cluster/afr: Fix missing name indices due to EEXIST error

PROBLEM:
Consider a volume with  granular-entry-heal and sharding enabled. When
a replica is down and a shard is created as part of a write, the name
index is correctly created under indices/entry-changes/<dot-shard-gfid>.
Now when a read on the same region triggers another MKNOD, the fop
fails on the online bricks with EEXIST. By virtue of this being a
symmetric error, the failed_subvols[] array is reset to all zeroes.
Because of this, before post-op, the GF_XATTROP_ENTRY_OUT_KEY will be
set, causing the name index, which was created in the previous MKNOD
operation, to be wrongly deleted in THIS MKNOD operation.

FIX:
The ideal fix would have been for a transaction to delete the name
index ONLY if it knows it is the one that created the index in the first
place. This would involve gathering information as to whether THIS xattrop
created the index from individual bricks, aggregating their responses and
based on the various posisble combinations of responses, decide whether to
delete the index or not. This is rather complex. Simpler fix would be
for post-op to examine local->op_ret in the event of no failed_subvols
to figure out whether to delete the name index or not. This can occasionally
lead to creation of stale name indices but they won't be affecting the IO path
or mess with pending changelogs in any way and self-heal in its crawl of
"entry-changes" directory would take care to delete such indices.

Change-Id: Ic1b5257f4dc9c20cb740a866b9598cf785a1affa
BUG: 1408712
Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
Reviewed-on: http://review.gluster.org/16286
Smoke: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>

3 files changed, 98 insertions, 0 deletions

diff --git a/tests/bugs/replicate/bug-1408712.t b/tests/bugs/replicate/bug-1408712.t
new file mode 100644
index 00000000000..b26e8a06923
--- /dev/null
+++ b/tests/bugs/replicate/bug-1408712.t
@@ -0,0 +1,87 @@
+#!/bin/bash
+
+. $(dirname $0)/../../include.rc
+. $(dirname $0)/../../volume.rc
+. $(dirname $0)/../../afr.rc
+
+cleanup
+
+TESTS_EXPECTED_IN_LOOP=12
+
+TEST glusterd
+TEST pidof glusterd
+
+TEST $CLI volume create $V0 replica 3 $H0:$B0/${V0}{0,1,2}
+TEST $CLI volume start $V0
+TEST $CLI volume set $V0 features.shard on
+TEST $CLI volume heal $V0 granular-entry-heal enable
+TEST $CLI volume set $V0 cluster.data-self-heal off
+TEST $CLI volume set $V0 cluster.metadata-self-heal off
+TEST $CLI volume set $V0 cluster.entry-self-heal off
+TEST $CLI volume set $V0 self-heal-daemon off
+TEST $CLI volume set $V0 performance.flush-behind off
+
+TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 $M0
+TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 $M1
+
+cd $M0
+TEST dd if=/dev/zero of=file bs=1M count=8
+
+# Kill brick-0.
+TEST kill_brick $V0 $H0 $B0/${V0}0
+
+TEST "dd if=/dev/zero bs=1M count=8 >> file"
+
+FILE_GFID=$(get_gfid_string $M0/file)
+
+# Test that the index associated with '/.shard' is created on B1 and B2.
+TEST stat $B0/${V0}1/.glusterfs/indices/entry-changes/$DOT_SHARD_GFID
+TEST stat $B0/${V0}2/.glusterfs/indices/entry-changes/$DOT_SHARD_GFID
+# Check for successful creation of granular entry indices
+for i in {2..3}
+do
+        TEST_IN_LOOP stat $B0/${V0}1/.glusterfs/indices/entry-changes/$DOT_SHARD_GFID/$FILE_GFID.$i
+        TEST_IN_LOOP stat $B0/${V0}2/.glusterfs/indices/entry-changes/$DOT_SHARD_GFID/$FILE_GFID.$i
+done
+
+cd ~
+TEST md5sum $M1/file
+
+# Test that the index associated with '/.shard' and the created shards do not disappear on B1 and B2.
+TEST stat $B0/${V0}1/.glusterfs/indices/entry-changes/$DOT_SHARD_GFID
+TEST stat $B0/${V0}2/.glusterfs/indices/entry-changes/$DOT_SHARD_GFID
+for i in {2..3}
+do
+        TEST_IN_LOOP stat $B0/${V0}1/.glusterfs/indices/entry-changes/$DOT_SHARD_GFID/$FILE_GFID.$i
+        TEST_IN_LOOP stat $B0/${V0}2/.glusterfs/indices/entry-changes/$DOT_SHARD_GFID/$FILE_GFID.$i
+done
+
+# Start the brick that was down
+TEST $CLI volume start $V0 force
+EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" afr_child_up_status $V0 0
+EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" afr_child_up_status $V0 1
+EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" afr_child_up_status $V0 2
+
+# Enable shd
+TEST gluster volume set $V0 cluster.self-heal-daemon on
+EXPECT_WITHIN $PROCESS_UP_TIMEOUT "Y" glustershd_up_status
+EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status_in_shd $V0 0
+EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status_in_shd $V0 1
+EXPECT_WITHIN $CHILD_UP_TIMEOUT "1" afr_child_up_status_in_shd $V0 2
+
+TEST $CLI volume heal $V0
+
+# Wait for heal to complete
+EXPECT_WITHIN $HEAL_TIMEOUT "^0$" get_pending_heal_count $V0
+
+# Now verify that there are no name indices left after self-heal
+TEST ! stat $B0/${V0}1/.glusterfs/indices/entry-changes/$DOT_SHARD_GFID
+TEST ! stat $B0/${V0}2/.glusterfs/indices/entry-changes/$DOT_SHARD_GFID
+
+for i in {2..3}
+do
+        TEST_IN_LOOP ! stat $B0/${V0}1/.glusterfs/indices/entry-changes/$DOT_SHARD_GFID/$FILE_GFID.$i
+        TEST_IN_LOOP ! stat $B0/${V0}2/.glusterfs/indices/entry-changes/$DOT_SHARD_GFID/$FILE_GFID.$i
+done
+
+cleanup
diff --git a/tests/include.rc b/tests/include.rc
index 74c279cb8e1..5b5804ea7ab 100644
--- a/tests/include.rc
+++ b/tests/include.rc
@@ -11,6 +11,7 @@ B0=${B0:=/d/backends};        # top level of brick directories
 WORKDIRS="$B0 $M0 $M1 $M2 $N0 $N1"
 
 ROOT_GFID="00000000-0000-0000-0000-000000000001"
+DOT_SHARD_GFID="be318638-e8a0-4c6d-977d-7a937aa84806"
 
 META_VOL=${META_VOL:=gluster_shared_storage}; # shared gluster storage volume used by snapshot scheduler, nfs ganesha and geo-rep.
 META_MNT=${META_MNT:=/var/run/gluster/shared_storage}; # Mount point of shared gluster volume.
diff --git a/xlators/cluster/afr/src/afr-transaction.c b/xlators/cluster/afr/src/afr-transaction.c
index adaad407254..9cfd3bf89ec 100644
--- a/xlators/cluster/afr/src/afr-transaction.c
+++ b/xlators/cluster/afr/src/afr-transaction.c
@@ -1226,6 +1226,16 @@ afr_changelog_populate_xdata (call_frame_t *frame, afr_xattrop_type_t op,
                                 need_entry_key_set = _gf_false;
                                 break;
                         }
+                        /* If the transaction itself did not fail and there
+                         * are no failed subvolumes, check whether the fop
+                         * failed due to a symmetric error. If it did, do
+                         * not set the ENTRY_OUT xattr which would end up
+                         * deleting a name index which was created possibly by
+                         * an earlier entry txn that may have failed on some
+                         * of the sub-volumes.
+                         */
+                        if (local->op_ret)
+                                need_entry_key_set = _gf_false;
                 } else {
                         key = GF_XATTROP_ENTRY_IN_KEY;
                 }