Fixes issue where context cancelling could result in an invalid error…

… being reported (#5329) (#5482)

* Fix issue.

* Update internal/pkg/agent/application/gateway/fleet/fleet_gateway.go

Co-authored-by: Paolo Chilà <paolo.chila@elastic.co>

* Remove check for failed as it no-longer gets marked failed.

---------

Co-authored-by: Paolo Chilà <paolo.chila@elastic.co>
Co-authored-by: Pierre HILBERT <pierre.hilbert@elastic.co>
Co-authored-by: Julien Lind <julien.lind@elastic.co>
Co-authored-by: Shaunak Kashyap <ycombinator@gmail.com>
(cherry picked from commit 96e02e0)

Co-authored-by: Blake Rouse <blake.rouse@elastic.co>

internal/pkg/agent/application/gateway/fleet/fleet_gateway.go

@@ -198,7 +198,13 @@ func (f *FleetGateway) doExecute(ctx context.Context, bo backoff.Backoff) (*flee
 			}
 
 			if !bo.Wait() {
-				// Something bad has happened and we log it and we should update our current state.
+				if ctx.Err() != nil {
+					// if the context is cancelled, break out of the loop
+					break
+				}
+
+				// This should not really happen, but just in-case this error is used to show that
+				// something strange occurred and we want to log it and report it.
 				err := errors.New(
 					"checkin retry loop was stopped",
 					errors.TypeNetwork,

testing/integration/endpoint_security_test.go

@@ -275,11 +275,6 @@ func testInstallAndUnenrollWithEndpointSecurity(t *testing.T, info *define.Info,
 				return false
 			}
 
-			if state.FleetState != client.Failed {
-				t.Logf("Fleet state has not been marked as failed yet!\n%+v", state)
-				return false
-			}
-
 			return true
 		},
 		endpointHealthPollingTimeout,

testing/integration/linux_rpm_test.go

@@ -44,8 +44,6 @@ func TestRpmLogIngestFleetManaged(t *testing.T) {
 		Sudo:  true,
 	})
 
-	t.Skip("Flaky https://github.com/elastic/elastic-agent/issues/5311")
-
 	ctx, cancel := testcontext.WithDeadline(t, context.Background(), time.Now().Add(10*time.Minute))
 	defer cancel()