summaryrefslogtreecommitdiffstats
path: root/xlators/cluster/dht
Commit message (Collapse)AuthorAgeFilesLines
* iatt: changes across the codebaseAnand V. Avati2010-03-168-237/+241
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - libglusterfs -- call-stub -- inode -- protocol - libglusterfsclient - cluster/replicate - cluster/{dht,nufa,switch} - cluster/unify - cluster/HA - cluster/map - cluster/stripe - debug/error-gen - debug/trace - debug/io-stats - encryption/rot-13 - features/filter - features/locks - features/path-converter - features/quota - features/trash - mount/fuse - performance/io-threads - performance/io-cache - performance/quick-read - performance/read-ahead - performance/stat-prefetch - performance/symlink-cache - performance/write-behind - protocol/client - protocol/server - storage-posix Signed-off-by: Anand V. Avati <avati@blackhole.gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 361 (GlusterFS 3.0 should work on Mac OS/X) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=361
* send lookup on every subvolume if local->hashed_subvol is NULL in nufa/switchAmar Tumballi2010-03-122-4/+6
| | | | | | | | Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 114 (Creating files or directories gives "Invalid argument") URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=114
* dht: preserve and return proper pre/postparent structures during unlinkAnand Avati2010-03-041-4/+7
| | | | | | | | Signed-off-by: Anand V. Avati <avati@blackhole.gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 689 (Segfault (11) on op UNLINK) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=689
* distribute: cleanup intermediate hardlink and linkfile on failed renameAnand Avati2010-03-041-12/+78
| | | | | | | | | | | | | implement a generic cleanup function which handles cleanup of temporary linkfiles and hardlinks created as part of the rename transaction, and callable from any stage of failure. Signed-off-by: Anand V. Avati <avati@blackhole.gluster.com> Signed-off-by: Shehjar Tikoo <shehjart@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 597 (miscellaneous fixes for xlators to work well with NFS xlator) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=597
* distribute: perform self-heal as rootShehjar Tikoo2010-03-043-0/+37
| | | | | | | | | | | prerserve original frame uid and gid and perform self-heal by changing uid/gid to root (0) temporarily. Signed-off-by: Shehjar Tikoo <shehjart@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 597 (miscellaneous fixes for xlators to work well with NFS xlator) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=597
* distribute, nufa: return 0 mtimes in preparent/postparentShehjar Tikoo2010-03-044-2/+58
| | | | | | | | | | | | | | Returning 0 times means the values are unreliable and not to be trusted for the purposes of caching. This is a temporary fix till we bring in proper times handling in DHT to return consistant values for parent inodes. Signed-off-by: Shehjar Tikoo <shehjart@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 597 (miscellaneous fixes for xlators to work well with NFS xlator) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=597
* distribute: Save ino/dev from first subvolume during mkdirShehjar Tikoo2010-03-041-0/+6
| | | | | | | | | | | | | | A self-heal on directory creation results in the subsequent call to dht_selfheal_new_directory to happen only for the last subvolume. But when this function is called for the last subvol it is possible that the st_ino returned in this function's callback is the st_ino from the hashed subvolume(assigned in dht_mkdir_hashed_cbk). Signed-off-by: Shehjar Tikoo <shehjart@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 597 (miscellaneous fixes for xlators to work well with NFS xlator) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=597
* distribute: Restore parent ino from local in symlinkShehjar Tikoo2010-03-041-0/+8
| | | | | | | | Signed-off-by: Shehjar Tikoo <shehjart@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 597 (miscellaneous fixes for xlators to work well with NFS xlator) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=597
* distribute: Restore inode from saved ino on readvShehjar Tikoo2010-03-041-1/+18
| | | | | | | | Signed-off-by: Shehjar Tikoo <shehjart@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 597 (miscellaneous fixes for xlators to work well with NFS xlator) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=597
* distribute: Respect end-of-dir on readdir only for last subvolShehjar Tikoo2010-03-043-0/+40
| | | | | | | | | | | | | | | | We need to ensure that only the last subvolume's end-of-directory notification is respected so that directory reading does not stop before all subvolumes have been read. That could happen because the posix for each subvolume sends a ENOENT on end-of-directory but in distribute we're not concerned only with a posix's view of the directory but the aggregated namespace' view of the directory, which is maintained by distribute. Signed-off-by: Shehjar Tikoo <shehjart@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 597 (miscellaneous fixes for xlators to work well with NFS xlator) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=597
* distribute: Return pre/post-parent on successful unlinkShehjar Tikoo2010-03-041-0/+2
| | | | | | | | Signed-off-by: Shehjar Tikoo <shehjart@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 597 (miscellaneous fixes for xlators to work well with NFS xlator) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=597
* distribute: Return ino from the first-up subvolume in lookupShehjar Tikoo2010-03-041-1/+1
| | | | | | | | | | | | | | Not returning this from the first-up subvolume results in different inode numbers being returned for a directory when the same directory is read again through the NFS readdir request. Signed-off-by: Shehjar Tikoo <shehjart@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 597 (miscellaneous fixes for xlators to work well with NFS xlator) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=597
* dht: fix READDIR semantics which got broken in ↵Anand Avati2010-02-231-34/+24
| | | | | | | | | | 512dcad823feb199d78f5b66a09bc404ccb899e8 Signed-off-by: Anand V. Avati <avati@blackhole.gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 584 (automatically configure 'lookup-unhashed' option) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=584
* dht: fix leak introduced in 137f94ed123b1c7c7ada541aaa599679e19888f6Anand Avati2010-02-221-1/+4
| | | | | | | | | | | dht_layout_t was getting leaked in dht_readdirp_cbk when dht_layout_get was called but was not unref'ed Signed-off-by: Anand V. Avati <avati@blackhole.gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 584 (automatically configure 'lookup-unhashed' option) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=584
* dht: unlink stale linkfiles in rmdir to prevent ENOTEMPTYAnand Avati2010-02-223-14/+244
| | | | | | | | | | | | Thanks to He Xaobing <allreol@gmail.com>, this patch is inspired by http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=188#c2 Signed-off-by: Anand V. Avati <avati@blackhole.gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 188 ([ glusterfs 2.0.6rc2 ] - "Directory not empty" on rm -rf) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=188
* dht: Remove pointer casting in layout handlingAnand Avati2010-02-072-9/+15
| | | | | | | | | | | | | | | | | | | | | | | | | Small changes by Anand Avati <avati@gluster.com> over original submission: Remove pointer casting (copy data to proper memory before using it). Fixes bug #493. Hello, See bug #493 for more information on this patch. http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=493#c2 This patch assumes that the scope of disk_layout is local wherever it is used. Seems correct, not really checked, though. Greetings, Hraban Luyat Signed-off-by: Hraban Luyat <hraban@0brg.net> Signed-off-by: Anand V. Avati <avati@blackhole.gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 493 (tcp + dht + armv5tel: ???brick: disk layout has invalid count 29696???) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=493
* 'lookup-unhashed' option of distribute should be 'auto' by default.Amar Tumballi2010-02-065-22/+76
| | | | | | | | | | | | | | * Added 'auto' option, older boolean options works as they used to. * This option should make 'create' rate faster, also handles self-healing of linkfile properly in case of scaling to more servers or filesystem is getting full. Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 584 (automatically configure 'lookup-unhashed' option) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=584
* cluster/dht: Remove linkfile before unlinking the datafile.Vikas Gorur2010-02-021-17/+63
| | | | | | | | | | | | | | | | dht_unlink: If a linkfile exists, remove it first before attempting to remove the datafile. This eliminates the case where dht would end up with the linkfile intact and the datafile gone, and the user not being able to remove the linkfile. Thanks to He Xiaobin <allreol@gmail.com> for debugging this issue. Signed-off-by: Vikas Gorur <vikas@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 188 ([ glusterfs 2.0.6rc2 ] - "Directory not empty" on rm -rf) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=188
* cluster/dht: Check for size 0 in check_is_linkfile()Vijay Bellur2010-01-291-1/+3
| | | | | | | | Signed-off-by: Vijay Bellur <vijay@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 574 (Sticky bit files are not listed in the output of ls on a dht volume) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=574
* dht: initialize layout typeAnand Avati2010-01-233-5/+6
| | | | | | | | | | | Instead of relying on CALLOC's zero initialization which happens to be the value of DHT_HASH_TYPE_DM, explicitly set the value Signed-off-by: Anand V. Avati <avati@amp.gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 523 (Hash type is not initialized in dht_layout_new) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=523
* distribute: Dont dereference sbuf on failed readlink_cbkShehjar Tikoo2009-12-181-0/+4
| | | | | | | | | | | This fixes a crash that was caused due to a NULL pointer deref because a NULL sbuf is returned if the readlink fop fails. Signed-off-by: Shehjar Tikoo <shehjart@gluster.com> Signed-off-by: Vijay Bellur <vijay@dev.gluster.com> BUG: 492 (distribute crashes in readlink) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=492
* cluster/dht: Check for pointers before de-referencing in dht_stat_merge()Vijay Bellur2009-12-161-0/+3
| | | | | | | | Signed-off-by: Vijay Bellur <vijay@gluster.com> Signed-off-by: Vijay Bellur <vijay@dev.gluster.com> BUG: 463 (Crash in dht_stat_merge ()) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=463
* Save st_dev properly in stbuf.st_dev from the same subvolume.Harshavardhana Ranganath2009-12-061-1/+1
| | | | | | | | Signed-off-by: Harshavardhana <harsha@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 433 (Posix conformance test failed on 3.0.0pre2 (Dec 3) release) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=433
* distribute: more cases of st_dev settingAnand Avati2009-12-061-1/+4
| | | | | | | | Signed-off-by: Anand V. Avati <avati@blackhole.gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 433 (Posix conformance test failed on 3.0.0pre2 (Dec 3) release) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=433
* distribute: return st_dev from the same subvol from where itransform is ↵Anand Avati2009-12-062-0/+5
| | | | | | | | | | | | | | performed distribute should return st_dev from the same subvolume it itransforms the inode number. this is to ensure that there will not be false replacements of inodes in inode management Signed-off-by: Anand V. Avati <avati@blackhole.gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 433 (Posix conformance test failed on 3.0.0pre2 (Dec 3) release) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=433
* THIS: set THIS pointers before forget/release/releasedir callbacksAnand Avati2009-12-061-1/+1
| | | | | Signed-off-by: Anand V. Avati <avati@blackhole.gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
* fixes in nufaAmar Tumballi2009-11-263-96/+92
| | | | | | | | | | | | | | | Now nufa uses dht_conf->private variable to store its contents, and doesn't pollute distribute's structures. It used to hang earlier as layout_lock was not INIT()'d. whitespace-cleanup done on the code too. Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 409 (implement a switch scheduler) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=409
* switch translator addedAmar Tumballi2009-11-263-1/+1059
| | | | | | | | | | | switch translator is a wrapper around distribute to work for a pattern based scheduling. Signed-off-by: Amar Tumballi <amar@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 409 (implement a switch scheduler) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=409
* cluster/distribute: copy the stat structure while copying dentries in ↵Raghavendra G2009-11-191-0/+2
| | | | | | | | | | | | | readdirp_cbk. - only the transformed inode number was being copied, leaving other fields of stat 0. Signed-off-by: Raghavendra G <raghavendra@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 371 (rm -rf fails on stat-prefetch.) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=371
* cluster/dht: Do readdir in dht_readdir instead of readdirpVikas Gorur2009-11-131-1/+1
| | | | | | | | | | | | The readdir call will not be used by applications (fuse will only ever send readdirp), but it will be used by afr to do directory self-heal at the time of opendir. Signed-off-by: Vikas Gorur <vikas@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 249 (Self heal of a file that does not exist on the first subvolume) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=249
* Add unhashed-sticky-bit option to volume_options structureSachidananda2009-11-121-0/+3
| | | | | | | | Signed-off-by: Sachidananda <sac@gluster.com> Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 365 (unhashed-sticky-bit is not recognized) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=365
* distribute - more NULL checkAnand V. Avati2009-10-161-3/+21
| | | | | | | Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 315 (generation number support) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=315
* distribute,nufa: layout handling changesAnand V. Avati2009-10-167-136/+209
| | | | | | | | | | changes to make revalidate not fail but instead perform fresh lookup and swap inode context (layout) safely Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 315 (generation number support) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=315
* fix double initialization of dht_local_tAnand V. Avati2009-10-161-7/+0
| | | | | | | | | | This was causing leak of the first initialized frame->local in every create call Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 315 (generation number support) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=315
* NULL checks on local->loc.parentAnand V. Avati2009-10-161-17/+36
| | | | | | | | | | | check for presence of local->loc.parent != NULL before defering to pick ino for post/preparent parameters. local->loc.parent can be NULL legally when loc is describing "/" Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 315 (generation number support) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=315
* shuffle hash layouts on directoriesAnand Avati2009-10-151-2/+44
| | | | | | | | | | | allow for hash layouts to be written differently for different directories to give a better spread for same filenames across directories Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 324 (distribute does not spread files of the same name among all servers) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=324
* cluster/dht: Change STACK_UNWIND to STACK_UNWIND_STRICT.Vikas Gorur2009-10-084-98/+107
| | | | | | | Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 269 (Add a specialized STACK_UNWIND macro for each FOP) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=269
* Changed occurrences of Z Research to Gluster.Vijay Bellur2009-10-0711-11/+11
| | | | Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
* distribute: Support readdirp fopShehjar Tikoo2009-10-062-5/+112
| | | | | | | Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 292 (Separate readdirp functionality from readdir fop) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=292
* distribute: NFS-friendly logic changesShehjar Tikoo2009-10-014-101/+320
| | | | | | | Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 145 (NFSv3 related additions to 2.1 task list) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=145
* Global: NFS-friendly prototype changesShehjar Tikoo2009-10-015-42/+185
| | | | | | | Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 145 (NFSv3 related additions to 2.1 task list) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=145
* Global: Introduce setattr and fsetattr fopsShehjar Tikoo2009-10-014-303/+167
| | | | | | | Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 146 (Add setattr FOP) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=146
* Dumop inodectx addedVijay Bellur2009-09-161-0/+29
| | | | | | | | | | Added dumpop inodectx. Support for dumop inodectx added in dht, locks and client-protocol. Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 213 (Support for process state dump) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=213
* cluster/dht: Support for dumpop priv.Vijay Bellur2009-09-161-0/+134
| | | | | | | Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 213 (Support for process state dump) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=213
* dht_stat_merge - use the highest uid when ambiguousAnand Avati2009-08-041-2/+3
| | | | | | | | | When directories on different subvolumes have different ownerships, use the highest uid/gid till self-heal resolves the inconsistency Signed-off-by: Anand V. Avati <avati@dev.gluster.com> BUG: 191 (random Permission denied errors) URL: http://bugs.gluster.com/cgi-bin/bugzilla3/show_bug.cgi?id=191
* dht_readdir_cbk: use stat attributes in the READDIRPLUS dirents toAnand V. Avati2009-06-261-22/+21
| | | | Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
* rename dht_first_up_child to dht_first_up_subvolAnand V. Avati2009-06-262-2/+3
| | | | Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
* local->loc should be filled in dht-mknod()Amar Tumballi2009-06-181-0/+9
| | | | | | | local->loc should be properly filled while creating a linkfile. otherwise this causes the segfault in underlying volume layer (mostly client-protocol). Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
* making dht and nufa's 'min-free-disk' option to take both percent and disk-sizeAmar Tumballi2009-06-184-38/+90
| | | | | | | | | | | Originally from Paul Rawson <plrca2@gmail.com> http://patches.gluster.com/patch/391/ : patch re-submitted with patching guidelines. cluster/distribute, and cluster/nufa uses new option PERCENT_OR_SIZET for its option 'min-free-disk'. Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
* change the log level of disk full message in dht-diskusageAmar Tumballi2009-06-181-3/+3
| | | | | | | | the current msg is getting printed in warning level. Instead changed it to debug, as the msg specific each node being full is already getting printed in higher priority, hence this msg looks excessive. Signed-off-by: Anand V. Avati <avati@dev.gluster.com>
generated by cgit (git 2.34.1) at 2024-04-19 20:16:45 +0000