What causes "procinfo: too few fields in stat for pid %d" in wavetermdev/waveterm?

Retry the read once — transient truncation during process teardown is possible

procinfo: too few fields in stat for pid %d

Error message

procinfo: too few fields in stat for pid %d

What it means

After stripping the comm field, /proc/<pid>/stat must contain at least 22 whitespace-separated fields for the parser to extract ppid, utime, stime, num_threads, and rss (it needs rest[0..21]). Fewer fields means the stat record was truncated or structurally different from what the kernel provides.

Source

Thrown at pkg/util/procinfo/procinfo_linux.go:85

	if lp < 0 || rp < 0 || rp <= lp {
		return nil, fmt.Errorf("procinfo: malformed stat for pid %d", pid)
	}

	pidStr := strings.TrimSpace(s[:lp])
	comm := s[lp+1 : rp]
	rest := strings.Fields(s[rp+1:])

	// rest[0] = field 3 (state), rest[1] = field 4 (ppid), ...
	// Fields after comm are numbered starting at 3, so rest[i] = field (i+3).
	// We need:
	//   rest[0]  = field  3  state
	//   rest[1]  = field  4  ppid
	//   rest[11] = field 14  utime
	//   rest[12] = field 15  stime
	//   rest[17] = field 20  num_threads
	//   rest[21] = field 24  rss (pages)
	if len(rest) < 22 {
		return nil, fmt.Errorf("procinfo: too few fields in stat for pid %d", pid)
	}

	parsedPid, err := strconv.ParseInt(pidStr, 10, 32)
	if err != nil {
		return nil, fmt.Errorf("procinfo: parse pid: %w", err)
	}

	statusChar := rest[0]
	status, ok := LinuxStatStatus[statusChar]
	if !ok {
		status = "unknown"
	}

	info := &ProcInfo{
		Pid:        int32(parsedPid),
		Command:    comm,
		Status:     status,
		CpuUser:    -1,

View on GitHub (pinned to a4447c1563)

Solutions

Retry the read once — transient truncation during process teardown is possible
Verify /proc/<pid>/stat has ~52 fields (cat /proc/<pid>/stat | wc -w)
If on a custom/embedded kernel, confirm its procfs stat layout matches mainline field ordering
Fix test fixtures to include at least 24 stat fields

Example fix

// before
if len(rest) < 22 {
    return nil, fmt.Errorf("procinfo: too few fields in stat for pid %d", pid)
}
// after: parse what is available instead of failing
if len(rest) < 22 {
    return info, nil // partial info; fields default to -1
}

Defensive patterns

Strategy: retry

Validate before calling

// verify field count before dependent processing
data, _ := os.ReadFile(fmt.Sprintf("/proc/%d/stat", pid))
if strings.Count(string(data), " ") < 24 {
    // truncated/insufficient stat; avoid processing
}

Try / catch

_, err := procinfo.GetProcInfo(ctx, nil, pid)
if err != nil && strings.Contains(err.Error(), "too few fields") {
    // retry after a short delay; persistent failure means kernel/fixture problem
}

Prevention

Use mainline Linux kernels or verify procfs layout on custom kernels
Include >= 24 fields in test stat fixtures
Retry transient truncations instead of failing the poll cycle
Sanity-check 'wc -w /proc/<pid>/stat' (~52) in deployment diagnostics

When it happens

Trigger: A truncated read of a stat file during process exit; reading a fake/fixture stat file with too few columns; an exotic kernel whose stat format omits trailing fields.

Common situations: Custom kernels or stripped-down embedded Linux builds with divergent procfs; unit-test fixtures missing fields; corrupted procfs output from virtualization.

Related errors

AI-assisted analysis of wavetermdev/waveterm@a4447c1563 (2026-09-01). Data as JSON: /api/errors/352b143b2f2d2439. Report an issue: GitHub.