fix(cloudfoundry): avoid typed-nil DCAClient in cf_vm workloadmeta collector

comp/core/workloadmeta/collectors/internal/cloudfoundry/vm/cf_vm.go

@@ -10,6 +10,7 @@ import (
 	"context"
 	"encoding/json"
 	"fmt"
+	"sync"
 	"time"
 
 	"go.uber.org/fx"
@@ -29,6 +30,10 @@ const (
 	componentName = "workloadmeta-cloudfoundry-vm"
 )
 
+// getClusterAgentClient is overridden in tests to avoid depending on the
+// real cluster agent connection singleton.
+var getClusterAgentClient = clusteragent.GetClusterAgentClient
+
 type dependencies struct {
 	fx.In
 
@@ -45,8 +50,9 @@ type collector struct {
 	gardenUtil cloudfoundry.GardenUtilInterface
 	nodeName   string
 
-	dcaClient  clusteragent.DCAClientInterface
-	dcaEnabled bool
+	dcaClientLock sync.Mutex
+	dcaClient     clusteragent.DCAClientInterface
+	dcaEnabled    bool
 }
 
 // NewCollector instantiates a CollectorProvider which can provide a CF container collector
@@ -108,7 +114,7 @@ func (c *collector) Pull(_ context.Context) error {
 
 	var allContainersTags map[string][]string
 	if dcaClient := c.getDCAClient(); dcaClient != nil {
-		allContainersTags, err = c.dcaClient.GetCFAppsMetadataForNode(c.nodeName)
+		allContainersTags, err = dcaClient.GetCFAppsMetadataForNode(c.nodeName)
 		if err != nil {
 			log.Debugf("Unable to fetch CF tags from cluster agent, CF tags will be missing, err: %v", err)
 		}
@@ -223,6 +229,9 @@ func (c *collector) Pull(_ context.Context) error {
 }
 
 func (c *collector) getDCAClient() clusteragent.DCAClientInterface {
+	c.dcaClientLock.Lock()
+	defer c.dcaClientLock.Unlock()
+
 	if !c.dcaEnabled {
 		return nil
 	}
@@ -231,13 +240,13 @@ func (c *collector) getDCAClient() clusteragent.DCAClientInterface {
 		return c.dcaClient
 	}
 
-	var err error
-	c.dcaClient, err = clusteragent.GetClusterAgentClient()
+	client, err := getClusterAgentClient()
 	if err != nil {
 		log.Debugf("Could not initialise the communication with the cluster agent, PCF tags may be missing, err: %v", err)
 		return nil
 	}
 
+	c.dcaClient = client
 	return c.dcaClient
 }
 

comp/core/workloadmeta/collectors/internal/cloudfoundry/vm/cf_vm_test.go

@@ -509,3 +509,40 @@ func TestPullAppNameWithGardenPropertiesWithoutDCA(t *testing.T) {
 
 	assert.Contains(t, container.CollectorTags, "container_name:"+"app-name-1")
 }
+
+// TestPullDCAUnreachableTwice reproduces the typed-nil interface trap: when
+// clusteragent.GetClusterAgentClient returns a concrete (*DCAClient)(nil)
+// alongside an error, assigning it into an interface field before the error
+// check poisons the field with a non-nil interface wrapping a nil pointer.
+// A second Pull would then see c.dcaClient != nil, skip re-init, and call a
+// method on a nil *DCAClient receiver.
+func TestPullDCAUnreachableTwice(t *testing.T) {
+	fakeGardenUtil := FakeGardenUtil{containers: nil}
+
+	orig := getClusterAgentClient
+	getClusterAgentClient = func() (*clusteragent.DCAClient, error) {
+		// Simulate GetClusterAgentClient's real error path: a nil concrete
+		// pointer alongside a non-nil error.
+		return nil, fmt.Errorf("cluster agent unreachable")
+	}
+	t.Cleanup(func() { getClusterAgentClient = orig })
+
+	workloadmetaStore := fxutil.Test[workloadmetamock.Mock](t, fx.Options(
+		core.MockBundle(),
+		workloadmetafxmock.MockModule(workloadmeta.NewParams()),
+	))
+
+	c := collector{
+		gardenUtil: &fakeGardenUtil,
+		store:      workloadmetaStore,
+		dcaEnabled: true,
+	}
+
+	require.NoError(t, c.Pull(context.TODO()))
+	require.Nil(t, c.dcaClient, "first Pull must not poison the dcaClient field on error")
+
+	require.NotPanics(t, func() {
+		require.NoError(t, c.Pull(context.TODO()))
+	})
+	require.Nil(t, c.dcaClient, "second Pull must not poison the dcaClient field on error")
+}