glusterd: replace-brick executing successfully when quorum does not met

Problem:
replace-brick command on a setup where quorum does not met executing
successfully.

Fix:
With the fix glusterd is validating whether server is in quorum or not
during replace-brick staging

>Reviewed-on: https://review.gluster.org/18068
>Smoke: Gluster Build System <jenkins@build.gluster.org>
>CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
>Reviewed-by: Atin Mukherjee <amukherj@redhat.com>

Change-Id: I8017154bb62bdcc6c6490e720ecfe9cde090c161
BUG: 1486110
Signed-off-by: Gaurav Yadav <gyadav@redhat.com>
Reviewed-on: https://review.gluster.org/18125
Smoke: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
Reviewed-by: Prashanth Pai <ppai@redhat.com>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>

2 files changed, 60 insertions, 0 deletions

diff --git a/tests/bugs/glusterd/bug-1483058-replace-brick-quorum-validation.t b/tests/bugs/glusterd/bug-1483058-replace-brick-quorum-validation.t
new file mode 100644
index 00000000000..28b2fbbb9cf
--- /dev/null
+++ b/tests/bugs/glusterd/bug-1483058-replace-brick-quorum-validation.t
@@ -0,0 +1,51 @@
+#!/bin/bash
+
+# Test case for quorum validation in glusterd for syncop framework
+
+. $(dirname $0)/../../include.rc
+. $(dirname $0)/../../volume.rc
+. $(dirname $0)/../../cluster.rc
+
+
+cleanup;
+
+TEST launch_cluster 3
+
+TEST $CLI_1 peer probe $H2;
+EXPECT_WITHIN $PROBE_TIMEOUT 1 peer_count
+
+TEST $CLI_1 peer probe $H3;
+EXPECT_WITHIN $PROBE_TIMEOUT 2 peer_count
+
+# Lets create the volume and set quorum type as a server
+TEST $CLI_1 volume create $V0 $H1:$B1/${V0}0 $H2:$B2/${V0}1 $H3:$B3/${V0}2
+TEST $CLI_1 volume set $V0 cluster.server-quorum-type server
+
+# Start the volume
+TEST $CLI_1 volume start $V0
+
+# Set quorum ratio 95. means 95 % or more than 95% nodes of total available node
+# should be available for performing volume operation.
+# i.e. Server-side quorum is met if the number of nodes that are available is
+# greater than or equal to 'quorum-ratio' times the number of nodes in the
+# cluster
+
+TEST $CLI_1 volume set all cluster.server-quorum-ratio 95
+# Bring down 2nd glusterd
+TEST kill_glusterd 2
+
+EXPECT_WITHIN $PROBE_TIMEOUT 1 peer_count
+
+# Now quorum is not meet. Now execute replace-brick command
+# This command should fail as cluster is not in quorum
+TEST ! $CLI_1 volume replace-brick $V0 $H2:$B2/${V0}1 $H1:$B1/${V0}1_new commit force
+
+# Bring 2nd glusterd up
+TEST start_glusterd 2
+
+EXPECT_WITHIN $PROBE_TIMEOUT 2 peer_count
+
+# Now quorum is met. replace-brick will execute successfuly
+TEST  $CLI_1 volume replace-brick $V0 $H2:$B2/${V0}1 $H1:$B1/${V0}1_new commit force
+
+#cleanup;
diff --git a/xlators/mgmt/glusterd/src/glusterd-replace-brick.c b/xlators/mgmt/glusterd/src/glusterd-replace-brick.c
index 887c3496475..ab38725ffb0 100644
--- a/xlators/mgmt/glusterd/src/glusterd-replace-brick.c
+++ b/xlators/mgmt/glusterd/src/glusterd-replace-brick.c
@@ -21,6 +21,7 @@
 #include "glusterd-nfs-svc.h"
 #include "glusterd-volgen.h"
 #include "glusterd-messages.h"
+#include "glusterd-server-quorum.h"
 #include "glusterd-mgmt.h"
 #include "run.h"
 #include "syscall.h"
@@ -210,6 +211,14 @@ glusterd_op_stage_replace_brick (dict_t *dict, char **op_errstr,
         if (ret)
                 goto out;
 
+        ret = glusterd_validate_quorum (this, gd_op,  dict, op_errstr);
+        if (ret) {
+                gf_msg (this->name, GF_LOG_CRITICAL, 0,
+                        GD_MSG_SERVER_QUORUM_NOT_MET,
+                        "Server quorum not met. Rejecting operation.");
+                goto out;
+        }
+
         if (strcmp (op, "GF_REPLACE_OP_COMMIT_FORCE")) {
                 ret = -1;
                 goto out;