summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* glusterd: Fix glusterd crash resulting from uninitialised pointerKrutika Dhananjay2012-08-021-0/+6
| | | | | | | | | | | | | | | | | | | | | | PROBLEM: glusterd crashes during replace-brick operation as the pointer old_brickinfo is dereferenced without being initialised. FIX: Have old_brickinfo initialised to point to the brick info of the brick being replaced before dereferencing it. TEST: Ran glusterd in debug mode before and after the fix; process crash ceased in the latter case. Change-Id: I53f21b5424f534bbdba3b3fc407d4501a2c0052f BUG: 844332 Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com> Reviewed-on: http://review.gluster.com/3777 Reviewed-by: Pranith Kumar Karampuri <pranithk@gluster.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/afr: Unwind with correct pre/post parent bufsPranith Kumar K2012-08-023-403/+147
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | RCA: In case of dir fops create, mknod, mkdir, link, symlink, rename if the fop fails on read-child then unwinds are happening with all-zero pre/post iatt-bufs. The bug occurs because the parent bufs are not saved if the response is not from read-child. Fix: Save the pre/post-bufs for the first response. If the response comes from read-child, overwrite whatever we have cached. Tests: Attached the mount process to gdb. Tested that the unwinds happen with proper pre/post iatt bufs in the following cases: 1) All success case 2) Failure on read-child 3) Failure on non-read-child 4) Failure on all children. Tested soft-link self-heal to test the change made in that. Tested errno ENOTEMPTY for rmdir, rename fops. Change-Id: I82882423d2d766b4f4a3044203bcb5dbcaee1755 BUG: 845242 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/3775 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* performance/write-behind: maintain a per-inode request queue instead of ↵Raghavendra G2012-08-022-596/+769
| | | | | | | | | | | | | | | | maintaining per-fd path based operations like stat etc, whose results will be affected by writes have to be ordered with writes. With request queues maintained in inode this can be done naturally, than when they are maintained per open fd. Change-Id: Ibdde3b81366f642d07531632fc9062cb44fad2e7 BUG: 765443 Signed-off-by: Raghavendra G <raghavendra@gluster.com> Reviewed-on: http://review.gluster.com/712 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* protocol/server: use pthread_mutex_trylock while dumping statedumpsRaghavendra Bhat2012-08-022-15/+37
| | | | | | | | | Change-Id: Iec6f539ee7d36c25aca3960abc6d2611707617c8 BUG: 843821 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com> Reviewed-on: http://review.gluster.com/3774 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* rpc-lib: fix rpc_clnt_reconfig log on remote_host changeyinyin2012-08-021-1/+1
| | | | | | | | | Change-Id: I5446d7fb975d4fb8105264e026fef165ae73a925 BUG: 764890 Signed-off-by: yinyin <maillistofyinyin@gmail.com> Reviewed-on: http://review.gluster.com/3773 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* cluster/afr: Handle child_up & fd not opened case in xactionPranith Kumar K2012-08-011-7/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | RCA: When an fd is opened while a brick is down, after the brick comes back up afr issues open on the other brick. It can fail for a number of reasons (enoent etc). While the system is in that state, inode/entrylks pre-op happen only on the brick that is up and fd is opened for fd-fops. post-op should consider only the bricks where both pre-op and fop succeeded as success, rest of them as failures. Code now marks only the children that are down as failures as opposed to child_down & fd-not-opened. This makes change-log appear as success on the subvolume where we did not do any fop leading to no change-log but differences in data/metadata for reg-files. Fix: Mark non-participants of fop as failure. This is tracked in transaction.pre_op[]. Tests: Simulated the scenario using err-gen on top of one of the client xlator which fails all fops always. Performed fops and the changelog represented pending fops on the brick with err-gen loaded. Tested the case of brick down and perform entry/metadata/data operations to confirm they still work as expected. Change-Id: I41905936126b19abba56ca581c0301a894507e1a BUG: 844987 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/3765 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* mem-pool: remove 'if' by asserting condition directly.Jules.Wang2012-08-011-22/+11
| | | | | | | | | | | | | | | | | | | before: if (!condition) { GF_ASSERT (0); } after: GF_ASSERT (condition); BUG: 764890 Change-Id: I9894fcc609579f5efd3f98dbff41ca1487ce6dd2 Signed-off-by: Jules Wang <lancelotds@163.com> Reviewed-on: http://review.gluster.com/3762 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cli: more clear separation of error messagesAmar Tumballi2012-08-011-70/+87
| | | | | | | | | | | | | | | | | the intention is to make CLI output more script friendly. use ':' as field separator per command and each field should be meaningful. this is phase-1 of the change, and hence RFC, once I get review comments, will be doing more changes, if required Change-Id: I11e70084f78a197ac50f9401dd7af312e63647ef Signed-off-by: Amar Tumballi <amarts@redhat.com> BUG: 815194 Reviewed-on: http://review.gluster.com/3753 Reviewed-by: Raghavendra Bhat <raghavendra@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* supportability: print all missing framesAmar Tumballi2012-08-012-8/+8
| | | | | | | | | | | | | | | | | | | earlier, the statedump and backtrace were printing frames only if it matched type GF_FOPs or GF_MGMTs (GF_MGMT is not in use now). This made all the frames from glusterd, or any other frames which got created outside the scope of 'fops' not getting printed. now, made two classifications, one as FOPs (which gets logged with the proper name), and another set is !FOPs (ie, else part), which gets logged with only the op number. Change-Id: Ifa18cdb75f7afefe6bb489d319b589cda8e64c1d Signed-off-by: Amar Tumballi <amarts@redhat.com> BUG: 845114 Reviewed-on: http://review.gluster.com/3768 Reviewed-by: Raghavendra Bhat <raghavendra@redhat.com> Reviewed-by: Anand Avati <avati@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* core: remove libglusterfs/src/scheduler.[ch]Amar Tumballi2012-08-013-116/+1
| | | | | | | | | | | | * as its not part of the build * confuses with some TAGS and 'git greps' Change-Id: I1f97307a8ce486ba56a4a3ad0151276b0a72e096 Signed-off-by: Amar Tumballi <amarts@redhat.com> BUG: 764890 Reviewed-on: http://review.gluster.com/3769 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* performance/read-ahead: use pthread_mutex_trylock to hold mutex in statedumpsRaghavendra Bhat2012-08-011-11/+18
| | | | | | | | | | Change-Id: I4491c96fc7d9879ba9040dcda3700c81d99ba184 BUG: 843796 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com> Reviewed-on: http://review.gluster.com/3766 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Pranith Kumar Karampuri <pranithk@gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* event-history: implement gf_log_eh for saving eventsRaghavendra Bhat2012-08-012-6/+47
| | | | | | | | | | | Change-Id: I2ddf4e7849bda91bb99423e27082c2bae6800a70 BUG: 795419 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com> Reviewed-on: http://review.gluster.com/3532 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Pranith Kumar Karampuri <pranithk@gluster.com> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Anand Avati <avati@redhat.com>
* mount/fuse: initialize entry and negative timeouts correctly.Raghavendra G2012-08-011-2/+2
| | | | | | | | | | | | There was a typo which resulted in updation of attribute_timeout for both options. Change-Id: I6e212eaf3570cadde88d2a2c3b44121c893c9427 BUG: 804592 Signed-off-by: Raghavendra G <raghavendra@gluster.com> Reviewed-on: http://review.gluster.com/3763 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amarts@redhat.com>
* protocol/client: Fix negative return in client_setvolumeKrutika Dhananjay2012-08-011-2/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | PROBLEM: The function dict_serialized_length could, owing to an error, return a negative integer (-EINVAL) that gets assigned to an unsigned int member 'dict_len' of gf_setvolume_req structure. FIX: Hold the value returned by dict_serialized_length in local variable ret (which is a signed int). Test if ret is negative, in which case the control would anyway branch to the label fail where the function returns. Otherwise dict_len is assigned with ret, in turn giving a more meaningful value to the attribute length. TEST: Attached gdb to glusterfs mount process, set breakpoint at client_setvolume, forced dict_serialized_length to return -EINVAL (indirectly by forcing _dict_serialized_length to return -EINVAL after setting count to -1 within its body) and checked the value of ret (which is now sure to contain a negative value) whose value will be appropriately tested to decide the next course of action within client_setvolume: whether to simply exit due to an error or execute the subsequent statements. Change-Id: Ib22ad8f30d8ae04acaf2ff5bfee9c348a2c47148 BUG: 789278 Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com> Reviewed-on: http://review.gluster.com/3755 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/afr: Handle failures in fop_cbk gracefullyPranith Kumar K2012-07-311-31/+46
| | | | | | | | | | | | | | | | | | | | | | | | | RCA: Afr crashes when a last fop response fails and 'fop output' arguments are NULL. Afr does not handle these gracefully. Fix: Changed the fops to not access the 'fop output' arguments in case of failures. Tests: Changed afr wind_cbk code to fail the last response by setting op_ret as -1 and op_errno as ENOMEM and setting all other output variables as NULL to test the change. Removed the code to verify success cases. No crashes or errors seen. Change-Id: Iad9bc54db093a162f85bfb8dbeeda5b95acd21d8 BUG: 844689 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/3760 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Anand Avati <avati@redhat.com>
* Fix issues in make distM S Vishwanath Bhat2012-07-311-2/+2
| | | | | | | | | | Change-Id: I2fb3b0b35e2b5fdb08ee1857bf31ec8f65889306 BUG: 844669 Signed-off-by: M S Vishwanath Bhat <msvbhat@gmail.com> Reviewed-on: http://review.gluster.com/3759 Reviewed-by: Amar Tumballi <amarts@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/afr: update loc inode after inode_linkPranith Kumar K2012-07-311-12/+24
| | | | | | | | | | | | | | | | | | | | | | | | RCA: inode passed to inode_link is not assigned any gfid if the inode with that gfid is already linked, so loc for opendir does not have a valid inode Fix: Use the linked_inode returned by inode_link in the loc to perform further operations on the entry. Tests: Checked that opendir comes with an loc with valid inode. Checked that re-opendir happens successfully. Tested index, full self-heal work fine with the fix. Change-Id: Idf4ced4cc2320133744962059d363e373af0e5ec BUG: 826580 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/3748 Reviewed-by: Raghavendra Bhat <raghavendra@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* rpc/socket: finish initialization in own threadJeff Darcy2012-07-301-7/+47
| | | | | | | | | | | | ...if we have one, that is. This addresses Avati's review comment on ea7759f1240b1e97684273b9369472695173a66. Change-Id: I71a272f7eeff8279981a75241d60620ca4975c99 BUG: 764731 Signed-off-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-on: http://review.gluster.com/3701 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* glusterd: Modifed replace-brick mount cleanup to use lazy umountKrishnan Parthasarathi2012-07-301-94/+135
| | | | | | | | | Change-Id: I83a985967cbaee4927f1b8ca4bc01c9fb0171522 BUG: 806890 Signed-off-by: Krishnan Parthasarathi <kparthas@redhat.com> Reviewed-on: http://review.gluster.com/3143 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/stripe: handle short writes and errors in writev callbackBrian Foster2012-07-303-38/+106
| | | | | | | | | | | | | | | | | | | | | | | cluster/stripe write callback handling is broken in the event of server side errors and short writes due to crudely summing up the return values from each node. This can produce incorrect results or cause an application to rewrite the wrong portions of a buffer in an attempt to handle this condition. Modify cluster/stripe writev handling to record the requested size of each write and use this data to return the number of consecutive bytes written from the original request. This allows an application to retry a write at the point of error (and potentially consume said error). BUG: 809975 Change-Id: Ic35cb1e092c29545205aa32e352485c507534ce0 Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-on: http://review.gluster.com/3700 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Shishir Gowda <sgowda@redhat.com> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Anand Avati <avati@redhat.com>
* performance/io-threads: dump the io-threads specific information in statedumpRaghavendra Bhat2012-07-302-2/+52
| | | | | | | | | Change-Id: Ib9b9eab2d2e90fce21ef32f0f884101938978859 BUG: 843377 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com> Reviewed-on: http://review.gluster.com/3736 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* geo-rep / gsycnd: restore the generic signature for sendmark_regularCsaba Henk2012-07-301-3/+3
| | | | | | | | | | | | | | | | | Earlier fixes to 842330 changed the generic (*a, **kw) signature, although that was not related to the issue. We restore the generic signature as it was used for a reason (proxy methods that do none or only algebraic transformations on passed arguments idiomatically have generic signature, both to serve as visual cue and agnosticism wrt. the inner API). Change-Id: Ib609a3a58be53d78b7f1221a3c162c6aec8fd488 BUG: 842330 Signed-off-by: Csaba Henk <csaba@redhat.com> Reviewed-on: http://review.gluster.com/3754 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* libglusterfs/statedump: add start and end markers in the statedump fileRaghavendra Bhat2012-07-301-2/+37
| | | | | | | | | | | | | Also do not overwrite the statedump file if statedump is taken on the same process again. Instead add the timestamp to the filename Change-Id: I451a0decf2dad5c398631cab1fc2a69811c69af8 BUG: 843408 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com> Reviewed-on: http://review.gluster.com/3738 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Anand Avati <avati@redhat.com>
* performance/io-threads: Provide option to turn off least-priorityPranith Kumar K2012-07-303-5/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | RCA: In cases when self-heal is in progress, self-heal fops are starved because of least-priority. This affects other fops with conflicting inode, entry locks with self-heal. Fix: This patch provides configuring enable/disable of least-priority. Additional changes: Moved RCHECKSUM fop to low instead of least because it will still affect the performance of other fops if RCHECKSUM is in LEAST priority. Tests: Tested that the enabling/disabling of fops is working fine. Tested that RCHECKSUM fop priority is assigned LOW when least-priority is disabled. Change-Id: I8d72872b33e2ac7e1ad3afb27582682b0eb98a80 BUG: 843704 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/3743 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* volgen: add option to add trace/error-gen in client volume fileAmar Tumballi2012-07-291-0/+6
| | | | | | | | | | | | | | | | bash# gluster volume set <VOLNAME> debug.trace client bash# gluster volume set <VOLNAME> debug.error-gen client would add the corresponding translator in client volume file just below io-stats translator. (gets added even in nfs volfile) Change-Id: I698eb9b348c59aaff9967a4e238fabfc40e69a11 Signed-off-by: Amar Tumballi <amarts@redhat.com> BUG: 782262 Reviewed-on: http://review.gluster.com/3699 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Kaushal M <kaushal@redhat.com> Reviewed-by: Anand Avati <avati@redhat.com>
* glusterd: Persisted hooks friendly user.* keysKrishnan Parthasarathi2012-07-294-48/+51
| | | | | | | | | | | | - Fixed validation of user.* keys in presence of multiple key, value pairs in a single volume set command Change-Id: I5b96de2d009fbc79772121308d9b4c0a552bac52 BUG: 825902 Signed-off-by: Krishnan Parthasarathi <kparthas@redhat.com> Reviewed-on: http://review.gluster.com/3715 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* features/locks: Fix statedump codePranith Kumar K2012-07-294-74/+52
| | | | | | | | | | | | | | | | | | | | | | | | RCA: Taking blocking mutex/spin locks lead to dead locks because of the locking order in statedumps. Also we were asked to remove gf_logs if possible to avoid extra cost in signal handlers. Fix: changed blocking mutes/spin locks to their non-blocking variants. Removed gf_logs in locks xlator statedump code-path. Tests: State-dump success cases are working fine. Triggered try-lock failures by putting statedumps in a while loop. In parallel did chown of the same file in a while loop. Change-Id: I81539a62f8216f267f57bb703ef132c85bfd557d BUG: 843781 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/3747 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cli: Prevent creation of volumes with tokens as volnamesPranith Kumar K2012-07-291-3/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | RCA: yyparse confuses volnames with tokens when any of the tokens 'volume', 'type', 'subvolumes', 'option', 'end-volume' are used as volnames. This happens because io-stats xlator name is same as volname in fuse volfile. Both nfs, fuse volfiles are affected by this problem. Fix: We could fix this also by changing io-stats xlator name to something other than 'volname'. But I am worried of the backward compatibility issues it may introduce. Disallowing creation of volumes with tokens as volname seems like a safer fix. Tests: All volume creation operations with tokens as volname give invalid volume name error. Change-Id: Ifc63a5c31375e92541b954ec133aa3c8e6a56a02 BUG: 844030 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/3745 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Anand Avati <avati@redhat.com>
* glusterd: Ensured 'store' data reaches disk.Krishnan Parthasarathi2012-07-291-15/+55
| | | | | | | | | | | | | - Opened temporary file(s) with O_SYNC flag to avoid explicit fsync'ing. - Sync'd directory entry after creation and rename of 'store' files. - Thanks to Jeff Moyer's article on http://lwn.net/Articles/457667/ Change-Id: I68a8672dc6a0b24d128de53f3b60c74dd08d8ab8 BUG: 765434 Signed-off-by: Krishnan Parthasarathi <kparthas@redhat.com> Reviewed-on: http://review.gluster.com/3726 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* Warnings: Fix set, unused variable warningsPranith Kumar K2012-07-292-5/+0
| | | | | | | | | Change-Id: If7ee85aa0d93f9018a76e7f60fcb7dcaf849dbdc BUG: 839950 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/3742 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amarts@redhat.com>
* acl: enable handling of FMODE_EXEC flagAmar Tumballi2012-07-273-1/+18
| | | | | | | | | | | | | | | | | | | | | | on linux systems, with open(), we can get below flag as per 'linux/fs.h'. /* File is opened for execution with sys_execve / sys_uselib */ Instead of adding '#include <linux/fs.h>, its better to copy this absolute number into other variable because then we have to deal with declaring fmode_t etc etc.. With the fix, we can handle the file with '0711' permissions in the same way as backend linux filesystems. Change-Id: Ib1097fc0d2502af89c92d561eb4123cba15713f5 Signed-off-by: Amar Tumballi <amarts@redhat.com> Reviewed-on: http://review.gluster.com/3739 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Niels de Vos <ndevos@redhat.com> Tested-by: Niels de Vos <ndevos@redhat.com> Reviewed-by: Anand Avati <avati@redhat.com>
* gfapi: unglobalize oldvol{file,len}Anand Avati2012-07-273-12/+35
| | | | | | | | | | | Multiple initialization of volumes works fine now. Change-Id: Ibc16957b39325817bc6f9ab90ebac9cefdafabf8 BUG: 839950 Signed-off-by: Anand Avati <avati@redhat.com> Reviewed-on: http://review.gluster.com/3734 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amarts@redhat.com>
* cluster/afr: Modified split-brain handlingPranith Kumar K2012-07-265-57/+146
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | RCA The bug is observed because the decision to mark a file in split-brain is taken outside appropriate locks. Lookup gathers xattrs outside any lock. The xattrs being in split-brain in lookup should only be taken as a hint. Appropriate inodelks should be taken before confirming a split-brain. Self-heal confirms this at the moment. If data/metadata self-heal is turned off, inspecting of xattrs could not be performed so split-brain behavior does not work correctly if the self-heal options are turned off. Fix Self-heals are launched to inspect xattrs even when the data/metadata self-heal options are turned off. The decision to heal data/metadata after the xattrs are inspected is based on whether the options are turned on/off. So decision to set/reset split-brain flag is taken inside appropriate locks. Testcases: tests 33-36 in https://github.com/pranithk/gluster-tests/blob/master/afr/self-heal.sh Change-Id: Ia8aeab08208b50c06609ad35a9d72f3d553ee343 BUG: 833727 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/3626 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/afr: Filter O_TRUNC in afr-fix-openPranith Kumar K2012-07-251-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | RCA: When open was done while a brick is down, afr opens the file after the brick comes backup. If this happens after the self-heal on the file is completed by self-heald etc, the file will end up in truncated state. Fix: Filter O_TRUNC while afr-fix-open because afr_open turns O_TRUNC into truncate transaction, so there will be pending changelog for the subvolume on which open fails. Testing: Had to simulate the race by stopping fix-open until self-heald completes self-heal on the file after brick online. Change-Id: I32759cc37f4bb34f206d01606a279f17b246dba4 BUG: 841840 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/3705 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Anand Avati <avati@redhat.com>
* performance/write-behind: detect short writes and pend an EIO errorBrian Foster2012-07-251-1/+15
| | | | | | | | | | | | | | | Write-behind returns write requests immediately and queues the request in memory for merging, etc. If a write is incomplete, pend an EIO error for the next fop. This ensures that write failures are not silent and potentially ignored. BUG: 809975 Change-Id: I0e0e6c8e710efab58ccfaf746501d00e459eb7ef Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-on: http://review.gluster.com/3712 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Anand Avati <avati@redhat.com>
* performance/write-behind: preserve lk-owner while syncing writes.Raghavendra G2012-07-251-66/+143
| | | | | | | | | | | | | - This patch also makes syncing of non-overlapping but consecutive writes parallel. Till now only contiguous writes were synced parallely. Change-Id: Icf0d5ea373f30c79fcdc90ba44b7e7a1bc5f0111 BUG: 765141 Signed-off-by: Raghavendra G <raghavendra@gluster.com> Reviewed-on: http://review.gluster.com/269 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster: fix crash on link of named pipe in stripe/replicate volBrian Foster2012-07-252-12/+11
| | | | | | | | | | | | | | | | | | | | | | | | A crash occurs when attempting to link a named pipe on a striped, replicated volume. The cause for this crash is attempting to deref a NULL inode pointer in stripe_link_cbk(). The RCA for this bug uncovered a couple of problems: - AFR ignores the inode pointer it receives on failure (returning NULL). - stripe assumes the inode pointer is valid on failure. Either one of these changes addresses the crash, but this patch includes both changes. AFR is modified to pass along the inode pointer it receives (which could still be NULL). stripe is modified to not assume the inode pointer is valid on fop failure. BUG: 842825 Change-Id: I9cb2cc918552620929c3ecbd69bc66d4635eafdc Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-on: http://review.gluster.com/3727 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/afr: Perform data self-heal for non regular filesPranith Kumar K2012-07-255-217/+162
| | | | | | | | | | | | | | | | | | | | | | | | | RCA: Data self-heal for non regular files open the files and then proceeds using that fd. This approach does not work for symlinks because open on symlink opens the file resolved by it. Fix: If the file is not a regular file then perform self-heal using loc. It needs to get 'big' lock and then perform lookup to get changelog then erase data part of chagelog, then unlock. Test cases: Automated at https://github.com/pranithk/gluster-tests/blob/master/afr/special-file-self-heal-test.sh Change-Id: I924a922f5135872efe2cccf2e712ada082c5689f BUG: 811317 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/3724 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: don't fail if no fctx on a non-regular fileBrian Foster2012-07-251-10/+14
| | | | | | | | | | | | | cluster/stripe broke directory rename. Only check for fctx on regular files. BUG: 842652 Change-Id: I8a1e7ff30d57c994082cb10471f610023713ee53 Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-on: http://review.gluster.com/3720 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Anand Avati <avati@redhat.com>
* features/index: make pthread_attr_t local to init()Pranith Kumar K2012-07-252-8/+6
| | | | | | | | | | | | | | | | | RC & FIX: No need to carry around pthread_attr_t in index_priv. So made it local to init() Tests: stepped through init() in gdb, it succeeded. Change-Id: I0525ac0676f9a329fccb0fd064933594ec117261 BUG: 843071 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/3729 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Kaleb KEITHLEY <kkeithle@redhat.com> Reviewed-by: Anand Avati <avati@redhat.com>
* cluster/distribute: Suppress user xattr mismatch log messageshishir gowda2012-07-251-1/+1
| | | | | | | | | | | | | | | | | Changing the log-level to DEBUG. Xattr mismatch can occur when parallel setxattr's race, or when one of the bricks was down. A subsequent setxattr will fix the condition when all the subvols are up. In this case, the 'user.swift' xattr used by ufo was out of sync, but did not cause any other error. Change-Id: I6fdff78869b8ff72c305bbe122033e6c1d9d3cff BUG: 838197 Signed-off-by: shishir gowda <sgowda@redhat.com> Reviewed-on: http://review.gluster.com/3722 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Mohammed Junaid <junaid@redhat.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* features/index: Fix race in this->private initializationPranith Kumar K2012-07-251-5/+27
| | | | | | | | | | | | | | | | | | | | | | | | | | RCA: In index_worker function at the time of assigning priv from this->private, this->private may not be set in init() function of index xlator. Fix: Set this->private before creating the thread. Additional Changes: Added code to handle error path completely. Test cases: Attached the process to gdb and simulated failures. executed fini in gdb using call fini(this) after init was successful. Change-Id: I1874a30d009a35352173b827574cf83daf431453 BUG: 843071 Signed-off-by: Pranith Kumar K <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/3728 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* gsyncd / geo-rep: Fix sendmark() invocation for Normal MixinVenky Shankar2012-07-251-1/+1
| | | | | | | | | Change-Id: I0ae81ab01418becba83e401ec36c6db5323945e8 BUG: 842330 Signed-off-by: Venky Shankar <vshankar@redhat.com> Reviewed-on: http://review.gluster.com/3725 Tested-by: Vijay Bellur <vbellur@redhat.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* libglusterfs: fix creation of spurious root inodesRaghavendra Bhat2012-07-241-5/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | Suppose lookup is sent on a newly created inode (gfid is root gfid), and in the call back inode_link is done, where if there is an inode already present in the inode table for that gfid, then we start using the present inode and the newly created inode supposedly gets destroyed when refcount comes down to zero through inode_unref calls. But inodes with root gfid are not unrefed. Now since in inode_link, the gfid has already been copied to the new inode and later the actual inode is found from the inode table, the new inode does not gets unrefed if its gfid is root gfid. Thus those spurious inodes with root gfid will be present throughout the lifetime of the filesystem. To fix this first search the inode table for an inode, with the gfid, present in the stat structure received as an argument. If the inode is not found, then copy the gfid to the newly created inode. Now whenever unref is called on on the new inode it gets unrefed and thus eventually gets destroyed Change-Id: I0f25f0a8dca3245abda1c322c216f063b52cf842 BUG: 841188 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com> Reviewed-on: http://review.gluster.com/3716 Reviewed-by: Amar Tumballi <amarts@redhat.com> Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* Fixes for gsyncd / geo-rep and FUSE listxattrVenky Shankar2012-07-242-8/+32
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch fixes two problems with recent changes to Geo-Replication First: ------ Recent changes to geo-replication relies on Rsync to tranfer extended attributes. Essentially Rsync would invoke a listxattr() and then getxattr() the set reutrned by listxattr() and finally transfer it to the remote slave. Xattrs like security.selinux would create problems as they are not allowed to be set explicitly (unless there's a rule that allows this). So, to make Rsync behave sanely we filter out all "*.selinux*" xattrs from listxattr() (which is getxattr() with ->name as NULL). Second: ------- Python's "if {..} else {..}" shortcut ".. and .. or .." was misused here. This is a straightforward fix by interchanging last two variables (classes in this case). Also fix a typo in sendmark_regular() definition. Change-Id: I097b5f5d88a36c7eef5560a78d4332948a545942 BUG: 842330 Signed-off-by: Venky Shankar <vshankar@redhat.com> Reviewed-on: http://review.gluster.com/3714 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* afr: pass back xdata in createBrian Foster2012-07-231-2/+5
| | | | | | | | | | | | | | | A striped, replicated volume spits an error on file creation because stripe requires xdata to process stripe information and AFR isn't passing it back. This fix was suggested by Amar Tumballi. BUG: 842373 Change-Id: Ia7063590ca5e873d4a4e155989cf067e8a07501f Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-on: http://review.gluster.com/3713 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@redhat.com>
* libgfxdr.so: change licenseAmar Tumballi2012-07-2026-366/+130
| | | | | | | | | Change-Id: I049441e465127195c192443ab0234548d2449e50 Signed-off-by: Amar Tumballi <amarts@redhat.com> Reviewed-on: http://review.gluster.com/3704 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Bharata B Rao <bharata.rao@gmail.com> Reviewed-by: Kaleb KEITHLEY <kkeithle@redhat.com>
* gsyncd / geo-rep: Fix typo in 'purge' flowVenky Shankar2012-07-201-1/+1
| | | | | | | | | Change-Id: I6c329b895178545d16b0cb9f01ad116f5342f752 BUG: 841855 Signed-off-by: Venky Shankar <vshankar@redhat.com> Reviewed-on: http://review.gluster.com/3706 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* geo-rep / gsyncd: convert ignore-deletes to a mixin tooCsaba Henk2012-07-191-3/+16
| | | | | | | | | Change-Id: I164a1d1dd5f15569afd6806834119a6844949df0 BUG: 841062 Signed-off-by: Csaba Henk <csaba@redhat.com> Reviewed-on: http://review.gluster.com/3684 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vbellur@redhat.com>
* geo-rep / gsyncd: add support for sending xtimes through rsyncCsaba Henk2012-07-193-14/+35
| | | | | | | | | | | | | | Note that in said mode metadata synchronization is best effort: rsync syncs metadata at last so if rsync is interrupted in between xattr sync and metadata sync stages, then file will be considered in sync Change-Id: I1c75eab33b0a1000abf3ad36b2d484a89eeda1bd BUG: 841062 Signed-off-by: Csaba Henk <csaba@redhat.com> Reviewed-on: http://review.gluster.com/3683 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Venky Shankar <vshankar@redhat.com>
generated by cgit (git 2.34.1) at 2024-11-20 03:36:11 +0000