1. Write a clear message to logs: "Gateway failed to start 3 times in 60s — possible config error. Check ~/.openclaw/openclaw.json. Run 'openclaw doctor' to diagnose." Option A (application-level): On startup, check a restart-sentinel file (already exists: src/infra/restart-sentinel.ts). If 3+ restarts within 60s, write error and exit with a distinct code (e.g. 78 = EX_CONFIG) that systemd's RestartPreventExitStatus can catch.

TL;DR

Implement a restart sentinel mechanism and configure systemd to prevent excessive restarts to address the gateway crash loop issue.

Guidance

• Implement Option A (application-level) by modifying the src/infra/restart-sentinel.ts file to check for 3+ restarts within 60s and exit with a distinct code (e.g., 78 = EX_CONFIG) that systemd's RestartPreventExitStatus can catch.
• Configure systemd to prevent excessive restarts by adding StartLimitBurst=3 and StartLimitIntervalSec=60 to the generated systemd unit file (src/daemon/node-service.ts).
• Verify that the gateway exits cleanly after 3 consecutive startup failures within 60 seconds and writes a clear error message to the logs.
• Test the implementation by intentionally introducing a config error and checking the gateway's behavior.

Example

// src/infra/restart-sentinel.ts
import fs from 'fs';

const restartSentinelFile = 'restart-sentinel.txt';
const maxRestarts = 3;
const restartInterval = 60 * 1000; // 60 seconds

let restartCount = 0;
let lastRestartTime = 0;

export function checkRestartSentinel() {
  if (fs.existsSync(restartSentinelFile)) {
    const stats = fs.statSync(restartSentinelFile);
    lastRestartTime = stats.mtimeMs;
    restartCount++;
  } else {
    fs.writeFileSync(restartSentinelFile, '');
    lastRestartTime = Date.now();
    restartCount = 1;
  }

  if (restartCount >= maxRestarts && Date.now() - lastRestartTime <= restartInterval) {
    console.error('Gateway failed to start 3 times in 60s — possible config error. Check ~/.openclaw/openclaw.json. Run \'openclaw doctor\' to diagnose.');
    process.exit(78); // EX_CONFIG
  }
}

Notes

The provided solution assumes that the restart-sentinel.ts file is already implemented and only needs to be modified to include the restart count and interval checks. Additionally, the systemd configuration changes require modifying the node-service.ts file to include the StartLimitBurst and StartLimitIntervalSec options.

Recommendation

Apply both Option A (application-level) and Option B (systemd unit) to ensure that the gateway exits cleanly after 3 consecutive startup failures within 60 seconds and provides user-visible feedback. This approach provides a safety net to prevent excessive restarts and helps diagnose config errors.