diff --git a/deploy/deploy.go b/deploy/deploy.go
index a8ce993d..b84e9959 100644
--- a/deploy/deploy.go
+++ b/deploy/deploy.go
@@ -220,25 +220,25 @@ func (d *Deployment) Deploy(ctx context.Context, kubecfg string) (rerr error) {
 		defer func() { finish(rerr) }()
 	}
 	if err := d.checkDependencies(); err != nil {
-		return err
+		return fmt.Errorf("failed to check for dependencies: %w", err)
 	}
 	log.Infof("Deploying cluster...")
 	if err := d.Cluster.Deploy(ctx); err != nil {
-		return err
+		return fmt.Errorf("failed to deploy cluster: %w", err)
 	}
 	log.Infof("Cluster deployed")
 	if err := d.Cluster.Healthy(); err != nil {
-		return err
+		return fmt.Errorf("failed to check cluster is healthy: %w", err)
 	}
 	log.Infof("Cluster healthy")
 	// Once cluster is up, set kClient
 	rCfg, err := clientcmd.BuildConfigFromFlags("", kubecfg)
 	if err != nil {
-		return err
+		return fmt.Errorf("failed to create k8s config: %w", err)
 	}
 	kClient, err := kubernetes.NewForConfig(rCfg)
 	if err != nil {
-		return err
+		return fmt.Errorf("failed to create k8s client: %w", err)
 	}
 
 	log.Infof("Checking kubectl versions.")
@@ -252,11 +252,11 @@ func (d *Deployment) Deploy(ctx context.Context, kubecfg string) (rerr error) {
 	}
 	kClientVersion, err := getVersion(kubeYAML.ClientVersion.GitVersion)
 	if err != nil {
-		return err
+		return fmt.Errorf("failed to parse k8s client version: %w", err)
 	}
 	kServerVersion, err := getVersion(kubeYAML.ServerVersion.GitVersion)
 	if err != nil {
-		return err
+		return fmt.Errorf("failed to parse k8s server version: %w", err)
 	}
 	origMajor := kClientVersion.Major
 	kClientVersion.Major -= 2
@@ -298,35 +298,35 @@ func (d *Deployment) Deploy(ctx context.Context, kubecfg string) (rerr error) {
 
 	log.Infof("Deploying ingress...")
 	if err := d.Ingress.Deploy(ctx); err != nil {
-		return err
+		return fmt.Errorf("failed to deploy ingress: %w", err)
 	}
 	tCtx, cancel := context.WithTimeout(ctx, healthTimeout)
 	defer cancel()
 	if err := d.Ingress.Healthy(tCtx); err != nil {
-		return err
+		return fmt.Errorf("failed to check ingress is healthy: %w", err)
 	}
 	log.Infof("Ingress healthy")
 	log.Infof("Deploying CNI...")
 	if err := d.CNI.Deploy(ctx); err != nil {
-		return err
+		return fmt.Errorf("failed to deploy CNI: %w", err)
 	}
 	d.CNI.SetKClient(kClient)
 	tCtx, cancel = context.WithTimeout(ctx, healthTimeout)
 	defer cancel()
 	if err := d.CNI.Healthy(tCtx); err != nil {
-		return err
+		return fmt.Errorf("failed to check CNI is healthy: %w", err)
 	}
 	log.Infof("CNI healthy")
 	for _, c := range d.Controllers {
 		log.Infof("Deploying controller...")
 		if err := c.Deploy(ctx); err != nil {
-			return err
+			return fmt.Errorf("failed to deploy controller: %w", err)
 		}
 		c.SetKClient(kClient)
 		tCtx, cancel = context.WithTimeout(ctx, healthTimeout)
 		defer cancel()
 		if err := c.Healthy(tCtx); err != nil {
-			return err
+			return fmt.Errorf("failed to check controller is healthy: %w", err)
 		}
 	}
 	log.Infof("Controllers deployed and healthy")
@@ -336,7 +336,7 @@ func (d *Deployment) Deploy(ctx context.Context, kubecfg string) (rerr error) {
 func (d *Deployment) Delete() error {
 	log.Infof("Deleting cluster...")
 	if err := d.Cluster.Delete(); err != nil {
-		return err
+		return fmt.Errorf("failed to delete cluster: %w", err)
 	}
 	log.Infof("Cluster deleted")
 	return nil
@@ -344,26 +344,26 @@ func (d *Deployment) Delete() error {
 
 func (d *Deployment) Healthy(ctx context.Context) error {
 	if err := d.Cluster.Healthy(); err != nil {
-		return err
+		return fmt.Errorf("failed to check cluster is healthy: %w", err)
 	}
 	log.Infof("Cluster healthy")
 	tCtx, cancel := context.WithTimeout(ctx, healthTimeout)
 	defer cancel()
 	if err := d.Ingress.Healthy(tCtx); err != nil {
-		return err
+		return fmt.Errorf("failed to check ingress is healthy: %w", err)
 	}
 	log.Infof("Ingress healthy")
 	tCtx, cancel = context.WithTimeout(ctx, healthTimeout)
 	defer cancel()
 	if err := d.CNI.Healthy(tCtx); err != nil {
-		return err
+		return fmt.Errorf("failed to check CNI is healthy: %w", err)
 	}
 	log.Infof("CNI healthy")
 	for _, c := range d.Controllers {
 		tCtx, cancel = context.WithTimeout(ctx, healthTimeout)
 		defer cancel()
 		if err := c.Healthy(tCtx); err != nil {
-			return err
+			return fmt.Errorf("failed to check controller is healthy: %w", err)
 		}
 	}
 	log.Infof("Controllers healthy")
@@ -488,7 +488,7 @@ func (k *KindSpec) checkDependencies() error {
 	if k.Version != "" {
 		wantV, err := getVersion(k.Version)
 		if err != nil {
-			return err
+			return fmt.Errorf("failed to parse desired kind version: %w", err)
 		}
 
 		stdout, err := outCommand("kind", "version")
@@ -545,8 +545,9 @@ func (k *KindSpec) create() error {
 		args = append(args, "--config", k.KindConfigFile)
 	}
 	log.Infof("Creating kind cluster with: %v", args)
+	// TODO
 	if err := logCommand("kind", args...); err != nil {
-		return fmt.Errorf("failed to create cluster: %w", err)
+		return err
 	}
 	log.Infof("Deployed kind cluster: %s", k.Name)
 	return nil
@@ -554,11 +555,11 @@ func (k *KindSpec) create() error {
 
 func (k *KindSpec) Deploy(ctx context.Context) error {
 	if err := k.checkDependencies(); err != nil {
-		return err
+		return fmt.Errorf("failed to check for dependencies: %w", err)
 	}
 
 	if err := k.create(); err != nil {
-		return err
+		return fmt.Errorf("failed to create kind cluster: %w", err)
 	}
 
 	// If the script is found, then run it. Else silently ignore it.
@@ -566,7 +567,7 @@ func (k *KindSpec) Deploy(ctx context.Context) error {
 	// be acceptable for the Cisco 8000e container.
 	if _, err := os.Stat(setPIDMaxScript); err == nil {
 		if err := logCommand(setPIDMaxScript); err != nil {
-			return err
+			return fmt.Errorf("failed to exec set_pid_max script: %w", err)
 		}
 	}
 
@@ -600,7 +601,7 @@ func (k *KindSpec) Delete() error {
 		args = append(args, "--name", k.Name)
 	}
 	if err := logCommand("kind", args...); err != nil {
-		return fmt.Errorf("failed to delete cluster using cli: %w", err)
+		return fmt.Errorf("failed to delete cluster: %w", err)
 	}
 	return nil
 }
@@ -820,13 +821,13 @@ func (m *MetalLBSpec) Deploy(ctx context.Context) error {
 	if m.dClient == nil {
 		m.dClient, err = dclient.NewClientWithOpts(dclient.FromEnv)
 		if err != nil {
-			return err
+			return fmt.Errorf("failed to create docker client: %w", err)
 		}
 	}
 	if m.mClient == nil {
 		m.mClient, err = metallbclientv1.NewForConfig(m.rCfg)
 		if err != nil {
-			return err
+			return fmt.Errorf("failed to create metallb client: %w", err)
 		}
 	}
 
@@ -851,7 +852,7 @@ func (m *MetalLBSpec) Deploy(ctx context.Context) error {
 	}
 	log.Infof("Deploying MetalLB from: %s", m.Manifest)
 	if err := logCommand("kubectl", "apply", "-f", m.Manifest); err != nil {
-		return err
+		return fmt.Errorf("failed to deploy metallb: %w", err)
 	}
 	if _, err := m.kClient.CoreV1().Secrets("metallb-system").Get(ctx, "memberlist", metav1.GetOptions{}); err != nil {
 		log.Infof("Creating metallb secret")
@@ -868,13 +869,13 @@ func (m *MetalLBSpec) Deploy(ctx context.Context) error {
 			},
 		}
 		if _, err := m.kClient.CoreV1().Secrets("metallb-system").Create(ctx, s, metav1.CreateOptions{}); err != nil {
-			return err
+			return fmt.Errorf("failed to create metallb secret: %w", err)
 		}
 	}
 
 	// Wait for metallb to be healthy
 	if err := m.Healthy(ctx); err != nil {
-		return err
+		return fmt.Errorf("metallb not healthy: %w", err)
 	}
 
 	if _, err = m.mClient.IPAddressPool("metallb-system").Get(ctx, "kne-service-pool", metav1.GetOptions{}); err != nil {
@@ -882,7 +883,7 @@ func (m *MetalLBSpec) Deploy(ctx context.Context) error {
 		// Get Network information from docker.
 		nr, err := m.dClient.NetworkList(ctx, dtypes.NetworkListOptions{})
 		if err != nil {
-			return err
+			return fmt.Errorf("failed to get docker network list: %w", err)
 		}
 		var network dtypes.NetworkResource
 		for _, v := range nr {
@@ -899,7 +900,7 @@ func (m *MetalLBSpec) Deploy(ctx context.Context) error {
 		for _, ipRange := range network.IPAM.Config {
 			_, ipNet, err := net.ParseCIDR(ipRange.Subnet)
 			if err != nil {
-				return err
+				return fmt.Errorf("failed to parse cidr: %w", err)
 			}
 			if ipNet.IP.To4() != nil {
 				n = ipNet
@@ -932,7 +933,7 @@ func (m *MetalLBSpec) Deploy(ctx context.Context) error {
 			},
 		}
 		if _, err = m.mClient.L2Advertisement("metallb-system").Create(ctx, l2Advert, metav1.CreateOptions{}); err != nil {
-			return err
+			return fmt.Errorf("failed to create metallb L2 advertisement: %w", err)
 		}
 	}
 	return nil
@@ -981,7 +982,7 @@ func (m *MeshnetSpec) Deploy(ctx context.Context) error {
 	}
 	log.Infof("Deploying Meshnet from: %s", m.Manifest)
 	if err := logCommand("kubectl", "apply", "-f", m.Manifest); err != nil {
-		return err
+		return fmt.Errorf("failed to deploy meshnet: %w", err)
 	}
 	log.Infof("Meshnet Deployed")
 	return nil
@@ -998,10 +999,10 @@ func (m *MeshnetSpec) Healthy(ctx context.Context) error {
 	for {
 		select {
 		case <-ctx.Done():
-			return fmt.Errorf("context canceled before healthy")
+			return fmt.Errorf("context canceled before meshnet healthy")
 		case e, ok := <-w.ResultChan():
 			if !ok {
-				return fmt.Errorf("watch channel closed before healthy")
+				return fmt.Errorf("watch channel closed before meshnet healthy")
 			}
 			d, ok := e.Object.(*appsv1.DaemonSet)
 			if !ok {
@@ -1055,7 +1056,7 @@ func (c *CEOSLabSpec) Deploy(ctx context.Context) error {
 	}
 	log.Infof("Deploying CEOSLab controller from: %s", c.Operator)
 	if err := logCommand("kubectl", "apply", "-f", c.Operator); err != nil {
-		return err
+		return fmt.Errorf("failed to deploy ceoslab operator: %w", err)
 	}
 	log.Infof("CEOSLab controller deployed")
 	return nil
@@ -1104,7 +1105,7 @@ func (l *LemmingSpec) Deploy(ctx context.Context) error {
 	}
 	log.Infof("Deploying Lemming controller from: %s", l.Operator)
 	if err := logCommand("kubectl", "apply", "-f", l.Operator); err != nil {
-		return err
+		return fmt.Errorf("failed to deploy lemming operator: %w", err)
 	}
 	log.Infof("Lemming controller deployed")
 	return nil
@@ -1153,7 +1154,7 @@ func (s *SRLinuxSpec) Deploy(ctx context.Context) error {
 	}
 	log.Infof("Deploying SRLinux controller from: %s", s.Operator)
 	if err := logCommand("kubectl", "apply", "-f", s.Operator); err != nil {
-		return err
+		return fmt.Errorf("failed to deploy srlinux operator: %w", err)
 	}
 	log.Infof("SRLinux controller deployed")
 	return nil
@@ -1204,7 +1205,7 @@ func (i *IxiaTGSpec) Deploy(ctx context.Context) error {
 	}
 	log.Infof("Deploying IxiaTG controller from: %s", i.Operator)
 	if err := logCommand("kubectl", "apply", "-f", i.Operator); err != nil {
-		return err
+		return fmt.Errorf("failed to deploy ixiatg operator: %w", err)
 	}
 
 	if i.ConfigMap == "" && i.ManifestDir != "" {
@@ -1238,7 +1239,7 @@ func (i *IxiaTGSpec) Deploy(ctx context.Context) error {
 	}
 	log.Infof("Deploying IxiaTG config map from: %s", i.ConfigMap)
 	if err := logCommand("kubectl", "apply", "-f", i.ConfigMap); err != nil {
-		return err
+		return fmt.Errorf("failed to deploy ixiatg config map: %w", err)
 	}
 	log.Infof("IxiaTG controller deployed")
 	return nil
@@ -1252,16 +1253,16 @@ func deploymentHealthy(ctx context.Context, c kubernetes.Interface, name string)
 	log.Infof("Waiting on deployment %q to be healthy", name)
 	w, err := c.AppsV1().Deployments(name).Watch(ctx, metav1.ListOptions{})
 	if err != nil {
-		return err
+		return fmt.Errorf("failed to create watcher for deployment %q", name)
 	}
 	ch := w.ResultChan()
 	for {
 		select {
 		case <-ctx.Done():
-			return fmt.Errorf("context canceled before healthy")
+			return fmt.Errorf("context canceled before %q healthy", name)
 		case e, ok := <-ch:
 			if !ok {
-				return fmt.Errorf("watch channel closed before healthy")
+				return fmt.Errorf("watch channel closed before %q healthy", name)
 			}
 			d, ok := e.Object.(*appsv1.Deployment)
 			if !ok {
diff --git a/topo/node/node.go b/topo/node/node.go
index 7072c493..f9dc7e6d 100644
--- a/topo/node/node.go
+++ b/topo/node/node.go
@@ -32,6 +32,7 @@ type Interface interface {
 	Name() string
 	GetNamespace() string
 	GetProto() *tpb.Node
+	String() string
 }
 
 type Implementation interface {
@@ -140,6 +141,10 @@ func (n *Impl) GetNamespace() string {
 	return n.Namespace
 }
 
+func (n *Impl) String() string {
+	return fmt.Sprintf("%q (vendor: %q, model: %q)", n.Proto.Name, n.Proto.Vendor, n.Proto.Model)
+}
+
 func (n *Impl) TopologySpecs(context.Context) ([]*topologyv1.Topology, error) {
 	proto := n.GetProto()
 
diff --git a/topo/topo.go b/topo/topo.go
index 11e39a08..36d95209 100644
--- a/topo/topo.go
+++ b/topo/topo.go
@@ -254,10 +254,10 @@ func (m *Manager) Create(ctx context.Context, timeout time.Duration) (rerr error
 		}()
 	}
 	if err := m.push(ctx); err != nil {
-		return err
+		return fmt.Errorf("failed to create topology %q: %w", m.topo.GetName(), err)
 	}
 	if err := m.checkNodeStatus(ctx, timeout); err != nil {
-		return err
+		return fmt.Errorf("failed to check status of nodes in topology %q: %w", m.topo.GetName(), err)
 	}
 	log.Infof("Topology %q created", m.topo.GetName())
 	return nil
@@ -279,12 +279,15 @@ func (m *Manager) Delete(ctx context.Context) error {
 	}
 
 	if err := m.deleteMeshnetTopologies(ctx); err != nil {
-		return err
+		return fmt.Errorf("failed to delete meshnet topologies for topology %q: %w", m.topo.GetName(), err)
 	}
 
 	// Delete namespace
 	prop := metav1.DeletePropagationForeground
-	return m.kClient.CoreV1().Namespaces().Delete(ctx, m.topo.Name, metav1.DeleteOptions{PropagationPolicy: &prop})
+	if err := m.kClient.CoreV1().Namespaces().Delete(ctx, m.topo.Name, metav1.DeleteOptions{PropagationPolicy: &prop}); err != nil {
+		return fmt.Errorf("failed to delete namespace %q: %w", m.topo.Name, err)
+	}
+	return nil
 }
 
 // Show returns the topology information including services and node health.
@@ -470,27 +473,27 @@ func (m *Manager) push(ctx context.Context) error {
 		}
 		sNs, err := m.kClient.CoreV1().Namespaces().Create(ctx, ns, metav1.CreateOptions{})
 		if err != nil {
-			return err
+			return fmt.Errorf("failed to create namespace %q: %w", ns, err)
 		}
 		log.Infof("Server Namespace: %+v", sNs)
 	}
 
 	if err := m.createMeshnetTopologies(ctx); err != nil {
-		return err
+		return fmt.Errorf("failed to create meshnet topologies: %w", err)
 	}
 
 	log.Infof("Creating Node Pods")
-	for k, n := range m.nodes {
+	for _, n := range m.nodes {
 		if err := n.Create(ctx); err != nil {
-			return err
+			return fmt.Errorf("failed to create node %s: %w", n, err)
 		}
-		log.Infof("Node %q resource created", k)
+		log.Infof("Node %s resource created", n)
 	}
 	for _, n := range m.nodes {
 		err := m.GenerateSelfSigned(ctx, n.Name())
 		switch {
 		default:
-			return fmt.Errorf("failed to generate cert for node %s: %w", n.Name(), err)
+			return fmt.Errorf("failed to generate cert for node %s: %w", n, err)
 		case err == nil, status.Code(err) == codes.Unimplemented:
 		}
 	}
@@ -549,10 +552,10 @@ func (m *Manager) checkNodeStatus(ctx context.Context, timeout time.Duration) er
 
 			phase, err := n.Status(ctx)
 			if err != nil || phase == node.StatusFailed {
-				return fmt.Errorf("Node %q: Status %s Reason %v", name, phase, err)
+				return fmt.Errorf("Node %s: Status %s Reason %v", n, phase, err)
 			}
 			if phase == node.StatusRunning {
-				log.Infof("Node %q: Status %s", name, phase)
+				log.Infof("Node %s: Status %s", n, phase)
 				processed[name] = true
 			} else {
 				foundAll = false
diff --git a/topo/topo_test.go b/topo/topo_test.go
index 16f1c4cc..4297ede0 100644
--- a/topo/topo_test.go
+++ b/topo/topo_test.go
@@ -502,7 +502,7 @@ func TestCreate(t *testing.T) {
 				},
 			},
 		},
-		wantErr: `Node "bad": Status FAILED`,
+		wantErr: `Node "bad" (vendor: "1002", model: ""): Status FAILED`,
 	}, {
 		desc: "failed to report metrics, create still passes",
 		opts: []Option{WithUsageReporting(true, "", "")},