From 6e15fca1621b06270983f57ac146f0f8e52f0797 Mon Sep 17 00:00:00 2001
From: Mohit Agrawal <[email protected]>
Date: Tue, 9 Jun 2020 15:38:12 +0530
Subject: [PATCH 449/449] test: Test case brick-mux-validation-in-cluster.t is
 failing on RHEL-8

Brick process are not properly attached on any cluster node while
some volume options are changed on peer node and glusterd is down on
that specific node.

Solution: At the time of restart glusterd it got a friend update request
from a peer node if peer node having some changes on volume.If the brick
process is started before received a friend update request in that case
brick_mux behavior is not workingproperly. All bricks are attached to
the same process even volumes options are not the same. To avoid the
issue introduce an atomic flag volpeerupdate and update the value while
glusterd has received a friend update request from peer for a specific
volume.If volpeerupdate flag is 1 volume is started by
glusterd_import_friend_volume synctask

> Change-Id: I4c026f1e7807ded249153670e6967a2be8d22cb7
> Credit: Sanju Rakaonde <[email protected]>
> fixes: #1290
> Signed-off-by: Mohit Agrawal <[email protected]>
> (Reviewed on upstream link https://review.gluster.org/#/c/glusterfs/+/24540/)
> (Cherry pick from commit 955bfd567329cf7fe63e9c3b89d333a55e5e9a20)

BUG: 1844359
Change-Id: I4c026f1e7807ded249153670e6967a2be8d22cb7
Signed-off-by: Mohit Agrawal <[email protected]>
Reviewed-on: https://code.engineering.redhat.com/gerrit/202812
Tested-by: Mohit Agrawal <[email protected]>
Reviewed-by: Sanju Rakonde <[email protected]>
Tested-by: RHGS Build Bot <[email protected]>
Reviewed-by: Sunil Kumar Heggodu Gopala Acharya <[email protected]>
---
 tests/bugs/glusterd/brick-mux-validation-in-cluster.t | 4 +---
 xlators/mgmt/glusterd/src/glusterd-utils.c            | 7 +++++--
 xlators/mgmt/glusterd/src/glusterd.h                  | 4 ++++
 3 files changed, 10 insertions(+), 5 deletions(-)

diff --git a/tests/bugs/glusterd/brick-mux-validation-in-cluster.t b/tests/bugs/glusterd/brick-mux-validation-in-cluster.t
index f088dbb..b6af487 100644
--- a/tests/bugs/glusterd/brick-mux-validation-in-cluster.t
+++ b/tests/bugs/glusterd/brick-mux-validation-in-cluster.t
@@ -100,10 +100,8 @@ $CLI_2 volume set $V0 performance.readdir-ahead on
 $CLI_2 volume set $V1 performance.readdir-ahead on
 
 TEST $glusterd_1;
+EXPECT_WITHIN $PROBE_TIMEOUT 2 peer_count
 
-sleep 10
-
-EXPECT 4 count_brick_processes
 EXPECT_WITHIN $PROCESS_UP_TIMEOUT 4 count_brick_pids
 EXPECT_WITHIN $PROCESS_UP_TIMEOUT 0  count_N/A_brick_pids
 
diff --git a/xlators/mgmt/glusterd/src/glusterd-utils.c b/xlators/mgmt/glusterd/src/glusterd-utils.c
index 2eb2a76..6f904ae 100644
--- a/xlators/mgmt/glusterd/src/glusterd-utils.c
+++ b/xlators/mgmt/glusterd/src/glusterd-utils.c
@@ -3758,6 +3758,7 @@ glusterd_compare_friend_volume(dict_t *peer_data, int32_t count,
                "Version of volume %s differ. local version = %d, "
                "remote version = %d on peer %s",
                volinfo->volname, volinfo->version, version, hostname);
+        GF_ATOMIC_INIT(volinfo->volpeerupdate, 1);
         *status = GLUSTERD_VOL_COMP_UPDATE_REQ;
         goto out;
     } else if (version < volinfo->version) {
@@ -4784,7 +4785,8 @@ glusterd_volinfo_stop_stale_bricks(glusterd_volinfo_t *new_volinfo,
          * or if it's part of the new volume and is pending a snap,
          * then stop the brick process
          */
-        if (ret || (new_brickinfo->snap_status == -1)) {
+        if (ret || (new_brickinfo->snap_status == -1) ||
+            GF_ATOMIC_GET(old_volinfo->volpeerupdate)) {
             /*TODO: may need to switch to 'atomic' flavour of
              * brick_stop, once we make peer rpc program also
              * synctask enabled*/
@@ -6490,7 +6492,8 @@ glusterd_brick_start(glusterd_volinfo_t *volinfo,
      * three different triggers for an attempt to start the brick process
      * due to the quorum handling code in glusterd_friend_sm.
      */
-    if (brickinfo->status == GF_BRICK_STARTING || brickinfo->start_triggered) {
+    if (brickinfo->status == GF_BRICK_STARTING || brickinfo->start_triggered ||
+        GF_ATOMIC_GET(volinfo->volpeerupdate)) {
         gf_msg_debug(this->name, 0,
                      "brick %s is already in starting "
                      "phase",
diff --git a/xlators/mgmt/glusterd/src/glusterd.h b/xlators/mgmt/glusterd/src/glusterd.h
index 1c6c3b1..f739b5d 100644
--- a/xlators/mgmt/glusterd/src/glusterd.h
+++ b/xlators/mgmt/glusterd/src/glusterd.h
@@ -523,6 +523,10 @@ struct glusterd_volinfo_ {
     pthread_mutex_t store_volinfo_lock; /* acquire lock for
                                          * updating the volinfo
                                          */
+    gf_atomic_t volpeerupdate;
+    /* Flag to check about volume has received updates
+       from peer
+    */
 };
 
 typedef enum gd_snap_status_ {
-- 
1.8.3.1