[signalfxexporter] Fix invalid and missing host metadata (#40218)

#### Description
Fix invalid and missing host metadata when the collector is deployed in
a containerized environment and the host root filesystem is mounted to
non-standard location.
This PR introduces the `root_path` config option which is used to create
gopsutil context using
https://github.com/open-telemetry/opentelemetry-collector-contrib/tree/main/internal/gopsutilenv

Ideally this should be merged after
#40213
To make sure gopsutilenv detects inconsistency if the root_path value
between the hostmetrics receiver and sfx exporter are different
#### Testing
Unit tests and manual deployment

#### Documentation
README updated

Signed-off-by: Dani Louca <[email protected]>

Author: dloucasfx

.chloggen/sfxexporterhostmetadata.yaml

@@ -0,0 +1,29 @@
+# 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. filelogreceiver)
+component: signalfxexporter
+
+# A brief description of the change.  Surround your text with quotes ("") if it needs to start with a backtick (`).
+note:  Fix invalid and missing host metadata when the collector is deployed in a containerized environment and |
+  the host root filesystem is mounted to non-standard location.
+
+# Mandatory: One or more tracking issues related to the change. You can use the PR number here if no issue exists.
+issues: [40218]
+
+# (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: Use the newly introduced `root_path` configuration option to specify the root path of the host filesystem.|
+  This is required when the host root filesystem is mounted to a non-standard location.
+
+# 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: []

cmd/opampsupervisor/go.mod

@@ -325,3 +325,5 @@ replace github.com/open-telemetry/opentelemetry-collector-contrib/exporter/stefe
 replace github.com/open-telemetry/opentelemetry-collector-contrib/receiver/stefreceiver => ../../receiver/stefreceiver
 
 replace github.com/open-telemetry/opentelemetry-collector-contrib/processor/deltatocumulativeprocessor => ../../processor/deltatocumulativeprocessor
+
+replace github.com/open-telemetry/opentelemetry-collector-contrib/internal/gopsutilenv => ../../internal/gopsutilenv

exporter/elasticsearchexporter/integrationtest/go.mod

@@ -348,3 +348,5 @@ replace github.com/open-telemetry/opentelemetry-collector-contrib/exporter/stefe
 replace github.com/open-telemetry/opentelemetry-collector-contrib/receiver/stefreceiver => ../../../receiver/stefreceiver
 
 replace github.com/open-telemetry/opentelemetry-collector-contrib/processor/deltatocumulativeprocessor => ../../../processor/deltatocumulativeprocessor
+
+replace github.com/open-telemetry/opentelemetry-collector-contrib/internal/gopsutilenv => ../../../internal/gopsutilenv

exporter/signalfxexporter/README.md

@@ -102,6 +102,12 @@ The following configuration options can also be configured:
   processor is enabled in the pipeline with one of the cloud provider detectors
   or environment variable detector setting a unique value to `host.name` attribute
   within your k8s cluster. And keep `override=true` in resourcedetection config.
+- `root_path`: Used by the host metadata to identify the root filesystem.
+  This is needed when running in a containerized environment and the host root
+  filesystem is not `/`. Example: if the root filesystem is mounted under `/hostfs`, set
+  `root_path` to `/hostfs`.
+  Note: all components using `root_path` must have the same value; this currently applies
+  to hostmetrics receiver and signalfx exporter.
 - `exclude_properties`: A list of property filters to limit dimension update content.
   Property filters can contain any number of the following fields, supporting (negated)
   string literals, re2 `/regex/`, and [glob](https://github.com/gobwas/glob) syntax values:

exporter/signalfxexporter/config.go

@@ -20,6 +20,7 @@ import (
 	"github.com/open-telemetry/opentelemetry-collector-contrib/exporter/signalfxexporter/internal/correlation"
 	"github.com/open-telemetry/opentelemetry-collector-contrib/exporter/signalfxexporter/internal/translation"
 	"github.com/open-telemetry/opentelemetry-collector-contrib/exporter/signalfxexporter/internal/translation/dpfilters"
+	"github.com/open-telemetry/opentelemetry-collector-contrib/internal/gopsutilenv"
 	"github.com/open-telemetry/opentelemetry-collector-contrib/internal/splunk"
 )
 
@@ -114,6 +115,9 @@ type Config struct {
 	//            And keep `override=true` in resourcedetection config.
 	SyncHostMetadata bool `mapstructure:"sync_host_metadata"`
 
+	// RootPath is the host's root directory used when syncing metadata; applies to linux only.
+	RootPath string `mapstructure:"root_path"`
+
 	// ExcludeMetrics defines dpfilter.MetricFilters that will determine metrics to be
 	// excluded from sending to SignalFx backend. If translations enabled with
 	// TranslationRules options, the exclusion will be applied on translated metrics.
@@ -224,6 +228,10 @@ func (cfg *Config) Validate() error {
 		return errors.New(`cannot have a negative "timeout"`)
 	}
 
+	if err := gopsutilenv.ValidateRootPath(cfg.RootPath); err != nil {
+		return fmt.Errorf("invalid root_path: %w", err)
+	}
+
 	return nil
 }
 

exporter/signalfxexporter/exporter.go

@@ -21,6 +21,7 @@ import (
 	"github.com/open-telemetry/opentelemetry-collector-contrib/exporter/signalfxexporter/internal/dimensions"
 	"github.com/open-telemetry/opentelemetry-collector-contrib/exporter/signalfxexporter/internal/hostmetadata"
 	"github.com/open-telemetry/opentelemetry-collector-contrib/exporter/signalfxexporter/internal/translation"
+	"github.com/open-telemetry/opentelemetry-collector-contrib/internal/gopsutilenv"
 	"github.com/open-telemetry/opentelemetry-collector-contrib/internal/splunk"
 	metadata "github.com/open-telemetry/opentelemetry-collector-contrib/pkg/experimentalmetricmetadata"
 )
@@ -157,7 +158,8 @@ func (se *signalfxExporter) start(ctx context.Context, host component.Host) (err
 
 	var hms *hostmetadata.Syncer
 	if se.config.SyncHostMetadata {
-		hms = hostmetadata.NewSyncer(se.logger, dimClient)
+		envMap := gopsutilenv.SetGoPsutilEnvVars(se.config.RootPath)
+		hms = hostmetadata.NewSyncer(se.logger, dimClient, envMap)
 	}
 	se.dimClient = dimClient
 	se.pushMetricsData = dpClient.pushMetricsData

exporter/signalfxexporter/go.mod

@@ -8,6 +8,7 @@ require (
 	github.com/gogo/protobuf v1.3.2
 	github.com/open-telemetry/opentelemetry-collector-contrib/internal/common v0.127.0
 	github.com/open-telemetry/opentelemetry-collector-contrib/internal/coreinternal v0.127.0
+	github.com/open-telemetry/opentelemetry-collector-contrib/internal/gopsutilenv v0.0.0-00010101000000-000000000000
 	github.com/open-telemetry/opentelemetry-collector-contrib/internal/splunk v0.127.0
 	github.com/open-telemetry/opentelemetry-collector-contrib/pkg/batchperresourceattr v0.127.0
 	github.com/open-telemetry/opentelemetry-collector-contrib/pkg/experimentalmetricmetadata v0.127.0
@@ -35,6 +36,7 @@ require (
 	go.uber.org/goleak v1.3.0
 	go.uber.org/multierr v1.11.0
 	go.uber.org/zap v1.27.0
+	golang.org/x/net v0.40.0
 	golang.org/x/sys v0.33.0
 	gopkg.in/yaml.v3 v3.0.1
 )
@@ -102,7 +104,6 @@ require (
 	go.opentelemetry.io/otel/sdk/metric v1.35.0 // indirect
 	go.opentelemetry.io/otel/trace v1.36.0 // indirect
 	golang.org/x/crypto v0.38.0 // indirect
-	golang.org/x/net v0.40.0 // indirect
 	golang.org/x/text v0.25.0 // indirect
 	google.golang.org/genproto/googleapis/rpc v0.0.0-20250218202821-56aae31c358a // indirect
 	google.golang.org/grpc v1.72.2 // indirect
@@ -136,3 +137,5 @@ replace github.com/open-telemetry/opentelemetry-collector-contrib/pkg/pdatautil
 replace github.com/go-openapi/spec v0.20.5 => github.com/go-openapi/spec v0.20.6
 
 replace github.com/open-telemetry/opentelemetry-collector-contrib/pkg/golden => ../../pkg/golden
+
+replace github.com/open-telemetry/opentelemetry-collector-contrib/internal/gopsutilenv => ../../internal/gopsutilenv

exporter/signalfxexporter/internal/hostmetadata/host.go

@@ -8,10 +8,6 @@ package hostmetadata // import "github.com/open-telemetry/opentelemetry-collecto
 
 import (
 	"context"
-	"errors"
-	"os"
-	"path/filepath"
-	"regexp"
 	"strconv"
 	"time"
 
@@ -20,23 +16,14 @@ import (
 	"github.com/shirou/gopsutil/v4/mem"
 )
 
-// etcPath is the path to host etc and can be set using the env var "HOST_ETC"
-// this is to maintain consistency with gopsutil
-var etcPath = func() string {
-	if etcPath := os.Getenv("HOST_ETC"); etcPath != "" {
-		return etcPath
-	}
-	return "/etc"
-}
-
 const cpuStatsTimeout = 10 * time.Second
 
 // Map library functions to unexported package variables for testing purposes.
 var (
 	cpuInfo          = cpu.InfoWithContext
 	cpuCounts        = cpu.CountsWithContext
-	memVirtualMemory = mem.VirtualMemory
-	hostInfo         = host.Info
+	memVirtualMemory = mem.VirtualMemoryWithContext
+	hostInfo         = host.InfoWithContext
 )
 
 // hostCPU information about the host
@@ -62,14 +49,14 @@ func (c *hostCPU) toStringMap() map[string]string {
 }
 
 // getCPU - adds information about the host cpu to the supplied map
-func getCPU() (info *hostCPU, err error) {
+func getCPU(ctx context.Context) (info *hostCPU, err error) {
 	info = &hostCPU{}
 
 	// get physical cpu stats
 	var cpus []cpu.InfoStat
 
 	// On Windows this can sometimes take longer than the default timeout (10 seconds).
-	ctx, cancel := context.WithTimeout(context.Background(), cpuStatsTimeout)
+	ctx, cancel := context.WithTimeout(ctx, cpuStatsTimeout)
 	defer cancel()
 
 	cpus, err = cpuInfo(ctx)
@@ -119,9 +106,9 @@ func (o *hostOS) toStringMap() map[string]string {
 }
 
 // getOS returns a struct with information about the host os
-func getOS() (info *hostOS, err error) {
+func getOS(ctx context.Context) (info *hostOS, err error) {
 	info = &hostOS{}
-	hInfo, err := hostInfo()
+	hInfo, err := hostInfo(ctx)
 	if err != nil {
 		return info, err
 	}
@@ -130,31 +117,10 @@ func getOS() (info *hostOS, err error) {
 	info.HostKernelName = hInfo.OS
 	// in gopsutil KernelVersion returns what we would expect for Kernel Release
 	info.HostKernelRelease = hInfo.KernelVersion
-	err = fillPlatformSpecificOSData(info)
+	err = fillPlatformSpecificOSData(ctx, info)
 	return info, err
 }
 
-// getLinuxVersion - adds information about the host linux version to the supplied map
-func getLinuxVersion() (string, error) {
-	etc := etcPath()
-	if value, err := getStringFromFile(`DISTRIB_DESCRIPTION="(.*)"`, filepath.Join(etc, "lsb-release")); err == nil {
-		return value, nil
-	}
-	if value, err := getStringFromFile(`PRETTY_NAME="(.*)"`, filepath.Join(etc, "os-release")); err == nil {
-		return value, nil
-	}
-	if value, err := os.ReadFile(filepath.Join(etc, "centos-release")); err == nil {
-		return string(value), nil
-	}
-	if value, err := os.ReadFile(filepath.Join(etc, "redhat-release")); err == nil {
-		return string(value), nil
-	}
-	if value, err := os.ReadFile(filepath.Join(etc, "system-release")); err == nil {
-		return string(value), nil
-	}
-	return "", errors.New("unable to find linux version")
-}
-
 // Memory stores memory collected from the host
 type Memory struct {
 	Total int
@@ -169,27 +135,15 @@ func (m *Memory) toStringMap() map[string]string {
 }
 
 // getMemory returns the amount of memory on the host as datatype.USize
-func getMemory() (*Memory, error) {
+func getMemory(ctx context.Context) (*Memory, error) {
 	m := &Memory{}
-	memoryStat, err := memVirtualMemory()
+	memoryStat, err := memVirtualMemory(ctx)
 	if err == nil {
 		m.Total = int(memoryStat.Total)
 	}
 	return m, err
 }
 
-func getStringFromFile(pattern string, path string) (string, error) {
-	var err error
-	var file []byte
-	reg := regexp.MustCompile(pattern)
-	if file, err = os.ReadFile(path); err == nil {
-		if match := reg.FindSubmatch(file); len(match) > 1 {
-			return string(match[1]), nil
-		}
-	}
-	return "", err
-}
-
 func bytesToKilobytes(b int) int {
 	return b / 1024
 }

exporter/signalfxexporter/internal/hostmetadata/host_linux.go

@@ -10,15 +10,22 @@ package hostmetadata // import "github.com/open-telemetry/opentelemetry-collecto
 
 import (
 	"bytes"
+	"errors"
+	"os"
+	"path/filepath"
+	"regexp"
 
+	"golang.org/x/net/context"
 	"golang.org/x/sys/unix"
+
+	"github.com/open-telemetry/opentelemetry-collector-contrib/internal/gopsutilenv"
 )
 
 // syscallUname maps to the golib system call, but can be modified for testing
 var syscallUname = unix.Uname
 
-func fillPlatformSpecificOSData(info *hostOS) error {
-	info.HostLinuxVersion, _ = getLinuxVersion()
+func fillPlatformSpecificOSData(ctx context.Context, info *hostOS) error {
+	info.HostLinuxVersion, _ = getLinuxVersion(ctx)
 
 	uname := &unix.Utsname{}
 	if err := syscallUname(uname); err != nil {
@@ -43,3 +50,36 @@ func fillPlatformSpecificCPUData(info *hostCPU) error {
 	info.HostProcessor = info.HostMachine
 	return nil
 }
+
+// getLinuxVersion - adds information about the host linux version to the supplied map
+func getLinuxVersion(ctx context.Context) (string, error) {
+	etc := gopsutilenv.GetEnvWithContext(ctx, "HOST_ETC", "/etc")
+	if value, err := getStringFromFile(`DISTRIB_DESCRIPTION="(.*)"`, filepath.Join(etc, "lsb-release")); err == nil {
+		return value, nil
+	}
+	if value, err := getStringFromFile(`PRETTY_NAME="(.*)"`, filepath.Join(etc, "os-release")); err == nil {
+		return value, nil
+	}
+	if value, err := os.ReadFile(filepath.Join(etc, "centos-release")); err == nil {
+		return string(value), nil
+	}
+	if value, err := os.ReadFile(filepath.Join(etc, "redhat-release")); err == nil {
+		return string(value), nil
+	}
+	if value, err := os.ReadFile(filepath.Join(etc, "system-release")); err == nil {
+		return string(value), nil
+	}
+	return "", errors.New("unable to find linux version")
+}
+
+func getStringFromFile(pattern string, path string) (string, error) {
+	var err error
+	var file []byte
+	reg := regexp.MustCompile(pattern)
+	if file, err = os.ReadFile(path); err == nil {
+		if match := reg.FindSubmatch(file); len(match) > 1 {
+			return string(match[1]), nil
+		}
+	}
+	return "", err
+}