summaryrefslogtreecommitdiffstats
path: root/xlators/cluster
Commit message (Collapse)AuthorAgeFilesLines
* afr: event gen changesRavishankar N2020-07-083-78/+25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The general idea of the changes is to prevent resetting event generation to zero in the inode ctx, since event gen is something that should follow 'causal order'. Change #1: For a read txn, in inode refresh cbk, if event_generation is found zero, we are failing the read fop. This is not needed because change in event gen is only a marker for the next inode refresh to happen and should not be taken into account by the current read txn. Change #2: The event gen being zero above can happen if there is a racing lookup, which resets even get (in afr_lookup_done) if there are non zero afr xattrs. The resetting is done only to trigger an inode refresh and a possible client side heal on the next lookup. That can be acheived by setting the need_refresh flag in the inode ctx. So replaced all occurences of resetting even gen to zero with a call to afr_inode_need_refresh_set(). Change #3: In both lookup and discover path, we are doing an inode refresh which is not required since all 3 essentially do the same thing- update the inode ctx with the good/bad copies from the brick replies. Inode refresh also triggers background heals, but I think it is okay to do it when we call refresh during the read and write txns and not in the lookup path. The .ts which relied on inode refresh in lookup path to trigger heals are now changed to do read txn so that inode refresh and the heal happens. Change-Id: Iebf39a9be6ffd7ffd6e4046c96b0fa78ade6c5ec Fixes: #1179 Signed-off-by: Ravishankar N <ravishankar@redhat.com> Reported-by: Erik Jacobson <erik.jacobson at hpe.com> (cherry picked from commit f0fcd909ad4535b60c9208d4804ebe6afe421a09)
* cluster/afr: Prioritize ENOSPC over other errorskarthik-us2020-07-082-47/+5
| | | | | | | | | | | | | | | | | | | | | | | Problem: In a replicate/arbiter volume if file creations or writes fails on quorum number of bricks and on one brick it is due to ENOSPC and on other brick it fails for a different reason, it may fail with errors other than ENOSPC in some cases. Fix: Prioritize ENOSPC over other lesser priority errors and do not set op_errno in posix_gfid_set if op_ret is 0 to avoid receiving any error_no which can be misinterpreted by __afr_dir_write_finalize(). Also removing the function afr_has_arbiter_fop_cbk_quorum() which might consider a successful reply form a single brick as quorum success in some cases, whereas we always need fop to be successful on quorum number of bricks in arbiter configuration. Change-Id: I106e267f8b9451f681022f1cccb410d9bc824c08 Fixes: #1254 Signed-off-by: karthik-us <ksubrahm@redhat.com> (cherry picked from commit fa63b45ca5edf172b1b89b28b5db3c5129cc57b6)
* afr: more quorum checks in lookup and new entry markingRavishankar N2020-07-083-11/+25
| | | | | | | | | | | | | | | | | Problem: See github issue for details. Fix: -In lookup if the entry exists in 2 out of 3 bricks, don't fail the lookup with ENOENT just because there is an entrylk on the parent. Consider quorum before deciding. -If entry FOP does not succeed on quorum no. of bricks, do not perform new entry mark. Fixes: #1303 Change-Id: I56df8c89ad53b29fa450c7930a7b7ccec9f4a6c5 Signed-off-by: Ravishankar N <ravishankar@redhat.com> (cherry picked from commit c4a6748f25d2c1ab3ebcf89952278ebf94c8d371)
* cluster/ec: Return correct error code and log messageAshish Pandey2020-07-081-2/+9
| | | | | | | | | | | | | In case of readdir was send with an FD on which opendir was failed, this FD will be useless and we return it with error. For now, we are returning it with EINVAL without logging any message in log file. Return a correct error code and also log the message to improve thing to debug. fixes: #1220 Change-Id: Iaf035254b9c5aa52fa43ace72d328be622b06169 (cherry picked from commit af70cb5eedd80207cd184e69f2a4fb252b72d070)
* afr: mark pending xattrs as a part of metadata healRavishankar N2020-04-201-1/+61
| | | | | | | | | | | | | | | | | | | | ...if pending xattrs are zero for all children. Problem: If there are no pending xattrs and a metadata heal needs to be performed, it can be possible that we end up with xattrs inadvertendly deleted from all bricks, as explained in the BZ. Fix: After picking one among the sources as the good copy, mark pending xattrs on all sources to blame the sinks. Now even if this metadata heal fails midway, a subsequent heal will still choose one of the valid sources that it picked previously. Updates: #1067 Change-Id: If1b050b70b0ad911e162c04db4d89b263e2b8d7b Signed-off-by: Ravishankar N <ravishankar@redhat.com> (cherry picked from commit 2d5ba449e9200b16184b1e7fc84cabd015f1f779)
* cluster/afr: fix race when bricks come upXavi Hernandez2020-04-203-6/+9
| | | | | | | | | | | | | | | | | | | | | | | | | The was a problem when self-heal was sending lookups at the same time that one of the bricks was coming up. In this case there was a chance that the number of 'up' bricks changes in the middle of sending the requests to subvolumes which caused a discrepancy in the expected number of replies and the actual number of sent requests. This discrepancy caused that AFR continued executing requests before all requests were complete. Eventually, the frame of the pending request was destroyed when the operation terminated, causing a use- after-free issue when the answer was finally received. In theory the same thing could happen in the reverse way, i.e. AFR tries to wait for more replies than sent requests, causing a hang. Backport of: > Change-Id: I7ed6108554ca379d532efb1a29b2de8085410b70 > Signed-off-by: Xavi Hernandez <xhernandez@redhat.com> > Fixes: bz#1808875 Change-Id: I7ed6108554ca379d532efb1a29b2de8085410b70 Signed-off-by: Xavi Hernandez <xhernandez@redhat.com> Fixes: bz#1809439
* cluster/ec: Change handling of heal failure to avoid crashAshish Pandey2020-02-282-13/+13
| | | | | | | | | | | | | | | | | Problem: ec_getxattr_heal_cbk was called with NULL as second argument in case heal was failing. This function was dereferencing "cookie" argument which caused crash. Solution: Cookie is changed to carry the value that was supposed to be stored in fop->data, so even in the case when fop is NULL in error case, there won't be any NULL dereference. Thanks to Xavi for the suggestion about the fix. Change-Id: I0798000d5cadb17c3c2fbfa1baf77033ffc2bb8c fixes: bz#1806836
* afr: prevent spurious entry heals leading to gfid split-brainRavishankar N2020-02-285-15/+7
| | | | | | | | | | | | | | | | | | | | | Problem: In a hyperconverged setup with granular-entry-heal enabled, if a file is recreated while one of the bricks is down, and an index heal is triggered (with the brick still down), entry-self heal was doing a spurious heal with just the 2 good bricks. It was doing a post-op leading to removal of the filename from .glusterfs/indices/entry-changes as well as erroneous setting of afr xattrs on the parent. When the brick came up, the xattrs were cleared, resulting in the renamed file not getting healed and leading to gfid split-brain and EIO on the mount. Fix: Proceed with entry heal only when shd can connect to all bricks of the replica, just like in data and metadata heal. fixes: bz#1804594 Change-Id: I916ae26ad1fabf259bc6362da52d433b7223b17e Signed-off-by: Ravishankar N <ravishankar@redhat.com> (cherry picked from commit 06453d77d056fbaa393a137ca277a20e38d2f67e)
* cluster/thin-arbiter: Wait for TA connection before ta-file lookupAshish Pandey2020-02-261-19/+21
| | | | | | | | | | | | | | | | | | | Problem: When we mount a ta volume, as soon as 2 data bricks are connected we consider that the mount is done and then send a lookup/create on ta file on ta node. However, this connection with ta node might not have been completed. Due to this delay, ta replica id file will not be created and we will see ENOTCONN error in log file if we do lookup. Solution: As we know that this ta node could have a higher latency, we should wait for reasonable time for connection to happen before sending lookup/create on replica id file. fixes: bz#1804546 Change-Id: I36f90865afe617e4e84cee57fec832a16f5dd6cc (cherry picked from commit a7fa54ddea3fe429f143b37e4de06a93b49d776a)
* cluster/ec: skip updating ctx->loc again when ec_fix_open/opendirKinglong Mee2020-02-262-10/+14
| | | | | | | | | | | | | The ec_manager_open/opendir memsets ctx->loc which causes memory/inode leak, and ec_fheal uses ctx->loc out of fd->lock that loc_copy may copy bad data when memset it. This patch skips updating ctx->loc when it is initilizaed. With it, ctx->loc is filled once, and never updated. Change-Id: I3bf5ffce4caf4c1c667f7acaa14b451d37a3550a fixes: bz#1806838 Signed-off-by: Kinglong Mee <mijinlong@horiscale.com>
* Cluster/afr: Don't treat all bricks having metadata pending as split-brainkarthik-us2020-02-252-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | Problem: We currently don't have a roll-back/undoing of post-ops if quorum is not met. Though the FOP is still unwound with failure, the xattrs remain on the disk. Due to these partial post-ops and partial heals (healing only when 2 bricks are up), we can end up in metadata split-brain purely from the afr xattrs point of view i.e each brick is blamed by atleast one of the others for metadata. These scenarios are hit when there is frequent connect/disconnect of the client/shd to the bricks. Fix: Pick a source based on the xattr values. If 2 bricks blame one, the blamed one must be treated as sink. If there is no majority, all are sources. Once we pick a source, self-heal will then do the heal instead of erroring out due to split-brain. This patch also adds restriction of all the bricks to be up to perform metadata heal to avoid any metadata loss. Removed the test case tests/bugs/replicate/bug-1468279-source-not-blaming-sinks.t as it was doing metadata heal even when only 2 of 3 bricks were up. Change-Id: I07a9d62f84ceda329dcab1f02a33aeed258dcb09 fixes: bz#1805097 Signed-off-by: karthik-us <ksubrahm@redhat.com>
* cluster/dht: Correct fd processing loopN Balachandran2019-12-301-22/+62
| | | | | | | | | | | | | | | | | The fd processing loops in the dht_migration_complete_check_task and the dht_rebalance_inprogress_task functions were unsafe and could cause an open to be sent on an already freed fd. This has been fixed. > Change-Id: I0a3c7d2fba314089e03dfd704f9dceb134749540 > Fixes: bz#1757399 > Signed-off-by: N Balachandran <nbalacha@redhat.com> > (cherry picked from commit 9b15867070b0cc241ab165886292ecffc3bc0aed) Change-Id: I0a3c7d2fba314089e03dfd704f9dceb134749540 Fixes: bz#1786983 Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
* cluster/ec: Update lock->good_mask on parent fop failurePranith Kumar K2019-10-302-0/+6
| | | | | | | | | | When discard/truncate performs write fop, it should do so after updating lock->good_mask to make sure readv happens on the correct mask fixes: bz#1739449 Change-Id: Idfef0bbcca8860d53707094722e6ba3f81c583b7 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
* cluster/ec: Fix reopen flags to avoid misbehaviorPranith Kumar K2019-10-302-3/+8
| | | | | | | | | | | | | | | | | | | | | | | Problem: when a file needs to be re-opened O_APPEND and O_EXCL flags are not filtered in EC. - O_APPEND should be filtered because EC doesn't send O_APPEND below EC for open to make sure writes happen on the individual fragments instead of at the end of the file. - O_EXCL should be filtered because shd could have created the file so even when file exists open should succeed - O_CREAT should be filtered because open happens with gfid as parameter. So open fop will create just the gfid which will lead to problems. Fix: Filter out these two flags in reopen. Change-Id: Ia280470fcb5188a09caa07bf665a2a94bce23bc4 Fixes: bz#1739450 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
* cluster/ec: Always read from good-maskPranith Kumar K2019-10-302-5/+25
| | | | | | | | | | There are cases where fop->mask may have fop->healing added and readv shouldn't be wound on fop->healing. To avoid this always wind readv to lock->good_mask updates: bz#1739449 Change-Id: I2226ef0229daf5ff315d51e868b980ee48060b87 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
* cluster/ec: inherit healing from lock when it has infoKinglong Mee2019-10-301-2/+3
| | | | | | | | | If lock has info, fop should inherit healing mask from it. Otherwise, fop cannot inherit right healing when changed_flags is zero. Change-Id: Ife80c9169d2c555024347a20300b0583f7e8a87f updates: bz#1739449 Signed-off-by: Kinglong Mee <mijinlong@horiscale.com>
* cluster/ec: Prevent double pre-op xattropsPranith Kumar K2019-10-301-6/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: Race: Thread-1 Thread-2 1) Does ec_get_size_version() to perform pre-op fxattrop as part of write-1 2) Calls ec_set_dirty_flag() in ec_get_size_version() for write-2. This sets dirty[] to 1 3) Completes executing ec_prepare_update_cbk leading to ctx->dirty[] = '1' 4) Takes LOCK(inode->lock) to check if there are any flags and sets dirty-flag because lock->waiting_flag is 0 now. This leads to fxattrop to increment on-disk dirty[] to '2' At the end of the writes the file will be marked for heal even when it doesn't need heal. Fix: Perform ec_set_dirty_flag() and other checks inside LOCK() to prevent dirty[] to be marked as '1' in step 2) above Updates bz#1739446 Change-Id: Icac2ab39c0b1e7e154387800fbededc561612865 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
* cluster/ec: fix EIO error for concurrent writes on sparse filesXavi Hernandez2019-10-241-9/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | EC doesn't allow concurrent writes on overlapping areas, they are serialized. However non-overlapping writes are serviced in parallel. When a write is not aligned, EC first needs to read the entire chunk from disk, apply the modified fragment and write it again. The problem appears on sparse files because a write to an offset implicitly creates data on offsets below it (so, in some way, they are overlapping). For example, if a file is empty and we read 10 bytes from offset 10, read() will return 0 bytes. Now, if we write one byte at offset 1M and retry the same read, the system call will return 10 bytes (all containing 0's). So if we have two writes, the first one at offset 10 and the second one at offset 1M, EC will send both in parallel because they do not overlap. However, the first one will try to read missing data from the first chunk (i.e. offsets 0 to 9) to recombine the entire chunk and do the final write. This read will happen in parallel with the write to 1M. What could happen is that half of the bricks process the write before the read, and the half do the read before the write. Some bricks will return 10 bytes of data while the otherw will return 0 bytes (because the file on the brick has not been expanded yet). When EC tries to recombine the answers from the bricks, it can't, because it needs more than half consistent answers to recover the data. So this read fails with EIO error. This error is propagated to the parent write, which is aborted and EIO is returned to the application. The issue happened because EC assumed that a write to a given offset implies that offsets below it exist. This fix prevents the read of the chunk from bricks if the current size of the file is smaller than the read chunk offset. This size is correctly tracked, so this fixes the issue. Also modifying ec-stripe.t file for Test #13 within it. In this patch, if a file size is less than the offset we are writing, we fill zeros in head and tail and do not consider it strip cache miss. That actually make sense as we know what data that part holds and there is no need of reading it from bricks. Backport of: > Patch:https://review.gluster.org/#/c/glusterfs/+/23066/ > Change-Id: Ic342e8c35c555b8534109e9314c9a0710b6225d6 > BUG: 1730715 > Signed-off-by: Xavi Hernandez <xhernandez@redhat.com> (cherry picked from commit b01a43586c5abc23a874e5528a063c508f952cbd) Change-Id: Ic342e8c35c555b8534109e9314c9a0710b6225d6 Fixes: bz#1739451 Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
* dht: Rebalance causing IO Error - File descriptor in bad stateMohit Agrawal2019-10-175-17/+116
| | | | | | | | | | | | | | | | | | | | | Problem : When a file is migrated, dht attempts to re-open all open fds on the new cached subvol. Earlier, if dht had not opened the fd, the client xlator would be unable to find the remote fd and would fall back to using an anon fd for the fop. That behavior changed with https://review.gluster.org/#/c/glusterfs/+/15804, causing fops to fail with EBADFD if the fd was not available on the cached subvol. The client xlator returns EBADFD if the remote fd is not found but dht only checks for EBADF before re-opening fds on the new cached subvol. Solution: Handle EBADFD at dht code path to avoid the issue >Change-Id: I43c51995cdd48d05b12e4b2889c8dbe2bb2a72d8 >Fixes: bz#1758579 >(cherry picked from commit 9314a9fbf487614c736cf6c4c1b93078d37bb9df) >(Reviewed on upstream https://review.gluster.org/#/c/glusterfs/+/23518/) Change-Id: I43c51995cdd48d05b12e4b2889c8dbe2bb2a72d8 Fixes: bz#1761907
* cluster/afr: Heal entries when there is a source & no healed_sinkskarthik-us2019-10-171-0/+15
| | | | | | | | | | | | | | | | | | | | Problem: In a situation where B1 blames B2, B2 blames B1 and B3 doesn't blame anything for entry heal, heal will not complete even though we have clear source and sinks. This will happen because while doing afr_selfheal_find_direction() only the bricks which are blamed by non-accused bricks are considered as sinks. Later in __afr_selfheal_entry_finalize_source() when it tries to mark all the non-sources as sinks it fails to do so because there won't be any healed_sinks marked, no witness present and there will be a source. Fix: If there is a source and no healed_sinks, then reset all the locked sources to 0 and healed sinks to 1 to do conservative merge. Change-Id: If40d8bc95d52a52b2730f55bdcf135109b421548 Fixes: bz#1760706 Signed-off-by: karthik-us <ksubrahm@redhat.com>
* afr: support split-brain CLI for replica 3Ravishankar N2019-10-171-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | Ever since we added quorum checks for lookups in afr via commit bd44d59741bb8c0f5d7a62c5b1094179dd0ce8a4, the split-brain resolution commands would not work for replica 3 because there would be no readables for the lookup fop. The argument was that split-brains do not occur in replica 3 but we do see (data/metadata) split-brain cases once in a while which indicate that there are a few bugs/corner cases yet to be discovered and fixed. Fortunately, commit 8016d51a3bbd410b0b927ed66be50a09574b7982 added GF_CLIENT_PID_GLFS_HEALD as the pid for all fops made by glfsheal. If we leverage this and allow lookups in afr when pid is GF_CLIENT_PID_GLFS_HEALD, split-brain resolution commands will work for replica 3 volumes too. Likewise, the check is added in shard_lookup as well to permit resolving split-brains by specifying "/.shard/shard-file.xx" as the file name (which previously used to fail with EPERM). Change-Id: I3c543dea79caf7cfbc1633e9089cb1cdd2538ba9 Fixes: bz#1760792 Signed-off-by: Ravishankar N <ravishankar@redhat.com> (cherry picked from commit 47dbd753187f69b3835d2e42fdbe7485874c4b3e)
* ctime/rebalance: Heal ctime xattr on directory during rebalanceKotresh HR2019-09-273-4/+7
| | | | | | | | | | | | | | | | | | | | | | | After add-brick and rebalance, the ctime xattr is not present on rebalanced directories on new brick. This patch fixes the same. Note that ctime still doesn't support consistent time across distribute sub-volume. This patch also fixes the in-memory inconsistency of time attributes when metadata is self healed. Backport of: > Patch: https://review.gluster.org/23127 > Change-Id: Ia20506f1839021bf61d4753191e7dc34b31bb2df > BUG: 1734026 > Signed-off-by: Kotresh HR <khiremat@redhat.com> Patch: https://review.gluster.org/23127 Change-Id: Ia20506f1839021bf61d4753191e7dc34b31bb2df fixes: bz#1752413 Signed-off-by: Kotresh HR <khiremat@redhat.com>
* dht: Custom xattrs are not healed in case of add-brickroot2019-09-271-8/+1
| | | | | | | | | | | | | | | | | | | | | | Problem: If any custom xattrs are set on the directory before add a brick, xattrs are not healed on the directory after adding a brick. Solution: xattr are not healed because dht_selfheal_dir_mkdir_lookup_cbk checks the value of MDS and if MDS value is not negative selfheal code path does not take reference of MDS xattrs.Change the condition to take reference of MDS xattr so that custom xattrs are populated on newly added brick Backport of: > Patch: https://review.gluster.org/22520 > BUG: bz#1702299 > Change-Id: Id14beedb98cce6928055f294e1594b22132e811c > Signed-off-by: Mohit Agrawal <moagrawal@redhat.com> fixes: bz#1753561 Change-Id: Id14beedb98cce6928055f294e1594b22132e811c Signed-off-by: Kotresh HR <khiremat@redhat.com>
* afr/lookup: Pass xattr_req in while doing a selfheal in lookupMohammed Rafi KC2019-09-233-5/+16
| | | | | | | | | | | | | | | | | We were not passing xattr_req when doing a name self heal as well as a meta data heal. Because of this, some xdata was missing which causes i/o errors Backport of > https://review.gluster.org/#/c/glusterfs/+/23024/ >Change-Id: Ibfb1205a7eb0195632dc3820116ffbbb8043545f >Fixes: bz#1728770 >Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com> Fixes: bz#1749307 Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com> (cherry picked from commit d026f0bcfd301712e4f0671ccf238f43f2e6dd30) Change-Id: Ibfb1205a7eb0195632dc3820116ffbbb8043545f
* afr: wake up index healer threadsRavishankar N2019-09-055-11/+24
| | | | | | | | | | | | | | | (Backport of https://review.gluster.org/#/c/glusterfs/+/23288/) ...whenever shd is re-enabled after disabling or there is a change in `cluster.heal-timeout`, without needing to restart shd or waiting for the current `cluster.heal-timeout` seconds to expire. See BZ 1743988 for more details. Change-Id: Ia5ebd7c8e9f5b54cba3199c141fdd1af2f9b9bfe fixes: bz#1743988 Reported-by: Glen Kiessling <glenk1973@hotmail.com> Signed-off-by: Ravishankar N <ravishankar@redhat.com>
* afr: restore timestamp of parent dir during entry-healRavishankar N2019-08-261-0/+2
| | | | | | | Fixes: bz#1741044 Change-Id: I29e338bac62104233a6f80212df8d0fb016affda Signed-off-by: Ravishankar N <ravishankar@redhat.com> (cherry picked from commit 8e9c53ebf16705b9a1db2fc486dc24a5cb244ddd)
* cluster/dht: Fix directory perms during selfhealN Balachandran2019-07-291-3/+5
| | | | | | | | | | | | | | Fixed a bug in the revalidate code path that wiped out directory permissions if no mds subvol was found. Backport of: > Change-Id: I8b4239ffee7001493c59d4032a2d3062586ea115 > fixes: bz#1716830 > Signed-off-by: N Balachandran <nbalacha@redhat.com> Change-Id: I8b4239ffee7001493c59d4032a2d3062586ea115 fixes: bz#1716848 Signed-off-by: N Balachandran <nbalacha@redhat.com>
* [RFC] change get_real_filename implementation to use ENOATTR instead of ENOENTMichael Adam2019-07-041-4/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | get_real_filename is implemented as a virtual extended attribute to help Samba implement the case-insensitive but case preserving SMB protocol more efficiently. It is implemented as a getxattr call on the parent directory with the virtual key of "get_real_filename:<entryname>" by looking for a spelling with different case for the provided file/dir name (<entryname>) and returning this correct spelling as a result if the entry is found. Originally (05aaec645a6262d431486eb5ac7cd702646cfcfb), the implementation used the ENOENT errno to return the authoritative answer that <entryname> does not exist in any case folding. Now this implementation is actually a violation or misuse of the defined API for the getxattr call which returns ENOENT for the case that the dir that the call is made against does not exist and ENOATTR (or the synonym ENODATA) for the case that the xattr does not exist. This was not a problem until the gluster fuse-bridge was changed to do map ENOENT to ESTALE in 59629f1da9dca670d5dcc6425f7f89b3e96b46bf, after which we the getxattr call for get_real_filename returned an ESTALE instead of ENOENT breaking the expectation in Samba. It is an independent problem that ESTALE should not leak out to user space but is intended to trigger retries between fuse and gluster. But nevertheless, the semantics seem to be incorrect here and should be changed. This patch changes the implementation of the get_real_filename virtual xattr to correctly return ENOATTR instead of ENOENT if the file/directory being looked up is not found. The Samba glusterfs_fuse vfs module which takes advantage of the get_real_filename over a fuse mount will receive a corresponding change to map ENOATTR to ENOENT. Without this change, it will still work correctly, but the performance optimization for nonexisting files is lost. On the other hand side, this change removes the distinction between the old not-implemented case and the implemented case. So Samba changed to treat ENOATTR like ENOENT will not work correctly any more against old servers that don't implement get_real_filename. I.e. existing files will be reported as non-existing Backport of: > Change-Id: I971b427ab8410636d5d201157d9af70e0d075b67 > fixes: bz#1722977 > Signed-off-by: Michael Adam <obnox@samba.org> Change-Id: I971b427ab8410636d5d201157d9af70e0d075b67 fixes: bz#1723659 Signed-off-by: Michael Adam <obnox@samba.org> (cherry picked from commit dc1b87fcfef08c9497b0c02b2410c9d18bbc2dba)
* cluster/dht: Fixed a memleak in dht_rename_cbkN Balachandran2019-07-031-11/+33
| | | | | | | | | | | | | Fixed a memleak in dht_rename_cbk when creating a linkto file. >Change-Id: I705adef3cb79e33806520fc2b15558e90e2c211c >fixes: bz#1722698 Change-Id: I705adef3cb79e33806520fc2b15558e90e2c211c fixes: bz#1726294 Signed-off-by: N Balachandran <nbalacha@redhat.com> (cherry picked from commit 532b0fc8b1ace9ad48fdaf643e0b1a34020b6cd8)
* cluster/ec: honor contention notifications for partially acquired locksXavi Hernandez2019-06-031-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | EC was ignoring lock contention notifications received while a lock was being acquired. When a lock is partially acquired (some bricks have granted the lock but some others not yet) we can receive notifications from acquired bricks, which should be honored, since we may not receive more notifications after that. Since EC was ignoring them, once the lock was acquired, it was not released until the eager-lock timeout, causing unnecessary delays on other clients. This fix takes into consideration the notifications received before having completed the full lock acquisition. After that, the lock will be releaed as soon as possible. Backport of: > BUG: bz#1708156 > Change-Id: I2a306dbdb29fb557dcab7788a258bd75d826cc12 > Signed-off-by: Xavi Hernandez <xhernandez@redhat.com> Fixes: bz#1714172 Change-Id: I2a306dbdb29fb557dcab7788a258bd75d826cc12 Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
* cluster/ec: Reopen shouldn't happen with O_TRUNCPranith Kumar K2019-05-151-1/+1
| | | | | | | | | | | | | Problem: Doing re-open with O_TRUNC will truncate the fragment even when it is not needed needing extra heals Fix: At the time of re-open don't use O_TRUNC. fixes bz#1709660 Change-Id: Idc6408968efaad897b95a5a52481c66e843d3fb8 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
* afr: thin-arbiter lock release fixesRavishankar N2019-05-153-47/+93
| | | | | | | | | | | | | | | | | | | | | - pass fop state instead of afr local to afr_ta_dom_lock_check_and_release() - avoid afr_lock_release_synctask() being called simultaneosuly from notify code path and transaction (post-op) code path due to races. - Check if the post-op on TA is valid based on event_gen checks. - Invalidate in-memory information when we get TA child down. Note: Thi patch addresses some pending review comments of commit 053b1309dc8fbc05fcde5223e734da9f694cf5cc (https://review.gluster.org/#/c/glusterfs/+/20095/) fixes: bz#1709130 Change-Id: I2ccd7e1b53362f9f3fed8680aecb23b5011eb18c Signed-off-by: Ravishankar N <ravishankar@redhat.com> (cherry picked from commit 9ab2747da78061882f6734df4b265bce11adaef1)
* cluster/afr : TA: Return actual error code in case of failureAshish Pandey2019-05-131-6/+6
| | | | | | | | | | In afr_ta_post_op_do, we were sending EIO for every failure. However, the original error code should be sent. Change-Id: I9fdc15dac00d758baf8e6f14db244f526481a63a updates: bz#1709143 Signed-off-by: Ashish Pandey <aspandey@redhat.com> (cherry picked from commit 63159cdb5374f458d7d2bffec24d4720ffc96d6c)
* cluster/dht: Refactor dht lookup functionsN Balachandran2019-05-091-74/+30
| | | | | | | | | | Part 2: Modify dht_revalidate_cbk to call dht_selfheal_directory instead of separate calls to heal attrs and xattrs. Change-Id: Id41ac6c4220c2c35484812bbfc6157fc3c86b142 fixes: bz#1707393 Signed-off-by: N Balachandran <nbalacha@redhat.com>
* cluster/dht: refactor dht lookup functionsN Balachandran2019-05-082-124/+119
| | | | | | | | | | Part 1: refactor the dht_lookup_dir_cbk and dht_selfheal_directory functions. Added a simple dht selfheal directory test Change-Id: I1410c26359e3c14b396adbe751937a52bd2fcff9 updates: bz#1707393 Signed-off-by: N Balachandran <nbalacha@redhat.com>
* cluster/ec: fix fd reopenPranith Kumar K2019-05-0814-274/+328
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently EC tries to reopen fd's that have been opened while a brick was down. This is done as part of regular write operations, just after having acquired the locks, and it's sent as a sub-fop of the main write fop. There were two problems: 1. The reopen was attempted on all UP bricks, even if a previous lock didn't succeed. This is incorrect because most probably the open will fail. 2. If reopen is sent and fails, the error is propagated to the main operation, causing it to fail when it shouldn't. To fix this, we only attempt reopens on bricks where the current fop owns a lock, and we prevent any error to be propagated to the main fop. To implement this behaviour an argument used to indicate the minimum number of required answers has overloaded to also include some flags. To make the change consistent, it has been necessary to rename the argument, which means that a lot of files have been changed. However there are no functional changes. This change has also uncovered a problem in discard code, which didn't correctely process requests of small sizes because no real discard fop was being processed, only a write of 0's on some region. In this case some fields of the fop remained uninitialized or with incorrect values. To fix this, a new function has been created to simulate success on a fop and it's used in the discard case. Thanks to Pranith for providing a test script that has also detected an issue in this patch. This patch includes a small modification of this script to force data to be written into bricks before stopping them. Backport of: > Change-Id: If272343873369186c2fb8f43c1d9c52c3ea304ec > BUG: bz#1699866 > Signed-off-by: Xavi Hernandez <xhernandez@redhat.com> Change-Id: If272343873369186c2fb8f43c1d9c52c3ea304ec Fixes: bz#1699917 Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
* cluster/afr: Remove local from owners_list on failure of lock-acquisitionPranith Kumar K2019-04-164-18/+14
| | | | | | | | | | | | | When eager-lock lock acquisition fails because of say network failures, the local is not being removed from owners_list, this leads to accumulation of waiting frames and the application will hang because the waiting frames are under the assumption that another transaction is in the process of acquiring lock because owner-list is not empty. Handled this case as well in this patch. Added asserts to make it easier to find these problems in future. fixes bz#1699731 Change-Id: I3101393265e9827755725b1f2d94a93d8709e923 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
* ec: fix truncate lock to cover the write in tuncate cleanKinglong Mee2019-04-161-2/+6
| | | | | | | | | | | ec_truncate_clean does writing under the lock granted for truncate, but the lock is calculated by ec_adjust_offset_up, so that, the write in ec_truncate_clean is out of lock. Updates: bz#1699499 Change-Id: Idbe1fd48d26afe49c36b77db9f12e0907f5a4134 Signed-off-by: Kinglong Mee <kinglongmee@gmail.com> (cherry picked from commit 0e1223491e964096384edfae5032ed0d50d028ad)
* cluster/afr: Thin-arbiter SHD fixeskarthik-us2019-04-162-13/+13
| | | | | | | | | This patch address post-merge review comments for commit 5784a00f997212d34bd52b2303e20c097240d91c Change-Id: I7ed954664a2ae8e1091d23ee3ceb9c66e83bfeac fixes: bz#1699319 Signed-off-by: karthik-us <ksubrahm@redhat.com>
* afr: thin-arbiter read txn fixesRavishankar N2019-04-163-22/+37
| | | | | | | | | | | | | | - Fixes afr_ta_read_txn() to handle inode refresh failures. code-path. - Fixes a double free issue of dict. Note: This patch address post-merge review comments for commit 69532c141be160b3fea03c1579ae4ac13018dcdf fixes: bz#1693992 Change-Id: Id5299b45b68569d47df6b73755918237a1592cb4 Signed-off-by: Ravishankar N <ravishankar@redhat.com> (cherry picked from commit 500bd0014128e6727e83b6cb77e8ac94304b8f4a)
* cluster/ec: Don't enqueue an entry if it is already healingAshish Pandey2019-04-165-30/+127
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: 1 - heal-wait-qlength is by default 128. If shd is disabled and we need to heal files, client side heal is needed. If we access these files that will trigger the heal. However, it has been observed that a file will be enqueued multiple times in the heal wait queue, which in turn causes queue to be filled and prevent other files to be enqueued. 2 - While a file is going through healing and a write fop from mount comes on that file, it sends write on all the bricks including healing one. At the end it updates version and size on all the bricks. However, it does not unset dirty flag on all the bricks, even if this write fop was successful on all the bricks. After healing completion this dirty flag remain set and never gets cleaned up if SHD is disabled. Solution: 1 - If an entry is already in queue or going through heal process, don't enqueue next client side request to heal the same file. 2 - Unset dirty on all the bricks at the end if fop has succeeded on all the bricks even if some of the bricks are going through heal. Change-Id: Ia61ffe230c6502ce6cb934425d55e2f40dd1a727 updates: bz#1693223 Signed-off-by: Ashish Pandey <aspandey@redhat.com> (cherry picked from commit 313dcefe7a62bd16cd794040df068f9bec9c6927)
* afr: add client-pid to all gf_event() callsRavishankar N2019-04-166-15/+28
| | | | | | | | | | client-pid for glustershd is GF_CLIENT_PID_SELF_HEALD client-pid for glfsheal is GF_CLIENT_PID_GLFS_HEALD updates: bz#1693155 Change-Id: Ib3a863af160ff48c822a5e6b0c27c575c9887470 Signed-off-by: Ravishankar N <ravishankar@redhat.com> (cherry picked from commit 8016d51a3bbd410b0b927ed66be50a09574b7982)
* cluster/ec: Fix handling of heal info cases without locksAshish Pandey2019-04-091-25/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When we use heal info command, it takes lot of time as in some cases it takes lock on entries to find out if the entry actually needs heal or not. There are some cases where we can avoid these locks and can conclude if the entry needs heal or not. 1 - We do a lookup (without lock) on an entry, which we found in .glusterfs/indices/xattrop, and find that lock count is zero. Now if the file contains dirty bit set on all or any brick, we can say that this entry needs heal. 2 - If the lock count is one and dirty is greater than 1, then it also means that some fop had left the dirty bit set which made the dirty count of current fop (which has taken lock) more than one. At this point also we can definitely say that this entry needs heal. This patch is modifying code to take into consideration above two points. It is also changing code to not to call ec_heal_inspect if ec_heal_do was called from client side heal. Client side heal triggeres heal only when it is sure that it requires heal. [We have changed the code to not to call heal for lookup] updates bz#1697764 Change-Id: I7f09f0ecd12f65a353297aefd57026fd2bebdf9c Signed-off-by: Ashish Pandey <aspandey@redhat.com> (cherry picked from commit da47caf2405c08c9abafc4a55525a8b2c2dd5bb8)
* cluster/afr: Send truncate on arbiter brick from SHDkarthik-us2019-03-121-15/+13
| | | | | | | | | | | | | | | | | | | Problem: In an arbiter volume configuration SHD will not send any writes onto the arbiter brick even if there is data pending marker for the arbiter brick. If we have a arbiter setup on the geo-rep master and there are data pending markers for the files on arbiter brick, SHD will not mark any data changelog during healing. While syncing the data from master to slave, if the arbiter-brick is considered as ACTIVE, then there is a chance that slave will miss out some data. If the arbiter brick is being newly added or replaced there is a chance of slave missing all the data during sync. Fix: If there is data pending marker for the arbiter brick, send truncate on the arbiter brick during heal, so that it will record truncate as the data transaction in changelog. Change-Id: I3242ba6cea6da495c418ef860d9c3359c5459dec fixes: bz#1687672 Signed-off-by: karthik-us <ksubrahm@redhat.com>
* cluster/dht: Request linkto xattrs in dht_rmdir opendirN Balachandran2019-02-221-1/+26
| | | | | | | | | | | | | | | | | | If parallel-readdir is enabled, the rda xlator is loaded below dht in the graph and proactively lists and caches entries when an opendir is performed. dht_rmdir checks if the directory being deleted contains stale linkto files by performing a readdirp on its child subvols. However, as the entries are actually read in during the opendir operation which does not request the linkto xattr,no linkto xattrs are present for the entries causing dht to incorrectly identify them as data files and fail the rmdir operation with ENOTEMPTY. DHT now always adds the linkto xattr in the list of xattrs requested in the opendir. Change-Id: I0711198e66c59146282eb8b88084170bedfb4018 fixes: bz#1679004 Signed-off-by: N Balachandran <nbalacha@redhat.com>
* dht: fix double extra unref of inode at heal pathKinglong Mee2019-02-221-1/+1
| | | | | | | | | | | | | The loc_wipe is done in the _out_ section, inode_unref(loc.parent) here casues a double extra unref of loc.parent. > Change-Id: I2dc809328d3d34bf7b02c7df9a4f97788af511e6 > Signed-off-by: Kinglong Mee <mijinlong@open-fs.com> (cherry-pick of https://review.gluster.org/#/c/glusterfs/+/21998/) Change-Id: I2dc809328d3d34bf7b02c7df9a4f97788af511e6 updates: bz#1679275 Signed-off-by: Susant Palai <spalai@redhat.com>
* cluster/dht: Fix lookup selfheal and rmdir raceN Balachandran2019-02-181-9/+25
| | | | | | | | | | | | | | A race between the lookup selfheal and rmdir can cause directories to be healed only on non-hashed subvols. This can prevent the directory from being listed from the mount point and in turn causes rm -rf to fail with ENOTEMPTY. Fix: Update the layout information correctly and reduce the call count only after processing the response. Change-Id: I812779aaf3d7bcf24aab1cb158cb6ed50d212451 fixes: bz#1677260 Signed-off-by: N Balachandran <nbalacha@redhat.com>
* cluster/dht: Do not use gfid-req in fresh lookupN Balachandran2019-02-022-8/+60
| | | | | | | | | | | | Fuse sets a random gfid-req value for a fresh lookup. Posix lookup will set this gfid on entries with missing gfids causing a GFID mismatch for directories. DHT will now ignore the Fuse provided gfid-req and use the GFID returned from other subvols to heal the missing gfid. Change-Id: I5f541978808f246ba4542564251e341ec490db14 fixes: bz#1670259 Signed-off-by: N Balachandran <nbalacha@redhat.com>
* core: make gf_thread_create() easier to useXavi Hernandez2019-02-011-6/+1
| | | | | | | | | | | | | | This patch creates a specific function to set the thread name using a string format and a variable argument list, like printf(). This function is used to set the thread name from gf_thread_create(), which now accepts a variable argument list to create the full name. It's not necessary anymore to use a local array to build the name of the thread. This is done automatically. Change-Id: Idd8d01fd462c227359b96e98699f8c6d962dc17c Updates: bz#1193929 Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
* cluster/thin-arbiter: Consider thin-arbiter before marking new entry changelogAshish Pandey2019-02-014-19/+87
| | | | | | | | | | | | | | | | If a fop to create an entry fails on one of the data brick, we mark the pending changelog on the entry on brick for which it was successful. This is done as part of post op phase to make sure that entry gets healed even if it gets renamed to some other path where its parent was not marked as bad. As it happens as part of post op, we should consider thin-arbiter to check if the brick, which was successful, is the good brick or not. This will avoide split brain and other issues. Change-Id: I12686675be98f02f70a5186b3ed748c541514d53 updates: bz#1662264 Signed-off-by: Ashish Pandey <aspandey@redhat.com>
generated by cgit (git 2.34.1) at 2024-05-09 14:23:01 +0000