Cgroups

.chloggen/cgroups.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: Add support for cgroup information as a new resource attribute on process metrics, as `process.cgroup`
+
+# Mandatory: One or more tracking issues related to the change. You can use the PR number here if no issue exists.
+issues: [29282]
+
+# (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/hostmetricsreceiver/internal/scraper/processscraper/config.go

@@ -31,6 +31,10 @@ type Config struct {
 	// the collector does not have permission for.
 	MuteProcessIOError bool `mapstructure:"mute_process_io_error,omitempty"`
 
+	// MuteProcessCgroupError is a flag that will mute the error encountered when trying to read the cgroup of a process
+	// the collector does not have permission for.
+	MuteProcessCgroupError bool `mapstructure:"mute_process_cgroup_error,omitempty"`
+
 	// ResilientProcessScraping is a flag that will let the collector continue reading a process even when
 	// the collector does not have permission to read it's executable path (Linux)
 	MuteProcessExeError bool `mapstructure:"mute_process_exe_error,omitempty"`

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

@@ -178,6 +178,7 @@ Process threads count.
 
 | Name | Description | Values | Enabled |
 | ---- | ----------- | ------ | ------- |
+| process.cgroup | cgroup associated with the process. | Any Str | false |
 | process.command | The command used to launch the process (i.e. the command name). On Linux based systems, can be set to the zeroth string in proc/[pid]/cmdline. On Windows, can be set to the first parameter extracted from GetCommandLineW. | Any Str | true |
 | process.command_line | The full command used to launch the process as a single string representing the full command. On Windows, can be set to the result of GetCommandLineW. Do not set this if you have to assemble it just for monitoring; use process.command_args instead. | Any Str | true |
 | process.executable.name | The name of the process executable. On Linux based systems, can be set to the Name in proc/[pid]/status. On Windows, can be set to the base name of GetProcessImageFileNameW. | Any Str | true |

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

@@ -28,6 +28,8 @@ all_set:
     process.threads:
       enabled: true
   resource_attributes:
+    process.cgroup:
+      enabled: true
     process.command:
       enabled: true
     process.command_line:
@@ -71,6 +73,8 @@ none_set:
     process.threads:
       enabled: false
   resource_attributes:
+    process.cgroup:
+      enabled: false
     process.command:
       enabled: false
     process.command_line:

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

@@ -46,6 +46,10 @@ resource_attributes:
     description: The username of the user that owns the process.
     enabled: true
     type: string
+  process.cgroup:
+    description: cgroup associated with the process.
+    enabled: false
+    type: string
 
 attributes:
   direction:

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

@@ -5,10 +5,14 @@ package processscraper // import "github.com/open-telemetry/opentelemetry-collec
 
 import (
 	"context"
+	"os"
+	"path/filepath"
 	"runtime"
+	"strconv"
 	"strings"
 	"time"
 
+	"github.com/shirou/gopsutil/v3/common"
 	"github.com/shirou/gopsutil/v3/cpu"
 	"github.com/shirou/gopsutil/v3/process"
 	"go.opentelemetry.io/collector/pdata/pcommon"
@@ -31,8 +35,9 @@ type processMetadata struct {
 }
 
 type executableMetadata struct {
-	name string
-	path string
+	name   string
+	path   string
+	cgroup string
 }
 
 type commandMetadata struct {
@@ -46,6 +51,7 @@ func (m *processMetadata) buildResource(rb *metadata.ResourceBuilder) pcommon.Re
 	rb.SetProcessParentPid(int64(m.parentPid))
 	rb.SetProcessExecutableName(m.executable.name)
 	rb.SetProcessExecutablePath(m.executable.path)
+	rb.SetProcessCgroup(m.executable.cgroup)
 	if m.command != nil {
 		rb.SetProcessCommand(m.command.command)
 		if m.command.commandLineSlice != nil {
@@ -91,14 +97,15 @@ type processHandle interface {
 	NumFDsWithContext(context.Context) (int32, error)
 	// If gatherUsed is true, the currently used value will be gathered and added to the resulting RlimitStat.
 	RlimitUsageWithContext(ctx context.Context, gatherUsed bool) ([]process.RlimitStat, error)
+	CgroupWithContext(ctx context.Context) (string, error)
 }
 
 type gopsProcessHandles struct {
-	handles []*process.Process
+	handles []wrappedProcessHandle
 }
 
 func (p *gopsProcessHandles) Pid(index int) int32 {
-	return p.handles[index].Pid
+	return p.handles[index].Process.Pid
 }
 
 func (p *gopsProcessHandles) At(index int) processHandle {
@@ -109,13 +116,51 @@ func (p *gopsProcessHandles) Len() int {
 	return len(p.handles)
 }
 
+type wrappedProcessHandle struct {
+	*process.Process
+}
+
+func (p wrappedProcessHandle) CgroupWithContext(ctx context.Context) (string, error) {
+	pid := p.Process.Pid
+	statPath := getEnvWithContext(ctx, string(common.HostProcEnvKey), "/proc", strconv.Itoa(int(pid)), "cgroup")
+	contents, err := os.ReadFile(statPath)
+	if err != nil {
+		return "", err
+	}
+
+	return strings.TrimSuffix(string(contents), "\n"), nil
+}
+
+// copied from gopsutil:
+// GetEnvWithContext retrieves the environment variable key. If it does not exist it returns the default.
+// The context may optionally contain a map superseding os.EnvKey.
+func getEnvWithContext(ctx context.Context, key string, dfault string, combineWith ...string) string {
+	var value string
+	if env, ok := ctx.Value(common.EnvKey).(common.EnvMap); ok {
+		value = env[common.EnvKeyType(key)]
+	}
+	if value == "" {
+		value = os.Getenv(key)
+	}
+	if value == "" {
+		value = dfault
+	}
+	segments := append([]string{value}, combineWith...)
+
+	return filepath.Join(segments...)
+}
+
 func getProcessHandlesInternal(ctx context.Context) (processHandles, error) {
 	processes, err := process.ProcessesWithContext(ctx)
 	if err != nil {
 		return nil, err
 	}
+	wrapped := make([]wrappedProcessHandle, len(processes))
+	for i, p := range processes {
+		wrapped[i] = wrappedProcessHandle{Process: p}
+	}
 
-	return &gopsProcessHandles{handles: processes}, nil
+	return &gopsProcessHandles{handles: wrapped}, nil
 }
 
 func parentPid(ctx context.Context, handle processHandle, pid int32) (int32, error) {

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

@@ -203,8 +203,15 @@ func (s *scraper) getProcessMetadata() ([]*processMetadata, error) {
 			}
 			continue
 		}
+		cgroup, err := getProcessCgroup(ctx, handle)
+		if err != nil {
+			if !s.config.MuteProcessCgroupError {
+				errs.AddPartial(1, fmt.Errorf("error reading process cgroup for pid %v: %w", pid, err))
+			}
+			continue
+		}
 
-		executable := &executableMetadata{name: name, path: exe}
+		executable := &executableMetadata{name: name, path: exe, cgroup: cgroup}
 
 		// filter processes by name
 		if (s.includeFS != nil && !s.includeFS.Matches(executable.name)) ||

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

@@ -38,6 +38,10 @@ func getProcessName(ctx context.Context, proc processHandle, _ string) (string,
 	return name, nil
 }
 
+func getProcessCgroup(_ context.Context, _ processHandle) (string, error) {
+	return "", nil
+}
+
 func getProcessExecutable(ctx context.Context, proc processHandle) (string, error) {
 	cmdline, err := proc.CmdlineWithContext(ctx)
 	if err != nil {

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

@@ -46,6 +46,15 @@ func getProcessExecutable(ctx context.Context, proc processHandle) (string, erro
 	return exe, nil
 }
 
+func getProcessCgroup(ctx context.Context, proc processHandle) (string, error) {
+	cgroup, err := proc.CgroupWithContext(ctx)
+	if err != nil {
+		return "", err
+	}
+
+	return cgroup, nil
+}
+
 func getProcessCommand(ctx context.Context, proc processHandle) (*commandMetadata, error) {
 	cmdline, err := proc.CmdlineSliceWithContext(ctx)
 	if err != nil {

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

@@ -25,6 +25,10 @@ func getProcessName(context.Context, processHandle, string) (string, error) {
 	return "", nil
 }
 
+func getProcessCgroup(ctx context.Context, proc processHandle) (string, error) {
+	return "", nil
+}
+
 func getProcessExecutable(context.Context, processHandle) (string, error) {
 	return "", nil
 }