afr: remove stale index entries

Backport of http://review.gluster.org/9714

Problem:
During pre-op phase, the index xlator
1. Creates the entry inside .glusterfs/indices/xattrop
2. Winds the xattrop fop to posix to mark dirty/pending changelogs.
If the brick crashes after 1, the xattrop entry becomes stale and never
gets removed by shd during subsequent crawls because there is nothing to
heal (changelogs are zero).

Though the stale entry does not get displayed in the output of 'heal info'
command, it nevertheless stays there forever unless a new write tansaction
is performed on the file.

Fix:
During index self-heal if afr xattrs are found to be clean (indicated by
ret value of 2 on a call to afr_shd_selfheal(), send a dummy
post-op with all 0s for the xattr values, which makes the index xlator
to unlink the stale entry.

Change-Id: Iffb171e40490abd8d44df09ccc058b5da67baafe
BUG: 1203081
Signed-off-by: Ravishankar N <ravishankar@redhat.com>
Reviewed-on: http://review.gluster.org/9920
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Reviewed-by: Raghavendra Bhat <raghavendra@redhat.com>

1 files changed, 57 insertions, 0 deletions

diff --git a/tests/bugs/bug-1190069-afr-stale-index-entries.t b/tests/bugs/bug-1190069-afr-stale-index-entries.t
new file mode 100644
index 00000000000..8d2871e8270
--- /dev/null
+++ b/tests/bugs/bug-1190069-afr-stale-index-entries.t
@@ -0,0 +1,57 @@
+#!/bin/bash
+. $(dirname $0)/../include.rc
+. $(dirname $0)/../volume.rc
+. $(dirname $0)/../afr.rc
+cleanup;
+
+#Stale entries in xattrop folder for files which do not need heal must be removed during the next index heal crawl.
+
+TEST glusterd;
+TEST pidof glusterd;
+TEST $CLI volume create $V0 replica 2 $H0:$B0/${V0}{0,1};
+TEST $CLI volume set $V0 cluster.self-heal-daemon off
+TEST $CLI volume start $V0;
+TEST glusterfs --volfile-id=/$V0 --volfile-server=$H0 $M0 --attribute-timeout=0 --entry-timeout=0
+TEST `echo hello>$M0/datafile`
+TEST touch $M0/mdatafile
+
+#Create split-brain and reset the afr xattrs, so that we have only the entry inside xattrop folder.
+#This is to simulate the case where the brick crashed just before pre-op happened, but index xlator created the entry inside xattrop folder.
+
+#Create data, metadata SB.
+TEST kill_brick $V0 $H0 $B0/$V0"1"
+TEST stat $M0/datafile
+TEST `echo append>>$M0/datafile`
+TEST chmod +x $M0/mdatafile
+TEST $CLI volume start $V0 force
+EXPECT_WITHIN $CHILD_UP_TIMEOUT '1' afr_child_up_status_meta $M0 $V0-replicate-0 1
+TEST kill_brick $V0 $H0 $B0/$V0"0"
+TEST stat $M0/datafile
+TEST `echo append>>$M0/datafile`
+TEST chmod +x $M0/mdatafile
+TEST $CLI volume start $V0 force
+EXPECT_WITHIN $CHILD_UP_TIMEOUT '1' afr_child_up_status_meta $M0 $V0-replicate-0 0
+TEST ! cat $M0/datafile
+
+TEST $CLI 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
+
+TEST $CLI volume heal $V0
+EXPECT_WITHIN $HEAL_TIMEOUT '2' echo $(count_sh_entries $B0/$V0"0")
+EXPECT_WITHIN $HEAL_TIMEOUT '2' echo $(count_sh_entries $B0/$V0"1")
+
+#Reset xattrs and trigger heal.
+TEST setfattr -n trusted.afr.$V0-client-1 -v 0x000000000000000000000000 $B0/${V0}0/datafile
+TEST setfattr -n trusted.afr.$V0-client-0 -v 0x000000000000000000000000 $B0/${V0}1/datafile
+
+TEST setfattr -n trusted.afr.$V0-client-1 -v 0x000000000000000000000000 $B0/${V0}0/mdatafile
+TEST setfattr -n trusted.afr.$V0-client-0 -v 0x000000000000000000000000 $B0/${V0}1/mdatafile
+
+TEST $CLI volume heal $V0
+EXPECT_WITHIN $HEAL_TIMEOUT "0" afr_get_pending_heal_count $V0
+EXPECT_WITHIN $HEAL_TIMEOUT '0' echo $(count_sh_entries $B0/$V0"0")
+EXPECT_WITHIN $HEAL_TIMEOUT '0' echo $(count_sh_entries $B0/$V0"1")
+
+cleanup