summaryrefslogtreecommitdiffstats
path: root/xlators
Commit message (Collapse)AuthorAgeFilesLines
* md-cache: fix time comparison to support 0 timeoutBrian Foster2012-04-021-2/+2
| | | | | | | | | | | | | | | | The time comparison currently causes caching when the current time matches the timeout. This causes caching to occur even when the timeout parameter is set to 0. Fix the condition to support the no caching scenario. A longer term approach is to use gettimeofday() for more accurate granularity, but this apparently fixes a known bug. Change-Id: I6552c5961743edb22a1aec2e6a3a6e041d0647b2 BUG: 809193 Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-on: http://review.gluster.com/3066 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
* nlm: print the reason of failure when clnt_create fails to create the client ↵Raghavendra Bhat2012-04-021-1/+2
| | | | | | | | | | | object Change-Id: Icc271e2166dbccc2cc12029e26c674fdb79fe975 BUG: 808390 Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com> Reviewed-on: http://review.gluster.com/3061 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* glusterd: Include port in nfs statusKaushal M2012-04-024-8/+25
| | | | | | | | | Change-Id: I679e491e801b694e8a0f93dd87cf540441dae927 BUG: 806877 Signed-off-by: Kaushal M <kaushal@redhat.com> Reviewed-on: http://review.gluster.com/3060 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* performance/quick-read: open shouldn't be worried about checkingRaghavendra G2012-04-022-73/+0
| | | | | | | | | | | | | | whether it is being done on a deleted dentry. lookup would've taken care of sending ENOENT, in case the entry was already deleted. Change-Id: I79c51e34ae3a202eb04123a2a58d0e819b6e94fa BUG: 808017 Signed-off-by: Raghavendra G <raghavendra@gluster.com> Reviewed-on: http://review.gluster.com/3044 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* mgmt/glusterd: unregister NLM service from portmapperRajesh Amaravathi2012-04-021-5/+14
| | | | | | | | | | | | Native NLM v4 service is unregistered from portmapper when Nfs is stopped. Change-Id: Id74cf9a31f6b00450504a1271bb47617aa5ec1fc BUG: 786345 Signed-off-by: Rajesh Amaravathi <rajesh@redhat.com> Reviewed-on: http://review.gluster.com/3052 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* performance/io-cache: Enhance state-dump outputshishir gowda2012-04-011-4/+10
| | | | | | | | | | | | | Along with weight, print uuid and path. If only the inode is cached, only weight, path and uuid will be printed. Change-Id: Ie7567832915319f26c6bab8a34576f6c43be7998 BUG: 806900 Signed-off-by: shishir gowda <shishirng@gluster.com> Reviewed-on: http://review.gluster.com/3014 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* mgmt/glusterd: Improvements for volume heal CLI.Vijay Bellur2012-04-014-23/+36
| | | | | | | | | | | | | | 1) Brought in a CLI improvement for volume heal cmd when it fails before initiating a transaction. 2) Minor improvements in UI messages related to self-heal-daemon. Change-Id: I983c1e54b48f2efe8793fa5b00fa50df2067683b BUG: 807261 Signed-off-by: Vijay Bellur <vijay@gluster.com> Reviewed-on: http://review.gluster.com/3057 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Pranith Kumar Karampuri <pranithk@gluster.com>
* cluster/afr: Fix frame leak in hardlink self-healPranith Kumar K2012-03-311-9/+2
| | | | | | | | | Change-Id: I8560fc2ba0d989eeb33226f8eb8c7fbad46dfe32 BUG: 765391 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/3056 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* glusterd: remove-brick validation behavior fixAmar Tumballi2012-03-311-14/+37
| | | | | | | | | | | | | | earlier one of the major validation case was missed if user provided a 'replica N' option for remove-brick where N is already existing replica count of the volume. This would have left the volume in inconsistent state, eventually crashing glusterd. Now fixed. Change-Id: I418f3bbb983d36aa51214c616a887e5a3ee98e74 Signed-off-by: Amar Tumballi <amarts@redhat.com> BUG: 803711 Reviewed-on: http://review.gluster.com/3050 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* features/locks: INIT the list before insert_mergePranith Kumar K2012-03-311-0/+1
| | | | | | | | | | Change-Id: I1612565e2bbf1811a85341016b86571bdf6e5062 BUG: 805994 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/3047 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/afr: Handle invalid inode in open_fd_fixPranith Kumar K2012-03-311-1/+5
| | | | | | | | | | Change-Id: Ie009fb4b8b7745ebd5b76f7a40287998be35eef3 BUG: 804914 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/3045 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* glusterd: Added init scripts like interface for gluster commandsKrishnan Parthasarathi2012-03-315-4/+287
| | | | | | | | | | | | | | | | | | | | | | | | | | Directories to hold the scripts mentioned above would be created for each gluster command under glusterd's 'working dir' as follows: (eg.) <glusterd-working-dir>/hooks/version/create/pre <glusterd-working-dir>/hooks/version/create/post where, version corresponds to the version of hook scripts. The scripts beginning with 'S' under the above directories would be run before (pre) and after (post) the volume created. The scripts would receive a command line argument as below: --volname=<volname> This can be optionally used by the script. Change-Id: Ia81b4ff9dd4477d99dd59b39c805aa645521edb0 BUG: 806996 Signed-off-by: Krishnan Parthasarathi <kp@gluster.com> Reviewed-on: http://review.gluster.com/3010 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Csaba Henk <csaba@redhat.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* glusterd: Used runner's RUN_PIPE to get inode size in xfs/ext3/ext4Krishnan Parthasarathi2012-03-311-32/+56
| | | | | | | | | Change-Id: I8057fd04baf36db86fe679a1beb86f76a1085b7c BUG: 786367 Signed-off-by: Krishnan Parthasarathi <kp@gluster.com> Reviewed-on: http://review.gluster.com/3033 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Csaba Henk <csaba@redhat.com>
* nlm: do not destroy the NULL client objectRaghavendra Bhat2012-03-301-1/+2
| | | | | | | | | | Change-Id: Ie1946b33353a91a266d2b21727d0c8eaa06af7f9 BUG: 808341 Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com> Reviewed-on: http://review.gluster.com/3046 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/stripe: minimum blk size log messageshylesh kumar2012-03-301-1/+2
| | | | | | | | | Change-Id: I16ba2e17796cd91e89a6216d4981c09f1bcdc5cb BUG: 807637 Signed-off-by: shylesh kumar <shylesh@gluster.com> Reviewed-on: http://review.gluster.com/3026 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/afr: Fix split-brain logPranith Kumar K2012-03-291-5/+8
| | | | | | | | | Change-Id: Id6869fb83bb083809a2a69227e233458e0b9e0bf BUG: 787653 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/3041 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/afr: Fix the split-brain logPranith Kumar K2012-03-291-3/+4
| | | | | | | | | | Change-Id: Ia4937dee009af7c945ec590a827c982b0e6ad307 BUG: 787653 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/3039 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* glusterd: add-brick: give proper reason in few error casesAmar Tumballi2012-03-291-0/+19
| | | | | | | | | | | | | had missed few 'else' cases, which were needed to send the right error message to the user regarding the reason for failure. Change-Id: I42d92782f4dec6f6bdffd42df254b7f38a715e36 Signed-off-by: Amar Tumballi <amarts@redhat.com> BUG: 772544 Reviewed-on: http://review.gluster.com/3032 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cli, glusterd : "volume status detail" fixesKaushal M2012-03-291-1/+1
| | | | | | | | | | | | | | | | Use _PATH_MOUNTED, which refers to /etc/mtab, in glusterd_add_brick_mount_details(), instead of _PATH_MNTTAB, which refers to /etc/fstab. More changes to cli to show available information. Change-Id: I9da97c0fec14b3372f281efc0be37b2799bc2d0d BUG: 808044 Signed-off-by: Kaushal M <kaushal@redhat.com> Reviewed-on: http://review.gluster.com/3037 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* core: fix all the iobuf related refs and unrefsAmar Tumballi2012-03-294-53/+98
| | | | | | | | | | | | | | reviewed the code against all the possible places where iobuf/iobref ref unref are present, and hopefully fixed most of the issues around memory leaks with respect to iobuf Change-Id: I9aa30326962991f8e23acedd389a0e962e097885 Signed-off-by: Amar Tumballi <amarts@redhat.com> BUG: 797875 Reviewed-on: http://review.gluster.com/2994 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cli,glusterd: more volume status improvementsKaushal M2012-03-294-161/+375
| | | | | | | | | | | | | | | | | | | | | | The major changes are, * "volume status" now supports getting details of the self-heal daemon processes for replica volumes. A new cli options "shd", similar to "nfs", has been introduced for this. "detail", "fd" and "clients" status ops are not supported for self-heal daemons. * The default/normal ouput of "volume status" has been enhanced to contain information about nfs-server and self-heal daemon processes as well. Some tweaks have been done to the cli output to show appropriate output. Also, changes have been done to rebalance/remove-brick status, so that hostnames are displayed instead of uuids. Change-Id: I3972396dcf72d45e14837fa5f9c7d62410901df8 BUG: 803676 Signed-off-by: Kaushal M <kaushal@redhat.com> Reviewed-on: http://review.gluster.com/3016 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Krishnan Parthasarathi <kp@gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* features/locks: check whether dict is NULL before refing itRaghavendra Bhat2012-03-291-1/+6
| | | | | | | | | Change-Id: I8906b80e584466aceb5a7402e3cee6ac2a552fab BUG: 808003 Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com> Reviewed-on: http://review.gluster.com/3036 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
* glusterd: Increase output_string size in glusterd_get_volopt_contentKaushal M2012-03-291-1/+1
| | | | | | | | | | | | | | The number of volume options with docs has increased, which causes the output_string to overflow and causes glusterd to crash. This change prevents glusterd from crashing. Change-Id: I79f8931f92ad55245aa547e7df15c9826537b615 BUG: 807914 Signed-off-by: Kaushal M <kaushal@redhat.com> Reviewed-on: http://review.gluster.com/3030 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/dht: send dict with linkto key for lookup in rmdir_readdirpshishir gowda2012-03-291-1/+23
| | | | | | | | | | | | | Without this, if there are any stale link files, rmdir fails with ENOEMPTY error. Change-Id: I7056b272122eb5540c22412856f46f638774c119 BUG: 807246 Signed-off-by: shishir gowda <shishirng@gluster.com> Reviewed-on: http://review.gluster.com/3035 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Amar Tumballi <amarts@redhat.com>
* cluster/afr: handle fstat failure in data-self-healPranith Kumar K2012-03-291-1/+1
| | | | | | | | | | | | | | | | The final fstat which makes the call_count 0 could be a failure. In that case the buf could either be NULL or buf is all zeros. If buf is NULL then it will crash, if it is all zeros buf->ia_type will be IA_INVAL and it proceeds to special file fix. sh->type is assigned with the ia_type of the file to be healed. I modified the code to depend on that instead. Change-Id: Icf7e19ff5908207128f2a1ee2963ad6b791c1ac5 BUG: 804645 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/3031 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* performance/quick-read: fix build errors.Raghavendra G2012-03-282-9/+13
| | | | | | | | | Change-Id: I8c4781bdd6c3cfd50cff95c31ecd6d653b916ac3 BUG: 765363 Signed-off-by: Raghavendra G <raghavendra@gluster.com> Reviewed-on: http://review.gluster.com/3028 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cli, glusterd: "volume set help" additionsKaushal M2012-03-282-21/+43
| | | | | | | | | | | | "auth.allow/reject" and "server.statedump" options are included in "volume set help" now. Change-Id: I7f9ba89d1782c26792347ffd2cd4042c3c396934 BUG: 783390 Signed-off-by: Kaushal M <kaushal@redhat.com> Reviewed-on: http://review.gluster.com/3025 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/afr: Handle afr data self-heal failures gracefullyPranith Kumar K2012-03-282-51/+138
| | | | | | | | | Change-Id: I5f91098111a8ca29982f3b4292e2109e4a36cce1 BUG: 765373 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/2662 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* performance/quick-read: make unlink aware of open-behind behaviour of ↵Raghavendra G2012-03-283-24/+336
| | | | | | | | | | | | | | | quick-read. quick-read should do an open of all the fds that are waiting for to be opened behind before doing an unlink on a path. Change-Id: I4c8d65c0511e8a37765ff57e30d99dd89b56c4fc BUG: 765363 Signed-off-by: Raghavendra G <raghavendra@gluster.com> Reviewed-on: http://review.gluster.com/506 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Shishir Gowda <shishirng@gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/stripe- do not ref if dict is nullshylesh kumar2012-03-281-1/+3
| | | | | | | | | Change-Id: Ifa4a5253482d837e2583de078e908b8c442f9dca BUG: 807214 Signed-off-by: shylesh kumar <shylesh@gluster.com> Reviewed-on: http://review.gluster.com/3024 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Shishir Gowda <shishirng@gluster.com>
* protocol/client: Set grace_timer_needed flag to false on a disconnect.Mohammed Junaid2012-03-281-0/+3
| | | | | | | | | | | | | | | | A grace timer is registered on a disconnect, but a reconnect timer sends a connect request after every 3sec and if the server is down, the client protocol receives disconnect and a grace timer will be registered which on timeout will increase the lk-version value. Its enough to register the grace timer once after the first disconnect and later just ignore other psuedo disconnects until the next successful reconnect. Change-Id: I7f34a98a79fdc25370ef15a6712bfa316a9b7c7c BUG: 807271 Signed-off-by: Mohammed Junaid <junaid@redhat.com> Reviewed-on: http://review.gluster.com/3020 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* performance/write-behind: queue setattr fop with writes always.v3.3.0qa32Raghavendra G2012-03-271-8/+1
| | | | | | | | | | | | | | stat returned in setattr_cbk can be cached by the kernel. Hence it is always necessary that we return correct stat, which implies that setattr should not be out of order with respect to write fops. Change-Id: I305feeb4802f8a41ffaf032100832cbd65dfc5c1 BUG: 765443 Signed-off-by: Raghavendra G <raghavendra@gluster.com> Reviewed-on: http://review.gluster.com/3011 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* glusterd: Fixed replace-brick commit_force algo.Krishnan Parthasarathi2012-03-273-26/+23
| | | | | | | | | | | | | | | | | | - commit force subcommand of replace-brick (rb) should be allowed even if source brick is (irrecoverably) offline. - modified rb_timer to be active only for start subcommand. This is important since, the rb timer event relies on src_brick and dst_brick objects to be 'alive' when it 'happens'. In the case of abort/commit/commit force it is very likely that src_brick and/or dst_brick objects could have been destroyed. Change-Id: Ib8b8a4d690fbdd6f99b8aff306490eb59c54a437 BUG: 772845 Signed-off-by: Krishnan Parthasarathi <kp@gluster.com> Reviewed-on: http://review.gluster.com/2620 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/dht : Acl fix for distribute directory selfhealKaushal M2012-03-263-6/+84
| | | | | | | | | | | | Send acl xattrs, if present in the xattrs returned during lookup, during directory self-heal. Change-Id: I5337bbd3f3963aeed500a8a552e5f6713089b53e BUG: 764787 Signed-off-by: Kaushal M <kaushal@redhat.com> Reviewed-on: http://review.gluster.com/737 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* mgmt/glusterd: native sort of info filesRajesh Amaravathi2012-03-261-13/+104
| | | | | | | | | | | | dependence on sort(1) has been removed, sorting of the files is now done using qsort(3). Change-Id: I90e162ba626506d01323ddfbfa594cf64b2e43c3 BUG: 790337 Signed-off-by: Rajesh Amaravathi <rajesh@redhat.com> Reviewed-on: http://review.gluster.com/2990 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* core: adding extra data for fopsAmar Tumballi2012-03-2292-5299/+7172
| | | | | | | | | | | | | with this change, the xlator APIs will have a dictionary as extra argument, which is passed between all the layers. This can be utilized for overloading in some of the operations. Change-Id: I58a8186b3ef647650280e63f3e5e9b9de7827b40 Signed-off-by: Amar Tumballi <amarts@redhat.com> BUG: 782265 Reviewed-on: http://review.gluster.com/2960 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* glusterd: bring back the 'non-synctask' behaviorAmar Tumballi2012-03-225-6/+6
| | | | | | | | | | | revert back to non synctask behavior Change-Id: Icfc9cbdeaf8dd6735df8366a6121c6598c56fade Signed-off-by: Amar Tumballi <amarts@redhat.com> BUG: 805802 Reviewed-on: http://review.gluster.com/3002 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* glusterd: fix bugs of syncop for operationsAmar Tumballi2012-03-224-178/+209
| | | | | | | | | | | | | | | | | | * free the stack created for synctask * use different key than 'operation' in dictionary as thats being used already by other glusterd operations * send proper frame to 'rpc_clnt_submit()' API, as it gets used internally * also make sure to destroy the above frame in all _cbk() * move everything specific to synctask into one file, so it is easy to maintain Change-Id: Ia1a4414fffec6f9e51700295947eea4a2104a8c2 Signed-off-by: Amar Tumballi <amarts@redhat.com> BUG: 805802 Reviewed-on: http://review.gluster.com/3000 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Krishnan Parthasarathi <kp@gluster.com>
* Add space for heal vol req log message.Rahul C S2012-03-211-1/+1
| | | | | | | | | | Change-Id: Iadf1d6304bcae8f666f2435ffa2d71c50dd1395a BUG: 764658 Signed-off-by: Rahul C S <rahulcs@redhat.com> Reviewed-on: http://review.gluster.com/2991 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Raghavendra Bhat <raghavendrabhat@gluster.com>
* glusterd: Disabled self-heal on clear-locks internal mountKrishnan Parthasarathi2012-03-212-5/+13
| | | | | | | | | | | | | | - Also, changed afr_get_xattr_clrlk_cbk to use dict_set_dynstr for clear-lock summary. Earlier, it was relying on 'str' passed from xlators below. Change-Id: I175f4542e6ef2c859c4811eecb9d8c5a7d25a283 BUG: 800779 Signed-off-by: Krishnan Parthasarathi <kp@gluster.com> Reviewed-on: http://review.gluster.com/2992 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Anand Avati <avati@redhat.com>
* glusterd: bring in feature to use syncop for mgmt opsAmar Tumballi2012-03-2112-30/+680
| | | | | | | | | | | | | | | | * new sycnop routines added to mgmt program * one should not use 'glusterd_op_begin()', instead can use the synctask framework, 'glusterd_op_begin_synctask()' * currently using for below operations: 'volume start', 'volume rebalance', 'volume quota', 'volume replace-brick' and 'volume add-brick' Change-Id: I0bee76d06790d5c5bb5db15d443b44af0e21f1c0 BUG: 762935 Signed-off-by: Amar Tumballi <amar@gluster.com> Reviewed-on: http://review.gluster.com/479 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* features/locks: Avoid race in pl_inode_getPranith Kumar K2012-03-211-21/+25
| | | | | | | | | Change-Id: I6c7b878fb43863137aac6552ee66f091534a6296 BUG: 770080 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/2998 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* features/locks: Avoid race in creating domainsPranith Kumar K2012-03-212-10/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Thread:1 | Thread:2 ----------------------------------------- 1) Does inodelk on inode1| 1) Does inodelk on inode1 2) It tries to get the | 2) It tries to get the dom object for the volume| dom object for the volume volname-replicate-0. But | volname-replicate-0. But the domain list is empty.| the domain list is empty. 3) Create the new domain | dom1 and add it to head | of the list of domains. | 4) inodelk happens on | 3) Create the new domain dom1 on range r1. | dom2 and add it to head | of the list of domains. | 4) inodelk happens on dom2 on range r2 5) unlock for r1 comes | 5) Unlock on r2 succeeds. on inode1. | 6) It tries to get the | domain, of the inodelk | for volume | volname-replicate-0, gets| dom2 but the lock on | range r1 is not present | so it fails. | subsequent inode lock, unlocks will keep happening on the domain dom2. The stale lock present in the dom1 on range r1 will live on. It wont cause any hangs, but the statedump will always be scary. Change-Id: I9adc120d33febf685b30859cc307768c2fc63ae5 BUG: 770080 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/2993 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Anand Avati <avati@redhat.com>
* protocol/client: memory leak fixes.v3.3.0qa31Raghavendra G2012-03-202-14/+4
| | | | | | | | | Signed-off-by: Raghavendra G <raghavendra@gluster.com> Change-Id: I804c934d79ed13ded9d1b741cef6597bc238e476 BUG: 803675 Reviewed-on: http://review.gluster.com/2987 Reviewed-by: Amar Tumballi <amarts@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* performance/read-ahead: destroy pages which survived from being flushedRaghavendra G2012-03-203-2/+10
| | | | | | | | | | | | | in flush_region because of frames waiting on them, after those frames are served. Change-Id: Ic8402f3da32058a1a11e62124b598b0ec3befeff BUG: 767948 Signed-off-by: Raghavendra G <raghavendra@gluster.com> Reviewed-on: http://review.gluster.com/2982 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* performance/io-cache: destroy the pages which survived inRaghavendra G2012-03-202-0/+6
| | | | | | | | | | | | __ioc_page_destroy because of non NULL waitq in ioc_fault_cbk. Change-Id: I00b83f8bf0a654b3a3a23cf0b25657fbfda17804 BUG: 767948 Signed-off-by: Raghavendra G <raghavendra@gluster.com> Reviewed-on: http://review.gluster.com/2981 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* protocol/server: remove the transport from the list irrespective of ↵Raghavendra Bhat2012-03-201-6/+11
| | | | | | | | | | | | | | | lock-heal is on/off Upon getting disconnect, remove the transport from the list of transports that protocol server maintains irrespective of whether lock-heal is on or off, since upon reconnect a new transport would be created. Change-Id: I18a269a93487d6e2cb11c345b6dde813d089809c BUG: 803815 Signed-off-by: Raghavendra Bhat <raghavendrabhat@gluster.com> Reviewed-on: http://review.gluster.com/2980 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amarts@redhat.com>
* features/marker: Replacing -1 with GF_CLIENT_PID_GSYNCD as part of code cleanup.Mohammed Junaid2012-03-201-1/+1
| | | | | | | | | Change-Id: Ia665060bff636b372904a726aba5249de9ade124 BUG: 769494 Signed-off-by: Mohammed Junaid <junaid@redhat.com> Reviewed-on: http://review.gluster.com/2971 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Csaba Henk <csaba@redhat.com>
* replicate: fix a glitch in up_count/down_count updates.Jeff Darcy2012-03-191-2/+24
| | | | | | | | | Change-Id: I4919a98191bf7fe5edad9a149a129bcd177cd4a8 BUG: 802522 Signed-off-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-on: http://review.gluster.com/2927 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* protocol/client: Handle failures in lock self healing gracefully (part2).Mohammed Junaid2012-03-191-49/+169
| | | | | | | | | | | | | | | During reopening of fd's and reacquiring of locks on the fd (after a reconnect), a release on a fd on which reacquiring of locks is in progress will free up fdctx. This patch will keep fdctx valid until the reacquiring of locks is in progress. Change-Id: I0fae27544a7f8ddaa26def4ee4d41a8a2b322521 BUG: 795386 Signed-off-by: Mohammed Junaid <junaid@redhat.com> Reviewed-on: http://review.gluster.com/2819 Reviewed-by: Amar Tumballi <amarts@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
generated by cgit (git 2.34.1) at 2024-05-21 18:10:03 +0000