cluster/afr: CLI for granular entry heal enablement/disablement

When there are already existing non-granular indices created that are
yet to be healed, if granular-entry-heal option is toggled from 'off' to
'on', AFR self-heal whenever it kicks in, will try to look for granular
indices in 'entry-changes'. Because of the absence of name indices,
granular entry healing logic will fail to heal these directories, and
worse yet unset pending extended attributes with the assumption that
are no entries that need heal.

To get around this, a new CLI is introduced which will invoke glfsheal
program to figure whether at the time an attempt is made to enable
granular entry heal, there are pending heals on the volume OR there
are one or more bricks that are down. If either of them is true, the
command will be failed with the appropriate error.

New CLI: gluster volume heal <VOL> granular-entry-heal {enable,disable}

Change-Id: I1f4fe8162813b9068e198965d94169fee4adc099
BUG: 1370410
Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
Reviewed-on: http://review.gluster.org/15747
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>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>

6 files changed, 149 insertions, 7 deletions

diff --git a/tests/basic/afr/granular-esh/add-brick.t b/tests/basic/afr/granular-esh/add-brick.t
index f3125d7fe7d..270cf1d32a6 100644
--- a/tests/basic/afr/granular-esh/add-brick.t
+++ b/tests/basic/afr/granular-esh/add-brick.t
@@ -14,7 +14,7 @@ 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 granular-entry-heal on
+TEST $CLI volume heal $V0 granular-entry-heal enable
 
 TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 $M0
 
diff --git a/tests/basic/afr/granular-esh/cli.t b/tests/basic/afr/granular-esh/cli.t
new file mode 100644
index 00000000000..a655180a095
--- /dev/null
+++ b/tests/basic/afr/granular-esh/cli.t
@@ -0,0 +1,142 @@
+#!/bin/bash
+
+. $(dirname $0)/../../../include.rc
+. $(dirname $0)/../../../volume.rc
+. $(dirname $0)/../../../afr.rc
+
+cleanup
+
+TESTS_EXPECTED_IN_LOOP=4
+
+TEST glusterd
+TEST pidof glusterd
+
+TEST   $CLI volume create $V0 replica 2 $H0:$B0/${V0}{0,1}
+# Test that enabling the option should work on a newly created volume
+TEST   $CLI volume set $V0 cluster.granular-entry-heal on
+TEST   $CLI volume set $V0 cluster.granular-entry-heal off
+
+#########################
+##### DISPERSE TEST #####
+#########################
+# Execute the same command on a disperse volume and make sure it fails.
+TEST $CLI volume create $V1 disperse 3 redundancy 1 $H0:$B0/${V1}{0,1,2}
+TEST $CLI volume start $V1
+TEST ! $CLI volume heal $V1 granular-entry-heal enable
+TEST ! $CLI volume heal $V1 granular-entry-heal disable
+
+#######################
+###### TIER TEST ######
+#######################
+# Execute the same command on a disperse + replicate tiered volume and make
+# sure the option is set on the replicate leg of the volume
+TEST $CLI volume attach-tier $V1 replica 2 $H0:$B0/${V1}{3,4}
+TEST $CLI volume heal $V1 granular-entry-heal enable
+EXPECT "enable" volume_get_field $V1 cluster.granular-entry-heal
+TEST $CLI volume heal $V1 granular-entry-heal disable
+EXPECT "disable" volume_get_field $V1 cluster.granular-entry-heal
+
+# Kill a disperse brick and make heal be pending on the volume.
+TEST kill_brick $V1 $H0 $B0/${V1}0
+
+# Now make sure that one offline brick in disperse does not affect enabling the
+# option on the volume.
+TEST $CLI volume heal $V1 granular-entry-heal enable
+EXPECT "enable" volume_get_field $V1 cluster.granular-entry-heal
+TEST $CLI volume heal $V1 granular-entry-heal disable
+EXPECT "disable" volume_get_field $V1 cluster.granular-entry-heal
+
+# Now kill a replicate brick.
+TEST kill_brick $V1 $H0 $B0/${V1}3
+# Now make sure that one offline brick in replicate causes the command to be
+# failed.
+TEST ! $CLI volume heal $V1 granular-entry-heal enable
+EXPECT "disable" volume_get_field $V1 cluster.granular-entry-heal
+
+######################
+### REPLICATE TEST ###
+######################
+TEST   $CLI volume start $V0
+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 that the volume-set way of enabling the option is disallowed
+TEST ! $CLI volume set $V0 granular-entry-heal on
+# Test that the volume-heal way of enabling the option is allowed
+TEST   $CLI volume heal $V0 granular-entry-heal enable
+# Volume-reset of the option should be allowed
+TEST   $CLI volume reset $V0 granular-entry-heal
+TEST   $CLI volume heal $V0 granular-entry-heal enable
+
+EXPECT "enable" volume_option $V0 cluster.granular-entry-heal
+
+TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 $M0
+
+# Kill brick-0.
+TEST kill_brick $V0 $H0 $B0/${V0}0
+
+# Disabling the option should work even when one or more bricks are down
+TEST $CLI volume heal $V0 granular-entry-heal disable
+# When a brick is down, 'enable' attempt should be failed
+TEST ! $CLI volume heal $V0 granular-entry-heal enable
+
+# Restart the killed brick
+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
+
+# When all bricks are up, it should be possible to enable the option
+TEST $CLI volume heal $V0 granular-entry-heal enable
+
+# Kill brick-0 again
+TEST kill_brick $V0 $H0 $B0/${V0}0
+
+# Create files under root
+for i in {1..2}
+do
+        echo $i > $M0/f$i
+done
+
+# Test that the index associated with '/' is created on B1.
+TEST stat $B0/${V0}1/.glusterfs/indices/entry-changes/$ROOT_GFID
+
+# Check for successful creation of granular entry indices
+for i in {1..2}
+do
+        TEST_IN_LOOP stat $B0/${V0}1/.glusterfs/indices/entry-changes/$ROOT_GFID/f$i
+done
+
+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
+
+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
+
+TEST $CLI volume heal $V0
+
+# Wait for heal to complete
+EXPECT_WITHIN $HEAL_TIMEOUT "^0$" get_pending_heal_count $V0
+
+# Test if data was healed
+for i in {1..2}
+do
+        TEST_IN_LOOP diff $B0/${V0}0/f$i $B0/${V0}1/f$i
+done
+
+# Now verify that there are no name indices left after self-heal
+TEST ! stat $B0/${V0}1/.glusterfs/indices/entry-changes/$ROOT_GFID/f1
+TEST ! stat $B0/${V0}1/.glusterfs/indices/entry-changes/$ROOT_GFID/f2
+TEST ! stat $B0/${V0}1/.glusterfs/indices/entry-changes/$ROOT_GFID
+
+# Perform a volume-reset-all-options operation
+TEST $CLI volume reset $V0
+# Ensure that granular entry heal is also disabled
+EXPECT "no" volume_get_field $V0 cluster.granular-entry-heal
+EXPECT "on" volume_get_field $V0 cluster.entry-self-heal
+
+cleanup
+#G_TESTDEF_TEST_STATUS_NETBSD7=BAD_TEST,BUG=1399038
diff --git a/tests/basic/afr/granular-esh/conservative-merge.t b/tests/basic/afr/granular-esh/conservative-merge.t
index b566a0ea4d3..b170e47e0cb 100644
--- a/tests/basic/afr/granular-esh/conservative-merge.t
+++ b/tests/basic/afr/granular-esh/conservative-merge.t
@@ -11,13 +11,13 @@ TESTS_EXPECTED_IN_LOOP=4
 TEST glusterd
 TEST pidof glusterd
 TEST $CLI volume create $V0 replica 2 $H0:$B0/${V0}{0,1}
+TEST $CLI volume start $V0
 TEST $CLI volume set $V0 self-heal-daemon off
 TEST $CLI volume set $V0 data-self-heal off
 TEST $CLI volume set $V0 metadata-self-heal off
 TEST $CLI volume set $V0 entry-self-heal off
-TEST $CLI volume set $V0 granular-entry-heal on
+TEST $CLI volume heal $V0 granular-entry-heal enable
 
-TEST $CLI volume start $V0
 TEST $GFS --volfile-id=$V0 -s $H0 $M0
 
 TEST mkdir $M0/dir
diff --git a/tests/basic/afr/granular-esh/granular-esh.t b/tests/basic/afr/granular-esh/granular-esh.t
index ee53878e004..de0e8f4290b 100644
--- a/tests/basic/afr/granular-esh/granular-esh.t
+++ b/tests/basic/afr/granular-esh/granular-esh.t
@@ -16,7 +16,7 @@ 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 granular-entry-heal on
+TEST $CLI volume heal $V0 granular-entry-heal enable
 
 TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 $M0
 
diff --git a/tests/basic/afr/granular-esh/granular-indices-but-non-granular-heal.t b/tests/basic/afr/granular-esh/granular-indices-but-non-granular-heal.t
index 2da90a98c76..1b5421bf4b6 100644
--- a/tests/basic/afr/granular-esh/granular-indices-but-non-granular-heal.t
+++ b/tests/basic/afr/granular-esh/granular-indices-but-non-granular-heal.t
@@ -17,7 +17,7 @@ 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 granular-entry-heal on
+TEST $CLI volume heal $V0 granular-entry-heal enable
 
 TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 $M0
 
@@ -40,7 +40,7 @@ do
 done
 
 # Now disable granular-entry-heal
-TEST $CLI volume set $V0 granular-entry-heal off
+TEST $CLI volume heal $V0 granular-entry-heal disable
 
 # Start the brick that was down
 TEST $CLI volume start $V0 force
diff --git a/tests/basic/afr/granular-esh/replace-brick.t b/tests/basic/afr/granular-esh/replace-brick.t
index aaa54da2a2c..639ed81b95c 100644
--- a/tests/basic/afr/granular-esh/replace-brick.t
+++ b/tests/basic/afr/granular-esh/replace-brick.t
@@ -12,7 +12,7 @@ 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 granular-entry-heal on
+TEST $CLI volume heal $V0 granular-entry-heal enable
 
 TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 $M0;