fix: [skip e2e] TestNodeDownOnSingleReplica has unstable result (#37288)

issue: #37289
those test case use search to verify replica's status, but if the search
gap is 1s, the node down's effect may be fixed up by balance.

This PR remove the 1 second gap between search operation.

Signed-off-by: Wei Liu <[email protected]>

tests/integration/replicas/balance/replica_test.go

@@ -106,7 +106,7 @@ func (s *ReplicaTestSuit) TestNodeDownOnSingleReplica() {
 			case <-stopSearchCh:
 				log.Info("stop search")
 				return
-			case <-time.After(time.Second):
+			default:
 				expr := fmt.Sprintf("%s > 0", integration.Int64Field)
 				nq := 10
 				topk := 10
@@ -131,9 +131,8 @@ func (s *ReplicaTestSuit) TestNodeDownOnSingleReplica() {
 
 	// stop qn in single replica expected got search failures
 	qn.Stop()
-	s.Eventually(func() bool {
-		return failCounter.Load() > 0
-	}, 30*time.Second, 1*time.Second)
+	time.Sleep(10 * time.Second)
+	s.True(failCounter.Load() > 0)
 
 	close(stopSearchCh)
 }
@@ -176,7 +175,7 @@ func (s *ReplicaTestSuit) TestNodeDownOnMultiReplica() {
 			case <-stopSearchCh:
 				log.Info("stop search")
 				return
-			case <-time.After(time.Second):
+			default:
 				expr := fmt.Sprintf("%s > 0", integration.Int64Field)
 				nq := 10
 				topk := 10