Skip to main content

Log Attachment

When a job fails, you can attach its stdout/stderr output directly to the run record. The log is stored in Crontify and surfaced in the run detail view — no need to dig through your server logs or a separate logging service to understand what went wrong.

How it works​

Pass a log string when calling monitor.fail() or sending a /fail HTTP ping. Crontify stores the log separately from the run record — it is never fetched in list queries and never bloats the monitor detail page. It's only loaded when you open a specific run.

SDK usage​

try {
  await doWork();
  await monitor.success();
} catch (error) {
  await monitor.fail({
    message: error.message,
    log: error.stack,   // Full stack trace attached to the run
  });
}

Automatic capture with wrap()​

When using wrap(), stack traces are captured automatically on error — you don't need to pass log manually:

await monitor.wrap(async () => {
  await doWork();
});
// If doWork() throws, error.stack is automatically attached as the log

Attaching richer output​

For more detailed diagnostics, you can capture and attach full process output:

import { execSync } from 'child_process';

try {
  const output = execSync('python3 job.py', { encoding: 'utf-8' });
  await monitor.success();
} catch (error) {
  await monitor.fail({
    message: 'Python job failed',
    log: error.stdout + error.stderr,  // Both streams
  });
}

Character limit​

Logs are truncated to 10,000 characters server-side. This is enforced as a hard limit regardless of what you send. If your output is longer, the first 10,000 characters are kept and the rest is discarded.

The SDK also applies this limit client-side before sending.

Viewing logs in the dashboard​

On the monitor detail page, click any failed run to open the run detail panel. If the run has a log attached, it appears in a collapsible code block below the error message. Logs longer than 400 characters are collapsed by default with an Expand button.

Log snippets in alerts​

When a fail alert fires for a run that has a log, the first 500 characters of the log are included in the alert as a snippet. Slack and Discord show it in a code block. Webhooks include it as a logSnippet field in the JSON payload.

The full log is always available on the dashboard — the snippet is just for quick context in the notification.

HTTP usage​

curl -X POST https://api.crontify.com/api/v1/ping/$MONITOR_ID/fail \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "errorMessage": "Job failed with exit code 1",
    "logOutput": "Traceback (most recent call last):\n  File \"job.py\", line 42, in <module>\n    ..."
  }'

Retrieving logs via API​

GET /api/v1/monitors/:monitorId/runs/:runId/log
Authorization: Bearer <jwt>
{
  "hasLog": true,
  "content": "Error: Connection refused\n    at ..."
}

If no log was attached, hasLog is false and content is null.