simplify log outputs and increase timeout for tests

pkg/client/node.go

@@ -427,6 +427,13 @@ func NodeWaitForLogMessage(ctx context.Context, runtime runtimes.Runtime, node *
 	for {
 		select {
 		case <-ctx.Done():
+			if ctx.Err() == context.DeadlineExceeded {
+				d, ok := ctx.Deadline()
+				if ok {
+					log.Debugf("NodeWaitForLogMessage: Context Deadline (%s) > Current Time (%s)", d, time.Now())
+				}
+				return fmt.Errorf("Context deadline exceeded while waiting for log message '%s' of node %s", message, node.Name)
+			}
 			return ctx.Err()
 		default:
 		}

pkg/runtimes/docker/container.go

@@ -161,8 +161,7 @@ func getNodeContainer(ctx context.Context, node *k3d.Node) (*types.Container, er
 		All:     true,
 	})
 	if err != nil {
-		log.Errorln("Failed to list containers")
+		return nil, fmt.Errorf("Failed to list containers: %+v", err)
-		return nil, err
 	}
 
 	if len(containers) > 1 {

pkg/runtimes/docker/node.go

@@ -191,8 +191,7 @@ func getContainersByLabel(ctx context.Context, labels map[string]string) ([]type
 		All:     true,
 	})
 	if err != nil {
-		log.Errorln("Failed to list containers")
+		return nil, fmt.Errorf("Failed to list containers: %+v", err)
-		return nil, err
 	}
 
 	return containers, nil

pkg/runtimes/docker/util.go

@@ -34,8 +34,8 @@ import (
 	"github.com/docker/docker/api/types/filters"
 	"github.com/docker/docker/client"
 	"github.com/docker/docker/pkg/archive"
-	k3d "github.com/rancher/k3d/v4/pkg/types"
 	"github.com/pkg/errors"
+	k3d "github.com/rancher/k3d/v4/pkg/types"
 	log "github.com/sirupsen/logrus"
 )
 

tests/assets/config_test_simple.yaml

@@ -39,7 +39,7 @@ registries:
 options:
   k3d:
     wait: true
-    timeout: "60s"
+    timeout: "360s" # should be pretty high for multi-server clusters to allow for a proper startup routine
     disableLoadbalancer: false
     disableImageVolume: false
   k3s:

tests/test_multi_master_start_stop.sh

@@ -34,7 +34,7 @@ info "Waiting for a bit..."
 sleep 5
 
 info "Restarting cluster (time: $(date -u +"%Y-%m-%d %H:%M:%S %Z"))..."
-$EXE cluster start multiserver --timeout 180s || failed "failed to restart cluster (timeout 180s)"
+$EXE cluster start multiserver --timeout 360s || failed "failed to restart cluster (timeout 360s)"
 
 info "Sleeping for 5 seconds to give the cluster enough time to get ready..."
 sleep 5