From e81bd15e992362062e2d60b614407667431cf248 Mon Sep 17 00:00:00 2001
From: Madelyn Olson <34459052+madolson@users.noreply.github.com>
Date: Wed, 30 Mar 2022 22:15:00 -0700
Subject: Prevent replica failover during manual takeover test (#10499)

During 11-manual-takeover.tcl, if the killing of the instances happens
too slowly, one of the replicas might be able to promote itself.
I'm not sure why it was slow, but it was observed taking 6 seconds
which is enough time to do an election.
I was able to verify the error locally by adding a small delay (1 second)
during ASAN CI. A fix is just to disable automated failover until all the
nodes are confirmed dead.
---
 tests/cluster/tests/11-manual-takeover.tcl | 24 ++++++++++++++++++------
 1 file changed, 18 insertions(+), 6 deletions(-)

(limited to 'tests/cluster')

diff --git a/tests/cluster/tests/11-manual-takeover.tcl b/tests/cluster/tests/11-manual-takeover.tcl
index f567c6962..78a0f858b 100644
--- a/tests/cluster/tests/11-manual-takeover.tcl
+++ b/tests/cluster/tests/11-manual-takeover.tcl
@@ -14,20 +14,32 @@ test "Cluster is writable" {
     cluster_write_test 0
 }
 
+# For this test, disable replica failover until
+# all of the primaries are confirmed killed. Otherwise
+# there might be enough time to elect a replica.
+set replica_ids { 5 6 7 }
+foreach id $replica_ids {
+    R $id config set cluster-replica-no-failover yes
+}
+
 test "Killing majority of master nodes" {
     kill_instance redis 0
     kill_instance redis 1
     kill_instance redis 2
 }
 
+foreach id $replica_ids {
+    R $id config set cluster-replica-no-failover no
+}
+
 test "Cluster should eventually be down" {
     assert_cluster_state fail
 }
 
 test "Use takeover to bring slaves back" {
-    R 5 cluster failover takeover
-    R 6 cluster failover takeover
-    R 7 cluster failover takeover
+    foreach id $replica_ids {
+        R $id cluster failover takeover
+    }
 }
 
 test "Cluster should eventually be up again" {
@@ -39,9 +51,9 @@ test "Cluster is writable" {
 }
 
 test "Instance #5, #6, #7 are now masters" {
-    assert {[RI 5 role] eq {master}}
-    assert {[RI 6 role] eq {master}}
-    assert {[RI 7 role] eq {master}}
+    foreach id $replica_ids {
+        assert {[RI $id role] eq {master}}
+    }
 }
 
 test "Restarting the previously killed master nodes" {
-- 
cgit v1.2.1