prometheus/node_exporter · error

failed to open procfs

Error message

failed to open procfs: %w

What it means

NewProcessStatCollector initializes a procfs filesystem handle via procfs.NewFS(*procPath) and fails the collector constructor when that call errors. This means the proc filesystem could not be opened at the configured path (default /proc), so the process collector cannot be created at all.

Solutions

1. Check that /proc (or the value of --path.procfs) exists and is a mounted proc filesystem: mount | grep proc.
2. Fix the --path.procfs flag if it was set incorrectly; the default should be /proc.
3. In containers, run with /proc mounted (default on Docker/Kubernetes) and avoid masking the whole proc mount.
4. When running tests, unpack the fixtures (make test unpacks collector/fixtures/sys.ttar) and point --path.procfs at collector/fixtures/sys.

Example fix

// before
node_exporter --path.procfs=/prco
// after
node_exporter --path.procfs=/proc

// before launching the exporter / constructing the collector
func procfsAvailable(path string) error {
    var st syscall.Statfs_t
    if err := syscall.Statfs(path, &st); err != nil {
        return err
    }
    if st.Type != 0x9fa0 { // PROC_SUPER_MAGIC
        return fmt.Errorf("%s is not a procfs mount", path)
    }
    return nil
}

collector, err := NewProcessStatCollector(logger)
if err != nil {
    log.Fatalf("process collector unavailable: %v (is %s a mounted procfs?)", err, *procPath)
}

When it happens

Trigger: procfs.NewFS is called with the --path.procfs flag value and returns an error, e.g. when the procfs mount point does not exist or statfs on it fails during collector construction at exporter startup.

Common situations: Running node_exporter on non-Linux setups or in containers where /proc is not mounted; passing --path.procfs pointing to a extracted fixture directory that was deleted; typos in --path.procfs such as --path.procfs=/prco.

Understand the failure class

Background: "open() failed", "failed to open file", "cannot create file" — what a file open error means and how to fix it — this error's family across 42 libraries.

Related errors

• failed to open procfs
• failed to open procfs
• failed to open procfs
• failed to open procfs
• failed to retrieve pressure stats

AI-assisted analysis of prometheus/node_exporter@17ddd77c59 (2026-09-07). Data as JSON: /api/errors/abe57e27fe6afda9. Report an issue: GitHub.

Appendix: source

Thrown at collector/processes_linux.go:51

	fs           procfs.FS
	threadAlloc  *prometheus.Desc
	threadLimit  *prometheus.Desc
	threadsState *prometheus.Desc
	procsState   *prometheus.Desc
	pidUsed      *prometheus.Desc
	pidMax       *prometheus.Desc
	logger       *slog.Logger
}

func init() {
	registerCollector("processes", defaultDisabled, NewProcessStatCollector)
}

// NewProcessStatCollector returns a new Collector exposing process data read from the proc filesystem.
func NewProcessStatCollector(logger *slog.Logger) (Collector, error) {
	fs, err := procfs.NewFS(*procPath)
	if err != nil {
		return nil, fmt.Errorf("failed to open procfs: %w", err)
	}
	subsystem := "processes"
	return &processCollector{
		fs: fs,
		threadAlloc: prometheus.NewDesc(
			prometheus.BuildFQName(namespace, subsystem, "threads"),
			"Allocated threads in system",
			nil, nil,
		),
		threadLimit: prometheus.NewDesc(
			prometheus.BuildFQName(namespace, subsystem, "max_threads"),
			"Limit of threads in the system",
			nil, nil,
		),
		threadsState: prometheus.NewDesc(
			prometheus.BuildFQName(namespace, subsystem, "threads_state"),
			"Number of threads in each state.",
			[]string{"thread_state"}, nil,

View on GitHub (pinned to 17ddd77c59)