Only execute upgrade tests with Infinispan 14.0.23.Final or higher to

avoid known clustering issues causing flakiness

Known issues:

- https://issues.redhat.com/browse/ISPN-14782
- https://issues.redhat.com/browse/ISPN-14793
- https://issues.redhat.com/browse/ISPN-15151
- https://issues.redhat.com/browse/ISPN-15191
- https://issues.redhat.com/browse/ISPN-15357

Author: ryanemerson

.github/workflows/test_hr_rolling_upgrades.yml

@@ -28,7 +28,7 @@ jobs:
       - name: Run Hot Rod Rolling Upgrade Tests
         run: make hotrod-upgrade-test
         env:
-          SUBSCRIPTION_STARTING_CSV: infinispan-operator.v2.3.2
+          SUBSCRIPTION_STARTING_CSV: infinispan-operator.v2.3.7
           TESTING_OPERAND_IGNORE_LIST: ${{ inputs.skipList }}
           INFINISPAN_CPU: 500m # prevent insufficient cpu error on test-rolling-upgrade pod start
 

.github/workflows/test_upgrades.yml

@@ -125,7 +125,9 @@ jobs:
               exit 1
             fi
           else
-            export SUBSCRIPTION_STARTING_CSV=infinispan-operator.v2.3.2
+            # We must start on a server version greater than 14.0.23.Final to prevent known clustering issues during upgrades
+            # See commit message for a detailed list
+            export SUBSCRIPTION_STARTING_CSV=infinispan-operator.v2.3.7
           fi
           make upgrade-test
         env:

test/e2e/upgrade/upgrade_operands_test.go

@@ -31,8 +31,9 @@ func TestOperandUpgrades(t *testing.T) {
 
 	startingOperandIdx := -1
 	for i, op := range versionManager.Operands {
-		// We must start at Infinispan 14.0.8 or higher due to ISPN-12224
-		if (op.UpstreamVersion.Major == 14 && op.UpstreamVersion.Patch >= 8) || op.UpstreamVersion.Major > 14 {
+		// We must start on a server version greater than 14.0.23.Final to prevent known clustering issues during upgrades
+		// See commit message for a detailed list
+		if op.UpstreamVersion.Major == 14 && op.UpstreamVersion.Patch > 22 || op.UpstreamVersion.Major > 14 {
 			startingOperandIdx = i
 			break
 		}