Problem: Lookup/creation of thin-arbiter ID file happens in background during mounting. On new volumes, if the ID file creation is in progress, and a FOP fails on data brick, a post-op (xattrop) is attemtped on TA. Since the TA file's gfid is null at this point, the ASSERT checks in protocol/ client causes a crash. Fix: Given that we decided to do Lookup/creation of thin-arbiter in background, fail the other AFR FOPS on TA if the ID file's gfid is null instead of winding it down to protocol/client. Also remove afr_changelog_thin_arbiter_post_op() which seems to be dead code. Updates: #763 Change-Id: I70dc666faf55cc5c8f7cf8e7d36085e4fa399c4d Signed-off-by: Ravishankar N <ravishankar@redhat.com> (cherry picked from commit f9b5074394e3d2f3b6728aab97230ba620879426)
Problem: AFR doesn't delay post-op for fsync fop. For fsync heavy workloads this leads to un-necessary fxattrop/finodelk for every fsync leading to bad performance. Fix: Have delayed post-op for fsync. Add special flag in xdata to indicate that afr shouldn't delay post-op in cases where either the process will terminate or graph-switch would happen. Otherwise it leads to un-necessary heals when the graph-switch/process-termination happens before delayed-post-op completes. Fixes: #1253 Change-Id: I531940d13269a111c49e0510d49514dc169f4577 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
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)
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)
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)
Support for gluster volume heal <volname> info healed/heal-failed was removed by commit bb02cfb56ae08f56df4452c2b948fa962ae1212b in release-3.6. cli parser will display the usage message in all the supported versions whenever these clis are run, leading to some dead code in the latest branches. Since support for these clis were removed long back, this should not give any backward compatibility issues as well. Hence removing the dead code from the code base which will lead to better code coverage by the regression runs as well. Updates: #1052 Change-Id: I0c2b061469caf233c06d9699b0d159ce48e240b9 Signed-off-by: karthik-us <ksubrahm@redhat.com>
...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. Fixes: #1067 Change-Id: If1b050b70b0ad911e162c04db4d89b263e2b8d7b Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Current implementation assumes that ping-event will come after connect event but that may not be the case in the cases where after socket connection fds need to be re-opened which would consume more time. So handle any order of the ping/child-up events. fixes: bz#1800583 Change-Id: I6bcdc0caa503bdc039ef2b4739fbf4afae121f05 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
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. Change-Id: I7ed6108554ca379d532efb1a29b2de8085410b70 Signed-off-by: Xavi Hernandez <xhernandez@redhat.com> Fixes: bz#1808875
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#1801624 Change-Id: I916ae26ad1fabf259bc6362da52d433b7223b17e Signed-off-by: Ravishankar N <ravishankar@redhat.com>