projectdiscovery/nuclei · error

max sleep count

Error message

max sleep count

What it means

instance.go installs maxBackoffSleeper(10) as the go-rod browser sleeper; every internal rod operation that waits (connecting to the browser, evaluating CDP calls) goes through it. After 10 sleeps (exponential 100ms-500ms, i.e. a few seconds total) the sleeper returns 'max sleep count', aborting whatever browser operation was pending. It almost always means Chrome started but its DevTools endpoint never became usable.

Source

Thrown at pkg/protocols/headless/engine/instance.go:70

	return i.engine.Close()
}

// SetInteractsh client
func (i *Instance) SetInteractsh(interactsh *interactsh.Client) {
	i.interactsh = interactsh
}

// maxBackoffSleeper is a backoff sleeper respecting max backoff values
func maxBackoffSleeper(max int) utils.Sleeper {
	count := 0
	backoffSleeper := utils.BackoffSleeper(100*time.Millisecond, 500*time.Millisecond, nil)

	return func(ctx context.Context) error {
		if ctx.Err() != nil {
			return ctx.Err()
		}
		if count == max {
			return errors.New("max sleep count")
		}
		count++
		return backoffSleeper(ctx)
	}
}

View on GitHub (pinned to 265b3a3dec)

Solutions

1. Kill stale chrome/chromium processes and retry the scan
2. Give the container more shared memory and CPU (docker run --shm-size=1g, raise memory limits)
3. Run fewer concurrent headless templates (-concurrency-limit, -headless-concurrency) so Chrome gets resources
4. Use -system-chrome with a known-good installed browser as an alternative to the downloaded Chromium

Example fix

# before
docker run --rm -it nuclei:v3 -tags headless -u file.txt
# after
docker run --rm -it --shm-size=1g --memory=2g nuclei:v3 -tags headless -u file.txt

pkill -f 'chrome.*remote-debugging-port' 2>/dev/null || true  # clear zombie browsers before a headless run

var berr *rod.ErrBrowser // or string-match on 'max sleep count'
if err != nil && strings.Contains(err.Error(), "max sleep count") {
    // browser stall: back off, kill stale chrome, retry once
    time.Sleep(5 * time.Second)
    _ = exec.Command("pkill", "-f", "remote-debugging-port").Run()
    return retryScan()
}

When it happens

Trigger: Chrome process launches but stalls: starved CI containers, tiny /dev/shm, low file-descriptor limits, or a stale/zombie chrome holding the debug port. Any rod call that keeps failing fast burns one of the 10 backoff slots.

Common situations: Docker with default 64MB /dev/shm running headless Chrome; low-memory GitHub Actions runners; leftover chrome processes from a previous crashed run; concurrent nuclei instances fighting for CPU.

Related errors

• js must be at least 1
• the chrome browser is not installed
• payload concurrency must be at least 1
• Invalid action type: %s
• Element did not appear in the given amount of time

AI-assisted analysis of projectdiscovery/nuclei@265b3a3dec (2026-08-15). Data as JSON: /api/errors/ed2ddf73b209a059. Report an issue: GitHub.