summaryrefslogtreecommitdiffstats
path: root/xlators/cluster
Commit message (Collapse)AuthorAgeFilesLines
* afr: Fixed backgroundness detection in self-heal algo.Krishnan Parthasarathi2011-11-231-0/+1
| | | | | | | | | | Change-Id: I9888d8a0b86fdaf6589885766f2de7222d8c8ba2 BUG: 3802 Reviewed-on: http://review.gluster.com/705 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Pranith Kumar Karampuri <pranithk@gluster.com> Reviewed-on: http://review.gluster.com/745 Reviewed-by: Anand Avati <avati@gluster.com>
* cluster/afr: Remove treating sh_frame as special loop_framePranith Kumar K2011-11-234-61/+129
| | | | | | | | Change-Id: I0d87f06f989b2d4b971967c52d4898331693a801 BUG: 3675 Reviewed-on: http://review.gluster.com/735 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/afr: Open fd fix should perform opendir for dirsPranith Kumar K2011-11-231-7/+22
| | | | | | | | Change-Id: Iee12828ca515d44ed71d9cf97dcb8627c85f0593 BUG: 3740 Reviewed-on: http://review.gluster.com/725 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/distribute: Add support for 'min-free-inodes' on each distribute ↵Harshavardhana2011-11-234-200/+257
| | | | | | | | | | | | | | | | | subvolume. This change is required as increasingly large number of small files would cause inodes to run out before they run out on available disk space. It is highly necessary to support algorithmic checking of inodes too just as we do for disk space. Change-Id: I9b87405328d443825e239ee80ab664aceb50ee68 BUG: 3799 Signed-off-by: Harshavardhana <fharshav@redhat.com> Reviewed-on: http://review.gluster.com/730 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Amar Tumballi <amar@gluster.com>
* cluster/afr: Fix memory leaksPranith Kumar K2011-11-232-16/+13
| | | | | | | | Change-Id: I79a1c70c47649fbcf236191f174d766d5806545c BUG: 3805 Reviewed-on: http://review.gluster.com/719 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/dht: set gfid in lookup locPranith Kumar K2011-11-231-0/+2
| | | | | | | | Change-Id: I59599cc88be1d973c955600fdd54e6c49c77b4a2 BUG: 3770 Reviewed-on: http://review.gluster.com/681 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/afr: Handle absence of gfid in lookupPranith Kumar K2011-11-231-6/+16
| | | | | | | | Change-Id: I6295245a7f40ba4f786f1f9f35b337f3f711128d BUG: 3783 Reviewed-on: http://review.gluster.com/739 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* Add quorum checks to avoid split-brain.Jeff Darcy2011-11-206-1/+90
| | | | | | | | Change-Id: I2f123ef93989862aa796903a45682981d5d7fc3c BUG: 3533 Reviewed-on: http://review.gluster.com/473 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/afr: lookup should honor gfid present in locPranith Kumar K2011-11-181-2/+7
| | | | | | | | Change-Id: I2319258743e478cc3a932d8ff0b2204a97cd4f8e BUG: 3760 Reviewed-on: http://review.gluster.com/680 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* dht,afr: Fixed gfid problemsPranith Kumar K2011-11-188-82/+68
| | | | | | | | | | | *) removed uuid_generate usage in pump and afr, self-heald *) filled the gfids for the fops which were sending no gfid in loc Change-Id: I85da3c10f5ee2006248b0123155a60867870d202 BUG: 3760 Reviewed-on: http://review.gluster.com/679 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/dht: Fix a typoHarshavardhana2011-11-161-3/+3
| | | | | | | | | | Change-Id: I6bcdc7d600ebb9ef68c60319f96cd9e28d12c861 BUG: 3809 Signed-off-by: Harshavardhana <fharshav@redhat.com> Reviewed-on: http://review.gluster.com/732 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Amar Tumballi <amar@gluster.com>
* glusterfs: An effort to fix all the spell mistakes and typoHarshavardhana2011-11-1610-18/+18
| | | | | | | | | | | | | | | in the entire glusterfs codebase. This patch fixes many of spell mistakes and typo in the entire glusterfs codebase and all supported modules. Change-Id: I83238a41aa08118df3cf4d1d605505dd3cda35a1 BUG: 3809 Signed-off-by: Harshavardhana <fharshav@redhat.com> Reviewed-on: http://review.gluster.com/731 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amar@gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* core: remove 'ino' variable from 'loc_t' structureAmar Tumballi2011-11-161-3/+2
| | | | | | | | Change-Id: I53b007fbdb42313d207d5d63fbfaaa6aaf033f95 BUG: 3518 Reviewed-on: http://review.gluster.com/523 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* core: remove 'ino' variable from 'inode_t' structureAmar Tumballi2011-11-165-8/+4
| | | | | | | | Change-Id: I0f078d1753db65d2f2e0380d1b0450c114cf40dd BUG: 3518 Reviewed-on: http://review.gluster.com/522 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/distribute: Trigger selfheal only if rmdir succeeded onceshishir gowda2011-11-162-2/+7
| | | | | | | | | | | A EACCES error also should not trigger a selfheal. Only if rmdir succeeded on any subvol, a selfheal should be triggered Change-Id: Ifebb980f0ebd1548adfd3be975ca52ca44c3f3ab BUG: 3786 Reviewed-on: http://review.gluster.com/716 Reviewed-by: Amar Tumballi <amar@gluster.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* Cluster/stripe setattr: Wind only to first_child for special filesshishir gowda2011-11-161-1/+9
| | | | | | | | Change-Id: Ib055578477461a24562d5da139809cf0965b931e BUG: 3807 Reviewed-on: http://review.gluster.com/724 Reviewed-by: Amar Tumballi <amar@gluster.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* cluster/distribute: Use local call_cnt while windingshishir gowda2011-11-163-16/+20
| | | | | | | | | | | | layout->cnt might be modified in cbk's or different threads, which will lead to corruptions Change-Id: Icfdab01ac583cb3d27d62f878b79e0098b597952 BUG: 3730 Reviewed-on: http://review.gluster.com/694 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Jeff Darcy <jdarcy@redhat.com> Reviewed-by: Amar Tumballi <amar@gluster.com>
* cluster/distribute lookup: send revalidate calls to all subvols for directoriesshishir gowda2011-11-161-2/+13
| | | | | | | | | | | | | | If mkdir fails on a subvolume, layout is set taking into account only the subvols where it was successful. stat does not trigger selfheal, as its layout based. Revalidate on directories needs to be sent to all subvols, to fix the error, and not just on the layout. Change-Id: If17055508ffcf268806258ed49e7d7500a89d0f2 BUG: 3793 Reviewed-on: http://review.gluster.com/693 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amar@gluster.com>
* Move some of the messages to log level `TRACE'.Sachidananda Urs2011-11-151-2/+2
| | | | | | | | Change-Id: I46133b5e2218b9d810251b3dadadd8f157ab07d7 BUG: 3761 Reviewed-on: http://review.gluster.com/643 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amar@gluster.com>
* stripe readv_cbk: Fix stat return valuesshishir gowda2011-11-082-1/+7
| | | | | | | | | | | | | | | | | | | | | | | Workaround - If the read request, does not fall to the subvolume with the largest file size set, then we never return the correct size. This leads to clients seeing a truncated file error. The work around is to wipe stat being returned as part of read call. Problem - We were passing the stbuf returned by the first child/index, which can be different to the size/blocks returned by stat. This led to applications viewing the file as being truncated. The stbuf size needs to be the largest of all results, and blocks the aggregation from all subvolumes. (similar to stat) BUG: 3774 Change-Id: I46c53c18b2b42b1f5b86b05555bbab73bf993476 Reviewed-on: http://review.gluster.com/666 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amar@gluster.com>
* cluster/afr : Fix self-heal of special filesKaushal M2011-10-313-12/+44
| | | | | | | | | | | | | | | | Fixes self-heal of special files like device files, fifo files, socket files etc. Does it by doing the following: * Prevent setting of pending data xattr on a special file during entry self-heal when a new fils is created. * Allow data self-heal to be started on all file types other than directories. During data self-heal, for special files just erase pending xattrs, if those xattrs were set by previous releases of glusterfs. Change-Id: I34d8121e23ad00e85371ae2a36ef30cf3bd5db7a BUG: 3525 Reviewed-on: http://review.gluster.com/618 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Pranith Kumar Karampuri <pranithk@gluster.com>
* cluster/afr: Remove unused 'ino' codePranith Kumar K2011-10-285-78/+0
| | | | | | | | Change-Id: I425e2d23e9e45f10ddeff2eacf918dd90f8baee7 BUG: 3744 Reviewed-on: http://review.gluster.com/639 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* pump: Change crawl to accommodate afr_lookup changesPranith Kumar K2011-10-283-51/+66
| | | | | | | | Change-Id: I600120252445c06d9cc3e7aa24022c2559b6abe2 BUG: 3747 Reviewed-on: http://review.gluster.com/638 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/afr: Take gfid-req from xattr_req only if inode->gfid is nullPranith Kumar K2011-10-281-29/+64
| | | | | | | | Change-Id: Iddf5b59d3534c517dcd3c0d7b819e3768f6e915a BUG: 3747 Reviewed-on: http://review.gluster.com/637 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* Stripe mknod: Always call mknod if REGULAR file on first child.shishir gowda2011-10-212-46/+138
| | | | | | | | | | | | This prevents a possible race between mknod(REGULUAR files) and remove. mknod first creates on the first_child, and only if successful, on other subvolumes. Change-Id: I0ddaeb92c1884c771c966bb151db052ab5735367 BUG: 3727 Reviewed-on: http://review.gluster.com/608 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amar@gluster.com>
* build: warning suppression (round n)Amar Tumballi2011-10-205-21/+34
| | | | | | | | | | with this patch, there are no more warnings with gcc (GCC) 4.6.1 20110908 Change-Id: Ice0d52d304b9846395f8a4a191c98eb53125f792 BUG: 2550 Reviewed-on: http://review.gluster.com/607 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@gluster.com>
* distribute: handle migration of symlink and special filesAmar Tumballi2011-10-203-58/+188
| | | | | | | | | | | | TODO: currently, wrt. rebalance/decommissioning, only pending thing is hardlink migration. Change-Id: I30cd06802e84c95601a5a081198f1f09c6d6bc01 BUG: 3714 Reviewed-on: http://review.gluster.com/578 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Shishir Gowda <shishirng@gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* pump: status shouldn't be cleared until migration is complete during commit.Krishnan Parthasarathi2011-10-191-8/+0
| | | | | | | | | | | Clearing pump status on migration complete is futile because we would be 'replacing' src brick with destination brick in the volume anyway. Change-Id: Ib12fee84bd5445c4a20dac1cf10555331d7b8ebd BUG: 3653 Reviewed-on: http://review.gluster.com/585 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/afr: Prevent truncation of mask uint64_t->gf_boolean_tPranith Kumar K2011-10-142-5/+9
| | | | | | | | Change-Id: If67f726f21b713fa9312dc499a1aca4cb00f71de BUG: 3682 Reviewed-on: http://review.gluster.com/589 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/dht: Handle 'user.ufo-test' sent by ufo in dht setxattr.Junaid2011-10-061-2/+53
| | | | | | | | | | In this case, send setxattr to all the bricks to test user xattr support and return error if setxattr fails on even one of the bricks. Change-Id: I9a9b27312a650d25bbc1f57e1ab8899d63d0593d BUG: 3185 Reviewed-on: http://review.gluster.com/542 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* Stripe: Prevent dir gfid mismatch and ENOTEMPTY errorsshishir gowda2011-10-052-103/+334
| | | | | | | | | | | | | | | | | | | | gfid mismatch issue- rmdir- first remove from non first child (FC), and then last from FC. If one fails, FC entry is not removed and op is failed. mkdir- first create dir from FC, followed by other childs. If FC is success then, we succeed, banking on entry self heal ENOTEMPTY issue- unlink: remove from non FC and then on success remove from FC. If non FC remove fails, fail unlink create- Create first on FC, and on success create on non FC. Change-Id: Ia8b9e92cd2aba891d249d1995e383fea0581ffca BUG: 3663 Reviewed-on: http://review.gluster.com/557 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amar@gluster.com>
* cluster/afr: set the read child ctx only if the self-heal source is validRaghavendra Bhat2011-10-031-4/+4
| | | | | | | | Change-Id: Ifdf0db71594ce526ad85c21103726798d9aceef4 BUG: 3639 Reviewed-on: http://review.gluster.com/556 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* core: made changes to return value of __is_root_gfid()Amar Tumballi2011-10-021-1/+1
| | | | | | | | | | | now returns 'true(1)' is gfid is root, 'false(0)' if not. earlier it was the inverse, which was bit confusing Change-Id: Id103f444ace048cbb0fccdc72c6646da06631584 BUG: 3518 Reviewed-on: http://review.gluster.com/549 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* statedump: do not print the inode number in the statedumpRaghavendra Bhat2011-10-013-89/+46
| | | | | | | | | | | | | Since gfid is used to uniquely identify a inode, in the statedump printing inode number is not necessary. Its suffecient if the gfid of the inode is printed. And do not print the the inodelks, entrylks and posixlks if the lock count is 0. Change-Id: Idac115fbce3a5684a0f02f8f5f20b194df8fb27f BUG: 3476 Reviewed-on: http://review.gluster.com/530 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amar@gluster.com>
* distribute rebalance: preserve proper mode in destinationAmar Tumballi2011-09-301-3/+11
| | | | | | | | | | | * don't set 'sticky' and 'sgid' bits to 0 without checking if the source had those bits prior to rebalance. Change-Id: Ia826cb3cfb55312cdbf00d3421f2bd06b3103ce6 BUG: 3656 Reviewed-on: http://review.gluster.com/541 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* rebalance process: propagate proper errors to userAmar Tumballi2011-09-301-8/+45
| | | | | | | | | | | | | | | | * cluster/distribute: while rebalance, differentiate between valid errors, validation check failures (may not be migrate failure), and success. * mgmt/glusterd: differentiate the errors based on errno. If a valid error has happened, mark the rebalance status as failure, and stop the rebalnace crawl. Change-Id: I2d7bd7b73d2b79bfaf892ad4364bc89830a0d5bb BUG: 3656 Reviewed-on: http://review.gluster.com/540 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/afr: Don't unlock sh_frame on failure in algoPranith Kumar K2011-09-302-8/+15
| | | | | | | | Change-Id: I0ef541c1f387c397c345e3f2bc9a57f1eff282a1 BUG: 3647 Reviewed-on: http://review.gluster.com/527 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/afr: set fd ctx on opendirPranith Kumar K2011-09-294-39/+59
| | | | | | | | Change-Id: Ica845035781f47de990e9dcfefdeb37bed99d515 BUG: 3637 Reviewed-on: http://review.gluster.com/536 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* Second round of warning suppression.Jeff Darcy2011-09-2910-43/+10
| | | | | | | | | | | | | | | | | | | | | | Used a #pragma to kill ~170 in rpcgen code. Added GF_UNUSED to deal with a few more from macros elsewhere. The remainder are function return values (mostly context and dict calls) that really should be checked. Those would be harder to fix without real understanding of the code where they occur, so they remain as reminders. (Patchset 2: deal with older gcc that doesn't handle #pragma GCC diagnostic) (Patchset 3: fix include paths in generated files) (Patchset 4: keep up with trunk, squash 9 new warnings) (Patchset 5: six more, all in AFR) Change-Id: I29760c8c81be4d7e6489312c5d0e92cc24814b7b BUG: 2550 Reviewed-on: http://review.gluster.com/378 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/afr: Handle files without gfid in self-healPranith Kumar K2011-09-299-654/+642
| | | | | | | | Change-Id: Ibcaaa9c928195939ff1e31b28b592e524e63a423 BUG: 3557 Reviewed-on: http://review.gluster.com/519 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/distribute: fixed a spurious inode refAmar Tumballi2011-09-271-4/+12
| | | | | | | | | | | | | | While bringing in support to open-fd migration, dht_local_init() itself started doing 'loc_copy()'. I had left one case in dht_lookup() where there is a extra loc_copy() on existing copied 'local->loc', which would cause 2 inode_refs on a given inode, and only one inode_unref() happens in dht_local_wipe(). Change-Id: Idd0375bdf9a6408db1e97e80389249ef8d802adb BUG: 3590 Reviewed-on: http://review.gluster.com/504 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Shishir Gowda <shishirng@gluster.com>
* glusterd: Implemented cmd to trigger self-heal on a replicate volume.Krishnan Parthasarathi2011-09-221-0/+8
| | | | | | | | | | | | | This cmd is used in the context of proactive self-heal for replicated volumes. User invokes the following cmd when (s)he suspects that self-heal needs to be done on a particular volume, gluster volume heal <VOLNAME>. Change-Id: I3954353b53488c28b70406e261808239b44997f3 BUG: 3602 Reviewed-on: http://review.gluster.com/454 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/distribute: validate buf before accessing.Rahul C S2011-09-221-5/+5
| | | | | | | | | | | The macro to check & reset rebalance flags was accessing the iatt structure even in case of failures leading to null dereference. Change-Id: I518f4cc9086cecbe6cf791c8a351287fe3613650 BUG: 3594 Reviewed-on: http://review.gluster.com/472 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Shishir Gowda <shishirng@gluster.com>
* cluster/afr: Make local->child_up immutablePranith Kumar K2011-09-216-164/+113
| | | | | | | | | | | | | | | | | | | | | | | Afr transaction performs lock, pre-op, op, post-op and unlock steps in that order. The child_up[] is overloaded with the information of where all the first two steps succeeded. This works perfectly fine for Transaction, but the locking/unlocking part of the code is re-used by data self-heal. In that each loop_frame does lock, rchecksum, read-from-source and write-to-sinks, unlock steps. Rchecksum fop assumes that the fop needs to happen on one source + all sinks and sets the call_count to that number. But if the lock step fails on any of the sinks it will mark the child_up of that child to 0, which will result in call_count mismatch and the frame will hang thinking that some more cbks need to come. When this happens loop_frame will never go to unlock step leading to hangs on that file. Change-Id: I3dd0449cc6193a980bacf637d935881f4b22210a BUG: 3597 Reviewed-on: http://review.gluster.com/474 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Amar Tumballi <amar@gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/distribute: minor fixes in open file migrationAmar Tumballi2011-09-195-37/+91
| | | | | | | | | | | * incorporated Avati's comments on the first patch. * send proper stat information while unwinding Change-Id: I36982cec610753c241c372272620ab2bd581fd9f BUG: 3071 Reviewed-on: http://review.gluster.com/408 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* Fix typo in log message.Sachidananda Urs2011-09-191-1/+1
| | | | | | | | Change-Id: Ia51ffe03c8b94ddfe21c6609bc0d54b5bd29eca7 BUG: 3158 Reviewed-on: http://review.gluster.com/392 Reviewed-by: Vijay Bellur <vijay@gluster.com> Tested-by: Gluster Build System <jenkins@build.gluster.com>
* Proactive self heal process implementationPranith Kumar K2011-09-1410-121/+631
| | | | | | | | Change-Id: I96db0d94566ceabf1649f890318363f738c06553 BUG: 2458 Reviewed-on: http://review.gluster.com/403 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* support for de-commissioning a node using 'remove-brick'Amar Tumballi2011-09-135-6/+141
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | to achieve this, we now create volume-file with 'decommissioned-nodes' option in distribute volume, then just perform the rebalance set of operations (with 'force' flag set). now onwards, the 'remove-brick' (with 'start' option) operation tries to migrate data from removed bricks to existing bricks. 'remove-brick' also supports similar options as of replace-brick. * (no options) -> works as 'force', will have the current behavior of remove-brick, ie., no data-migration, volume changes. * start (starts remove-brick with data-migration/draining process, which takes care of migrating data and once complete, will commit the changes to volume file) * pause (stop data migration, but keep the volume file intact with extra options whatever is set) * abort (stop data-migration, and fall back to old configuration) * commit (if volume is stopped, commits the changes to volumefile) * force (stops the data-migration and commits the changes to volume file) Change-Id: I3952bcfbe604a0952e68b6accace7014d5e401d3 BUG: 1952 Reviewed-on: http://review.gluster.com/118 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* distribute rebalance: handle the open file migrationAmar Tumballi2011-09-1213-1651/+2898
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Complexity involved: To migrate a file with open fd, we have to notify the other client process which has the open fd, and make sure the write()s happening on that fd is properly synced to the migrated file. Once the migration is complete, the client process which has open-fd should get notified and it should start performing all the operations on the new subvolume, instead of earlier cached volume. How to solve the notification part: We can overload the 'postbuf' attribute in the _cbk() function to understand if a file is 'under-migration' or 'migration-complete' state. (This will be something similar to deciding whether a file is DHT-linkfile by its 'mode'). Overall change includes below mentioned major changes: 1. dht_linkfile is decided by only 2 factors (mode(01000), xattr(trusted.glusterfs.dht.linkto)), instead of earlier 3 factors (size==0) 2. in linkfile self-heal part (in 'dht_lookup_everywhere_cbk()'), don't delete a linkfile if there is a open-fd on it. It means, there may be a migration in progress. 3. if a file's revalidate fails with ENOENT, it may be due to file migration, and hence need a lookup_everywhere() 4. There will be 2 phases of file-migration. -> Phase 1: Migration in progress * The source data file will have SGID and STICKY bit set in its mode. * The source data file will have a 'linkto' xattr pointing the destination. * Destination file will have mode set to '01000', and 'linkto' xattr set to itself. -> Phase 2: File migration Complete * The source data file will have mode '01000', and will be 'truncated' to size 0. * The destination file will have inherited mode from the source. (without sgid and sticky bit) and its 'linkto' attribute will be removed. 4. Changes in distribute to work smoothly with a file which is in migration / got migrated. The 'fops' are divided into 3 categories, inode-read, inode-write and others. inode-read fops need to handle only 'phase 2' notification, where as, the inode-write fops need to handle both 'phase 1' and phase2. The inode-write operations will be done on source file, and if any of 'file-migration' procedures are detected in _cbk(), then the operations should be performed on the destination too. when a phase-2 is detected, then the inode-ctx itself should be changed to represent a new layout. With these changes, the open file migration will work smoothly with multiple clients. Change-Id: I512408463814e650f34c62ed009bf2101d016fd6 BUG: 3071 Reviewed-on: http://review.gluster.com/209 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Vijay Bellur <vijay@gluster.com>
* cluster/afr: perform self-heal with least priorityPranith Kumar K2011-09-091-0/+7
| | | | | | | | Change-Id: Id8a1dffa3c3200234ad154d1749278a2d7c7021b BUG: 3502 Reviewed-on: http://review.gluster.com/336 Tested-by: Gluster Build System <jenkins@build.gluster.com> Reviewed-by: Anand Avati <avati@gluster.com>
generated by cgit (git 2.34.1) at 2024-06-03 05:22:05 +0000