hostmetrics: use WMI to fetch ppid

.chloggen/braydonk_hostmetrics_wmi_parent_pid.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: enhancement
+
+# The name of the component, or a single word describing the area of concern, (e.g. filelogreceiver)
+component: hostmetricsreceiver
+
+# A brief description of the change.  Surround your text with quotes ("") if it needs to start with a backtick (`).
+note: Use Windows Management Interface to fetch process.ppid by default. Add `disable_wmi` config option to fallback to old behaviour.
+
+# Mandatory: One or more tracking issues related to the change. You can use the PR number here if no issue exists.
+issues: [32947]
+
+# (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: This also made a change where the parent process ID will not be fetched if the resource attribute is disabled.
+
+# 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: [user]

receiver/hostmetricsreceiver/README.md

@@ -123,6 +123,7 @@ process:
   mute_process_io_error: <true|false>
   mute_process_user_error: <true|false>
   mute_process_cgroup_error: <true|false>
+  wmi_enabled: <true|false>
   scrape_process_delay: <time>
 ```
 
@@ -133,6 +134,19 @@ The following settings are optional:
 - `mute_process_cgroup_error` (default: false): mute the error encountered when trying to read the cgroup of a process the collector does not have permission to read. This flag is ignored when `mute_process_all_errors` is set to true as all errors are muted.
 - `mute_process_exe_error` (default: false): mute the error encountered when trying to read the executable path of a process the collector does not have permission to read (Linux only). This flag is ignored when `mute_process_all_errors` is set to true as all errors are muted.
 - `mute_process_user_error` (default: false): mute the error encountered when trying to read a uid which doesn't exist on the system, eg. is owned by a user that only exists in a container. This flag is ignored when `mute_process_all_errors` is set to true as all errors are muted.
+- `wmi_enabled` (default: true): allow the scraper to use [Windows Management Instrumentation (WMI)](https://learn.microsoft.com/en-us/windows/win32/wmisdk/wmi-start-page) to fetch some information on Windows. This option has no effect on non-Windows environments.
+
+#### High CPU Usage On Windows
+
+Getting the Parent Process ID of all processes on Windows is a very expensive operation. There are two options to combat this:
+* Allow the collector to use WMI, this is the default behaviour with the `wmi_enabled` configuration option
+* Disable Parent Process ID collection like so:
+```yaml
+process:
+  resource_attributes:
+    process.parent_pid:
+      enabled: false
+```
 
 ## Advanced Configuration
 

receiver/hostmetricsreceiver/config_test.go

@@ -108,23 +108,48 @@ func TestLoadConfig(t *testing.T) {
 	}
 }
 
-func TestLoadInvalidConfig_NoScrapers(t *testing.T) {
-	factory := NewFactory()
-	cfg := factory.CreateDefaultConfig()
-
-	cm, err := confmaptest.LoadConf(filepath.Join("testdata", "config-noscrapers.yaml"))
-	require.NoError(t, err)
+func TestLoadInvalidConfig(t *testing.T) {
+	testCases := []struct {
+		name          string
+		configPath    string
+		failMarshal   bool
+		errorContains string
+	}{
+		{
+			name:          "no scrapers",
+			configPath:    "config-noscrapers.yaml",
+			errorContains: "must specify at least one scraper when using hostmetrics receiver",
+		},
+		{
+			name:          "invalid scraper key",
+			configPath:    "config-invalidscraperkey.yaml",
+			errorContains: "invalid scraper key: invalidscraperkey",
+			failMarshal:   true,
+		},
+		{
+			name:          "handles enabled wmi disabled",
+			configPath:    "config-handles-no-wmi.yaml",
+			errorContains: processscraper.ErrProcessHandlesRequiresWMI.Error(),
+		},
+	}
 
-	require.NoError(t, cm.Unmarshal(cfg))
-	require.ErrorContains(t, xconfmap.Validate(cfg), "must specify at least one scraper when using hostmetrics receiver")
-}
+	for _, tc := range testCases {
+		tc := tc
+		t.Run(tc.name, func(t *testing.T) {
+			t.Parallel()
 
-func TestLoadInvalidConfig_InvalidScraperKey(t *testing.T) {
-	factory := NewFactory()
-	cfg := factory.CreateDefaultConfig()
+			factory := NewFactory()
+			cfg := factory.CreateDefaultConfig()
 
-	cm, err := confmaptest.LoadConf(filepath.Join("testdata", "config-invalidscraperkey.yaml"))
-	require.NoError(t, err)
+			cm, err := confmaptest.LoadConf(filepath.Join("testdata", tc.configPath))
+			require.NoError(t, err)
 
-	require.ErrorContains(t, cm.Unmarshal(cfg), "invalid scraper key: invalidscraperkey")
+			if tc.failMarshal {
+				require.ErrorContains(t, cm.Unmarshal(cfg), tc.errorContains)
+			} else {
+				require.NoError(t, cm.Unmarshal(cfg))
+				require.ErrorContains(t, xconfmap.Validate(cfg), tc.errorContains)
+			}
+		})
+	}
 }

receiver/hostmetricsreceiver/integration_test.go

@@ -6,13 +6,18 @@
 package hostmetricsreceiver
 
 import (
+	"context"
 	"os/exec"
 	"path/filepath"
+	"runtime"
 	"testing"
 	"time"
 
 	"github.com/stretchr/testify/require"
 	"go.opentelemetry.io/collector/component"
+	"go.opentelemetry.io/collector/component/componenttest"
+	"go.opentelemetry.io/collector/consumer/consumertest"
+	"go.opentelemetry.io/collector/receiver/receivertest"
 
 	"github.com/open-telemetry/opentelemetry-collector-contrib/internal/coreinternal/scraperinttest"
 	"github.com/open-telemetry/opentelemetry-collector-contrib/internal/filter/filterset"
@@ -116,3 +121,71 @@ func Test_ProcessScrapeWithBadRootPathAndEnvVar(t *testing.T) {
 		),
 	).Run(t)
 }
+
+func Test_Windows_ProcessScrapeWMIInformation(t *testing.T) {
+	if runtime.GOOS != "windows" {
+		t.Skip("this integration test is windows exclusive")
+	}
+
+	factory := NewFactory()
+
+	rCfg := &Config{}
+	rCfg.CollectionInterval = time.Second
+	f := processscraper.NewFactory()
+	pCfg := f.CreateDefaultConfig().(*processscraper.Config)
+	pCfg.Metrics.ProcessHandles.Enabled = true
+	pCfg.ResourceAttributes.ProcessParentPid.Enabled = true
+	rCfg.Scrapers = map[component.Type]component.Config{
+		f.Type(): pCfg,
+	}
+	cfg := component.Config(rCfg)
+
+	sink := new(consumertest.MetricsSink)
+	r, err := factory.CreateMetrics(context.Background(), receivertest.NewNopSettings(factory.Type()), cfg, sink)
+	require.NoError(t, err)
+
+	// Start the receiver and give it an extra 250 milliseconds after the
+	// collection interval to perform the scrape.
+	r.Start(context.Background(), componenttest.NewNopHost())
+	time.Sleep(rCfg.CollectionInterval + 250*time.Millisecond)
+	r.Shutdown(context.Background())
+
+	// The actual results of the test are non-deterministic, but
+	// all we want to know is whether the handles and parent PID
+	// metrics are being retrieved successfully on at least some
+	// metrics (it doesn't need to work for every process).
+	metrics := sink.AllMetrics()
+	var foundValidParentPid, foundValidHandles int
+	for _, m := range metrics {
+		rms := m.ResourceMetrics()
+		for i := 0; i < rms.Len(); i++ {
+			rm := rms.At(i)
+
+			// Check if the resource attributes has the parent PID.
+			ppid, ok := rm.Resource().Attributes().Get("process.parent_pid")
+			if ok && ppid.Int() > 0 {
+				foundValidParentPid++
+			}
+
+			sms := rm.ScopeMetrics()
+			for j := 0; j < sms.Len(); j++ {
+				sm := sms.At(j)
+				ms := sm.Metrics()
+				for k := 0; k < ms.Len(); k++ {
+					m := ms.At(k)
+
+					// Check if this is a process.handles metric
+					// with a non-zero datapoint.
+					if m.Name() == "process.handles" &&
+						m.Sum().DataPoints().Len() > 0 &&
+						m.Sum().DataPoints().At(0).IntValue() > 0 {
+						foundValidHandles++
+					}
+				}
+			}
+		}
+	}
+
+	require.Positive(t, foundValidHandles)
+	require.Positive(t, foundValidParentPid)
+}

receiver/hostmetricsreceiver/internal/scraper/processscraper/config.go

@@ -4,12 +4,15 @@
 package processscraper // import "github.com/open-telemetry/opentelemetry-collector-contrib/receiver/hostmetricsreceiver/internal/scraper/processscraper"
 
 import (
+	"errors"
 	"time"
 
 	"github.com/open-telemetry/opentelemetry-collector-contrib/internal/filter/filterset"
 	"github.com/open-telemetry/opentelemetry-collector-contrib/receiver/hostmetricsreceiver/internal/scraper/processscraper/internal/metadata"
 )
 
+var ErrProcessHandlesRequiresWMI = errors.New("the process.handles metric requires WMI to be enabled")
+
 // Config relating to Process Metric Scraper.
 type Config struct {
 	// MetricsBuilderConfig allows to customize scraped metrics/attributes representation.
@@ -53,6 +56,15 @@ type Config struct {
 	// ScrapeProcessDelay is used to indicate the minimum amount of time a process must be running
 	// before metrics are scraped for it.  The default value is 0 seconds (0s).
 	ScrapeProcessDelay time.Duration `mapstructure:"scrape_process_delay"`
+
+	WMIEnabled bool `mapstructure:"wmi_enabled"`
+}
+
+func (cfg *Config) Validate() error {
+	if !cfg.WMIEnabled && cfg.Metrics.ProcessHandles.Enabled {
+		return ErrProcessHandlesRequiresWMI
+	}
+	return nil
 }
 
 type MatchConfig struct {

receiver/hostmetricsreceiver/internal/scraper/processscraper/documentation.md

@@ -114,7 +114,7 @@ Number of disk operations performed by the process.
 
 Number of handles held by the process.
 
-This metric is only available on Windows.
+This metric is only available on Windows. It requires Windows Management Interface to be enabled.
 
 | Unit | Metric Type | Value Type | Aggregation Temporality | Monotonic |
 | ---- | ----------- | ---------- | ----------------------- | --------- |

receiver/hostmetricsreceiver/internal/scraper/processscraper/factory.go

@@ -35,6 +35,7 @@ func NewFactory() scraper.Factory {
 func createDefaultConfig() component.Config {
 	return &Config{
 		MetricsBuilderConfig: metadata.DefaultMetricsBuilderConfig(),
+		WMIEnabled:           true,
 	}
 }