[receiver/hostmetrics] Add process.paging.faults metric

.chloggen/add-process-paging-faults-metric.yaml

@@ -0,0 +1,16 @@
+# 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: Add a new metric `process.paging.faults` to the `process` scraper of the `hostmetrics` receiver.
+
+# One or more tracking issues related to the change
+issues: [14084]
+
+# (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:

receiver/hostmetricsreceiver/hostmetrics_receiver_test.go

@@ -176,8 +176,10 @@ func assertIncludesExpectedMetrics(t *testing.T, got pmetric.Metrics) {
 		return
 	}
 
-	assert.Equal(t, len(resourceMetrics), len(returnedResourceMetrics))
-	for _, expected := range resourceMetrics {
+	var expectedResourceMetrics []string
+	expectedResourceMetrics = append(expectedResourceMetrics, resourceMetrics...)
+	assert.Equal(t, len(expectedResourceMetrics), len(returnedResourceMetrics))
+	for _, expected := range expectedResourceMetrics {
 		assert.Contains(t, returnedResourceMetrics, expected)
 	}
 }

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

@@ -12,6 +12,7 @@ These are the metrics available for this scraper.
 | **process.disk.io** | Disk bytes transferred. | By | Sum(Int) | <ul> <li>direction</li> </ul> |
 | **process.memory.physical_usage** | The amount of physical memory in use. | By | Sum(Int) | <ul> </ul> |
 | **process.memory.virtual_usage** | Virtual memory size. | By | Sum(Int) | <ul> </ul> |
+| process.paging.faults | Number of page faults the process has made. This metric is only available on Linux. | {faults} | Sum(Int) | <ul> <li>type</li> </ul> |
 | process.threads | Process threads count. | {threads} | Sum(Int) | <ul> </ul> |
 
 **Highlighted metrics** are emitted by default. Other metrics are optional and not emitted by default.
@@ -41,3 +42,4 @@ metrics:
 | ---- | ----------- | ------ |
 | direction | Direction of flow of bytes (read or write). | read, write |
 | state | Breakdown of CPU usage by type. | system, user, wait |
+| type | Type of memory paging fault. | major, minor |

receiver/hostmetricsreceiver/internal/scraper/processscraper/metadata.yaml

@@ -47,6 +47,10 @@ attributes:
     description: Breakdown of CPU usage by type.
     enum: [system, user, wait]
 
+  type:
+    description: Type of memory paging fault.
+    enum: [major, minor]
+
 metrics:
   process.cpu.time:
     enabled: true
@@ -86,6 +90,16 @@ metrics:
       monotonic: true
     attributes: [direction]
 
+  process.paging.faults:
+    enabled: false
+    description: Number of page faults the process has made. This metric is only available on Linux.
+    unit: "{faults}"
+    sum:
+      value_type: int
+      aggregation: cumulative
+      monotonic: true
+    attributes: [type]
+
   process.threads:
     enabled: false
     description: Process threads count.

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

@@ -93,6 +93,7 @@ type processHandle interface {
 	NumThreads() (int32, error)
 	CreateTime() (int64, error)
 	Parent() (*process.Process, error)
+	PageFaults() (*process.PageFaultsStat, error)
 }
 
 type gopsProcessHandles struct {

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

@@ -33,9 +33,10 @@ const (
 	cpuMetricsLen    = 1
 	memoryMetricsLen = 2
 	diskMetricsLen   = 1
+	pagingMetricsLen = 1
 	threadMetricsLen = 1
 
-	metricsLen = cpuMetricsLen + memoryMetricsLen + diskMetricsLen + threadMetricsLen
+	metricsLen = cpuMetricsLen + memoryMetricsLen + diskMetricsLen + pagingMetricsLen + threadMetricsLen
 )
 
 // scraper for Process Metrics
@@ -113,6 +114,10 @@ func (s *scraper) scrape(_ context.Context) (pmetric.Metrics, error) {
 			errs.AddPartial(diskMetricsLen, fmt.Errorf("error reading disk usage for process %q (pid %v): %w", md.executable.name, md.pid, err))
 		}
 
+		if err = s.scrapeAndAppendPagingMetric(now, md.handle); err != nil {
+			errs.AddPartial(pagingMetricsLen, fmt.Errorf("error reading memory paging info for process %q (pid %v): %w", md.executable.name, md.pid, err))
+		}
+
 		if err = s.scrapeAndAppendThreadsMetrics(now, md.handle); err != nil {
 			errs.AddPartial(threadMetricsLen, fmt.Errorf("error reading thread info for process %q (pid %v): %w", md.executable.name, md.pid, err))
 		}
@@ -229,6 +234,21 @@ func (s *scraper) scrapeAndAppendDiskIOMetric(now pcommon.Timestamp, handle proc
 	return nil
 }
 
+func (s *scraper) scrapeAndAppendPagingMetric(now pcommon.Timestamp, handle processHandle) error {
+	if !s.config.Metrics.ProcessPagingFaults.Enabled {
+		return nil
+	}
+
+	pageFaultsStat, err := handle.PageFaults()
+	if err != nil {
+		return err
+	}
+
+	s.mb.RecordProcessPagingFaultsDataPoint(now, int64(pageFaultsStat.MajorFaults), metadata.AttributeTypeMajor)
+	s.mb.RecordProcessPagingFaultsDataPoint(now, int64(pageFaultsStat.MinorFaults), metadata.AttributeTypeMinor)
+	return nil
+}
+
 func (s *scraper) scrapeAndAppendThreadsMetrics(now pcommon.Timestamp, handle processHandle) error {
 	if !s.config.Metrics.ProcessThreads.Enabled {
 		return nil