[receiver/vcenter] add per-object fallback when perf batch query fails

.chloggen/fix_PerfMetricsQueryFollowUp.yaml

@@ -0,0 +1,27 @@
+# Use this changelog template to create an entry for release notes.
+
+# One of 'breaking', 'deprecation', 'new_component', 'enhancement', 'bug_fix'
+change_type: bug_fix
+
+# The name of the component, or a single word describing the area of concern, (e.g. receiver/filelog)
+component: receiver/vcenter
+
+# A brief description of the change.  Surround your text with quotes ("") if it needs to start with a backtick (`).
+note: added logic so that if a query fails due to a bad object it does not fail the entire scrape
+
+# Mandatory: One or more tracking issues related to the change. You can use the PR number here if no issue exists.
+issues: [46977]
+
+# (Optional) One or more lines of additional information to render under the primary note.
+# These lines will be padded with 2 spaces and then inserted directly into the document.
+# Use pipe (|) for multiline entries.
+subtext:
+
+# If your change doesn't affect end users or the exported elements of any package,
+# you should instead start your pull request title with [chore] or use the "Skip Changelog" label.
+# Optional: The change log or logs in which this entry should be included.
+# e.g. '[user]' or '[user, api]'
+# Include 'user' if the change is relevant to end users.
+# Include 'api' if there is a change to a library API.
+# Default: '[user]'
+change_logs: []

receiver/vcenterreceiver/client.go

@@ -327,7 +327,6 @@ func (vc *vcenterClient) PerfMetricsQuery(
 		return &perfMetricsQueryResult{}, nil
 	}
 	vc.pm.Sort = true
-
 	batchSize := vc.batchSizeForMetrics(len(names))
 	if batchSize <= 0 || batchSize >= len(objs) {
 		batchSize = max(len(objs), 1)
@@ -337,7 +336,22 @@ func (vc *vcenterClient) PerfMetricsQuery(
 	for batch := range slices.Chunk(objs, batchSize) {
 		sample, err := vc.pm.SampleByName(ctx, spec, names, batch)
 		if err != nil {
-			return nil, err
+			for _, obj := range batch {
+				singleSample, singleErr := vc.pm.SampleByName(ctx, spec, names, []vt.ManagedObjectReference{obj})
+				if singleErr != nil {
+					continue
+				}
+
+				singleResult, singleErr := vc.pm.ToMetricSeries(ctx, singleSample)
+				if singleErr != nil {
+					continue
+				}
+
+				for i := range singleResult {
+					resultsByRef[singleResult[i].Entity.Value] = &singleResult[i]
+				}
+			}
+			continue
 		}
 		result, err := vc.pm.ToMetricSeries(ctx, sample)
 		if err != nil {
@@ -348,6 +362,7 @@ func (vc *vcenterClient) PerfMetricsQuery(
 			resultsByRef[result[i].Entity.Value] = &result[i]
 		}
 	}
+
 	return &perfMetricsQueryResult{
 		resultsByRef: resultsByRef,
 	}, nil

receiver/vcenterreceiver/client_test.go

@@ -218,6 +218,30 @@ func TestPerfMetricsQuery(t *testing.T) {
 	}, esx)
 }
 
+func TestPerfMetricsQuery_FallbackOnBatchError(t *testing.T) {
+	esx := simulator.ESX()
+	simulator.Test(func(ctx context.Context, c *vim25.Client) {
+		pm := performance.NewManager(c)
+		m := view.NewManager(c)
+		finder := find.NewFinder(c)
+		client := vcenterClient{
+			vimDriver: c,
+			vm:        m,
+			pm:        pm,
+			finder:    finder,
+		}
+		hs, err := finder.DefaultHostSystem(ctx)
+		require.NoError(t, err)
+
+		spec := types.PerfQuerySpec{Format: string(types.PerfFormatNormal), IntervalId: int32(20)}
+
+		metrics, err := client.PerfMetricsQuery(ctx, spec, []string{"invalid.metric.name"}, []types.ManagedObjectReference{hs.Reference()})
+		require.NoError(t, err)
+		require.NotNil(t, metrics)
+		require.Empty(t, metrics.resultsByRef)
+	}, esx)
+}
+
 func TestPerfMetricsQueryBatching(t *testing.T) {
 	vpx := simulator.VPX()
 	vpx.Host = 10