes for 3.10.3
- tier: port value missing on cli parsing
- glusterd: Eliminate race in brick compatibility checking stage
- perf/ioc: Fix race causing crash when accessing freed page
- Tier: removing port allocated for tier
- glusterfsd: send PARENT_UP on brick attach
- Fixes upgrade issue in quota.conf
- rpc: log more about socket disconnects
- rpc: avoid logging success on failure
- Tier/cli: detach status xml output
- glusterd: Don't spawn new glusterfsds on node reboot with brick-mux
- glusterd: ignore incorrect uuid validation if uuid_str is empty
- cluster/afr: Return the list of node_uuids for the subvolume
- cluster/dht: Fix crash in dht_selfheal_dir_setattr
- build: ensure autoconf template expansions resolve variables
- afr: propagate correct errno for fop failures in arbiter
- cluster/dht: Rebalance on all nodes should migrate files
- cluster/dht: Fix crash in dht rmdir
- common-ha: adding a node to existing cluster failed to start pacemaker
-
Tue May 16 2017 Joe Jin <joe.jin@oracle.com> - 3.10.2-1
- release notes for 3.10.2
- nfs/nlm: remove lock request from the list after cancel
- nfs/nlm: free the nlm_client upon RPC_DISCONNECT
- nfs/nlm: log the caller_name if nlm_client_t can be found
- nfs/nlm: ignore notify when there is no matching rpc request
- nfs/nlm: unref rpc-client after nlm4svc_send_granted()
- dht: Add missing braces in dht_opendir
- afr: send the correct iatt values in fsync cbk
- Fixes quota aux mount failure
- rpc: fix a routine to destory RDMA qp(queue-pair)
- glusterd: Make reset-brick work correctly if brick-mux is on
- Tier: Watermark check for hi and low value being equal
- performance/read-ahead: prevent stale data being returned to application.
- posix: Send SIGKILL in 2nd attempt
- rpc: fix transport add/remove race on port probing
- cluster/ec: fix incorrect answer check in seek fop
- glusterd: cleanup pidfile on pmap signout
- glusterd: socketfile & pidfile related fixes for brick multiplexing feature
- cluster/dht: Fix ret check
- TIER/TESTS: improving regression test for tier
- glusterd: disallow rebalance & remove-brick on a sharded volume
- ganesha : minor improvments for commit e91cdf4 (17081)
- geo-rep: filter out xtime attribute during getxattr
- glusterd/ganesha : add proper NULL check in manage_export_config
- eventsapi: Fix webhook-test when no schema specified in URL
- glusterd/geo-rep: Fix snapshot create in geo-rep setup
- ganesha/scripts : remove dependency over export configuration file for unexport
- glusterd/ganesha : return proper value in pre_setup()
- glusterd/ganesha : update cache invalidation properly during volume stop
- cluster/dht: Pass the correct xdata in fremovexattr fop
- glusterd/ganesha : perform removal of ganesha.conf on nodes only in ganesha cluster
- cluster/dht Remove redundant logs in dht rmdir
- tests: remove tests/bugs/core/bug-1421590-brick-mux-reuse-ports.t
- cluster/dht: Pass the req dict instead of NULL in dht_attr2()
- cluster/dht: rm -rf fails if dir has stale linkto files
- dht: The xattrs sent in readdirp should be sent in opendir aswell
- afr: don't do a post-op on a brick if op failed
- gfapi: Fix inode ref leak in anonymous fd I/O APIs
- geo-rep: Retry on EBUSY
- geo-rep: Fix EBUSY traceback
- glusterd: Fix snapshot failure in non-root geo-rep setup
- nfs: make subdir mounting work for Solaris 10 clients
- glusterd: fix glusterd_wait_for_blockers to go in infinite loop
- glusterd: set conn->reconnect to null on timer cancellation
- glusterd: hold off volume deletes while still restarting bricks
- explicitly specify executor to be bash for tests
- ganesha : allow refresh-config and volume export/unexport in failover state
- common-ha: Use UpdateExports dbus msg for refresh-config
- ganesha/scripts : Stop ganesha process on all nodes if cluster setup fails
- glusterd/ganesha : throw proper error for "gluster nfs-ganesha disable"
- glusterd/ganesha : During volume delete remove the ganesha export configuration file
- ganesha/scripts : Remove export entries from ganesha.conf during cleanup
-
Thu Apr 20 2017 Joe Jin <joe.jin@oracle.com> - 3.10.1-1
- Oracle-GlusterFS 3.10