At the time of pre-op, pre_op_xdata is populted with the xattrs we get from the disk and at the time of post-op it gets over-written without unreffing the previous value stored leading to a leak. This is a regression we missed in https://review.gluster.org/#/q/ba149bac92d169ae2256dbc75202dc9e5d06538e Originally: > Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> > (cherry picked from commit e7b79c59590c203c65f7ac8548b30d068c232d33) Change-Id: I0456f9ad6f77ce6248b747964a037193af3a3da7 Fixes: bz#1613512 Signed-off-by: Amar Tumballi <amarts@redhat.com>
Backport of: https://review.gluster.org/#/c/20029/ 3.12 still supports quorum-reads, hence modified afr_inode_refresh_done() to support that. If inode refresh failed on all children of afr due to ENOENT (say file migrated by dht), it resets the readables to zero. Any inflight txn which then later comes on the inode fails with EIO because no readable children present for the inode. Fix: Don't update readables when inode refresh fails on *all* children of afr. In that way any inflight txns will either proceed with its own inode refresh if needed and fail it with the right errno or use the old value of readables and continue with the txn. Also, add quorum checks to the beginning of afr_transaction(). Otherwise, we seem to be winding the lock and checking for quorum only in pre-op pahse. Note: This should ideally fix BZ 1329505 since the stop gap fix for it is has been reverted at https://review.gluster.org/#/c/20028. Change-Id: I82990769f01be918a073fec83fc67ba4b3be24b1 BUG: 1599247 Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Backport of https://review.gluster.org/#/c/20271/ (only change is in .t) commit 20fa80057eb430fd72b4fa31b9b65598b8ec1265 introduced a regression wherein if a file is present in only 1 brick of replica *and* doesn't have a gfid associated with it, it doesn't get healed upon the next lookup from the client. Fix it. Change-Id: I7d1111dcb45b1b8b8340a7d02558f05df70aa599 BUG: 1598121 fixes: bz#1598121 Signed-off-by: Ravishankar N <ravishankar@redhat.com> (cherry picked from commit eb472d82a083883335bc494b87ea175ac43471ff)
Backport of https://review.gluster.org/#/c/20036/ Note: We need to update inode context's write_subvol even in case of compound fops. This is not there in master and 4.1 since compound FOPS was removed in it. Problem: In the .t, when the only good brick was brought down, writes on the fd were still succeeding on the bad bricks. The inflight split-brain check was marking the write as failure but since the write succeeded on all the bad bricks, afr_txn_nothing_failed() was set to true and we were unwinding writev with success to DHT and then catching the failure in post-op in the background. Fix: Don't wind the FOP phase if the write_subvol (which is populated with readable subvols obtained in pre-op cbk) does not have at least 1 good brick which was up when the transaction started. Change-Id: I4a1fef4569609c31cffeaef591a64c10870e8d0b BUG: 1598720 Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Backport of https://review.gluster.org/#/c/19781/ Problem: We seem to be winding the FOP if pre-op did not succeed on quorum bricks and then failing the FOP with EROFS since the fop did not meet quorum. This essentially masks the actual error due to which pre-op failed. (See BZ). Fix: Skip FOP phase if pre-op quorum is not met and go to post-op. Change-Id: Ie58a41e8fa1ad79aa06093706e96db8eef61b6d9 BUG: 1597154 Signed-off-by: Ravishankar N <ravishankar@redhat.com>
If the post-op phase of txn did not meet quorm checks, use that errno to unwind the FOP rather than blindly setting ENOTCONN. Change-Id: I0cb0c8771ec75a45f9a25ad4cd8601103deddf0c BUG: 1597120 Signed-off-by: Ravishankar N <ravishankar@redhat.com> (cherry picked from commit 440a048f24b006c80af3d7bcd0a1f13fe3459d87)
afr relies on pending changelog xattrs to identify source and sinks and the setting of these xattrs happen in post-op. So if post-op fails, we need to unwind the write txn with a failure. Change-Id: I0f019ac03890108324ee7672883d774918b20be1 BUG: 1597120 Signed-off-by: Ravishankar N <ravishankar@redhat.com> (cherry picked from commit a40a87ec3b226ae86a6ed8f4af25b45965a20cad)
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 split-brain purely from the afr xattrs point of view i.e each brick is blamed by atleast one of the others. These scenarios are hit when there is frequent connect/disconnect of the client/shd to the bricks while I/O or heal are in progress. Fix: Instead of undoing the post-op, 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. Change-Id: I3d0224b883eb0945785ade0e9697a1c828aec0ae BUG: 1597123 Signed-off-by: Ravishankar N <ravishankar@redhat.com> (cherry picked from commit 0e6e8216823c2d9dafb81aae0f6ee3497c23d140)
Backport of https://review.gluster.org/19045 Problem: Setting the write_subvol value to read_subvol in case of metadata transaction during pre-op (commit 19f9bcff4aada589d4321356c2670ed283f02c03) might lead to the original problem of arbiter becoming source. Scenario: 1) All bricks are up and good 2) 2 writes w1 and w2 are in progress in parallel 3) ctx->read_subvol is good for all the subvolumes 4) w1 succeeds on brick0 and fails on brick1, yet to do post-op on the disk 5) read/lookup comes on the same file and refreshes read_subvols back to all good 6) metadata transaction happens which makes ctx->write_subvol to be assigned with ctx->read_subvol which is all good 7) w2 succeeds on brick1 and fails on brick0 and this will update the brick in reverse order leading to arbiter becoming source Fix: Instead of setting the ctx->write_subvol to ctx->read_subvol in the pre-op statge, if there is a metadata transaction, check in the function __afr_set_in_flight_sb_status() if it is a data/metadata transaction. Use the value of ctx->write_subvol if it is a data transactions and ctx->read_subvol value for other transactions. With this patch we assign the value of ctx->write_subvol in the afr_transaction_perform_fop() with the on disk value, instead of assigning it in the afr_changelog_pre_op() with the in memory value. Change-Id: Id2025a7e965f0578af35b1abaac793b019c43cc4 BUG: 1566131 Signed-off-by: karthik-us <ksubrahm@redhat.com> Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Backport of https://review.gluster.org/#/c/18049/ Problem: When eager-lock is on, and two writes happen in parallel on a FD we were observing the following behaviour: - First write fails on one data brick - Since the post-op is not yet happened, the inode refresh will get both the data bricks as readable and set it in the inode context - In flight split brain check see both the data bricks as readable and allows the second write - Second write fails on the other data brick - Now the post-op happens and marks both the data bricks as bad and arbiter will become source for healing Fix: Adding one more variable called write_suvol in inode context and it will have the in memory representation of the writable subvols. Inode refresh will not update this value and its lifetime is pre-op through unlock in the afr transaction. Initially the pre-op will set this value same as read_subvol in inode context and then in the in flight split brain check we will use this value instead of read_subvol. After all the checks we will update the value of this and set the read_subvol same as this to avoid having incorrect value in that. Change-Id: I2ef6904524ab91af861d59690974bbc529ab1af3 BUG: 1566131 Signed-off-by: karthik-us <ksubrahm@redhat.com>