fix(e2e): Fix flaky Test_VerifyComponentsAreSuccessfullyStarted_WithRuntimeConfigLoad (#7502)

When a service fails during WaitReady (container starts but crashes due to
runtime config validation), it remains registered in the scenario's services
slice. The next attempt to start a service with the same name then fails with
"another service with the same name has already been started".

Fix by:
1. Calling s.Stop() after expected StartAndWaitReady failures to unregister
   the service before retrying with a new instance.
2. Making ConcreteService.Stop() and Kill() tolerant of already-removed
   containers (started with --rm flag) by treating "No such container"
   errors as successful stops.

Signed-off-by: Ben Ye <benye@amazon.com>

Author: yeya24

integration/e2e/service.go

@@ -156,6 +156,12 @@ func (s *ConcreteService) Stop() error {
 	logger.Log("Stopping", s.name)
 
 	if out, err := RunCommandAndGetOutput("docker", "stop", "--time=30", s.containerName()); err != nil {
+		// If the container has already exited and been removed (e.g., started
+		// with --rm), treat it as a successful stop.
+		if strings.Contains(string(out), "No such container") {
+			s.usedNetworkName = ""
+			return nil
+		}
 		logger.Log(string(out))
 		return err
 	}
@@ -181,6 +187,12 @@ func (s *ConcreteService) Kill() error {
 	logger.Log("Killing", s.name)
 
 	if out, err := RunCommandAndGetOutput("docker", "kill", s.containerName()); err != nil {
+		// If the container has already exited and been removed (e.g., started
+		// with --rm), treat it as a successful kill.
+		if strings.Contains(string(out), "No such container") {
+			s.usedNetworkName = ""
+			return nil
+		}
 		logger.Log(string(out))
 		return err
 	}

integration/runtime_config_test.go

@@ -215,6 +215,10 @@ overrides:
 		"-querier.store-gateway-addresses": strings.Join([]string{storeGateway.NetworkGRPCEndpoint()}, ","),
 	}), "")
 	require.Error(t, s.StartAndWaitReady(querier))
+	// Stop the failed service to unregister it before retrying with the same name.
+	// Ignore error: if the container crashed before Start() completed, the service
+	// was never registered and Stop() returns "does not exist" which is fine.
+	_ = s.Stop(querier)
 
 	// Start Query frontend
 	queryFrontend := e2ecortex.NewQueryFrontendWithConfigFile("query-frontend", "", flags, "")
@@ -231,6 +235,8 @@ overrides:
 	// Ruler start, but fail with "-distributor.shard-by-all-labels": "false"
 	ruler := e2ecortex.NewRuler("ruler", consul.NetworkHTTPEndpoint(), mergeFlags(flags, RulerFlags()), "")
 	require.Error(t, s.StartAndWaitReady(ruler))
+	// Stop the failed service to unregister it before retrying with the same name.
+	_ = s.Stop(ruler)
 
 	// Ruler start, should success with "-distributor.shard-by-all-labels": "true"
 	ruler = e2ecortex.NewRuler("ruler", consul.NetworkHTTPEndpoint(), mergeFlags(flags, RulerFlags(), map[string]string{
@@ -249,6 +255,8 @@ overrides:
 	// Distributor start, but fail with "-distributor.shard-by-all-labels": "false"
 	distributor := e2ecortex.NewQuerier("distributor", e2ecortex.RingStoreConsul, consul.NetworkHTTPEndpoint(), mergeFlags(flags, map[string]string{}), "")
 	require.Error(t, s.StartAndWaitReady(distributor))
+	// Stop the failed service to unregister it before retrying with the same name.
+	_ = s.Stop(distributor)
 
 	// Distributor start, should success with "-distributor.shard-by-all-labels": "true"
 	distributor = e2ecortex.NewQuerier("distributor", e2ecortex.RingStoreConsul, consul.NetworkHTTPEndpoint(), mergeFlags(flags, map[string]string{