Cron Jobs
Cron jobs run scheduled commands inside your application's environment — for example clearing caches, sending digests, or pruning data. You define them in .tragwerk/config.xml and Tragwerk runs them on the server in a dedicated scheduler container.
Defining cron jobs
Cron jobs live in the XML <crons> block. Each job has a name, a schedule (standard cron expression), and a command. See the configuration reference: Crons configuration.
<crons>
<cron name="cache-clear" schedule="0 * * * *" command="bin/console cache:clear" />
<cron name="digest" schedule="0 6 * * *" command="bin/console app:send-digest" />
</crons>How it runs on the server
For each application Tragwerk starts a supercronic sidecar container named {app}-cron. Supercronic is a cron implementation built for containers: it reads the generated crontab, runs each command on schedule inside the app's runtime, and logs structured (JSON) output.
INFO
Supercronic is used instead of the system cron daemon because it logs to stdout in a machine-readable form, runs as a non-root container process, and needs no MTA — which is exactly what Tragwerk needs to capture run history.
Run history and live logs
The Cron Jobs view shows three things:
- Definitions — the jobs declared in your config (name, schedule, command).
- Run history — the last execution time and result (success/failure) per job.
- Live logs — output streamed as a job runs.
Behind the scenes Tragwerk parses the {app}-cron container's supercronic logs, persists each run (start time, finish time, success flag, captured output), and streams updates so the UI reflects runs live. A run shows as in-progress until its finishing log line is ingested. This happens automatically.
Example
A job that runs every five minutes:
<crons>
<cron name="heartbeat" schedule="*/5 * * * *" command="bin/console app:heartbeat" />
</crons>After deployment, open Cron Jobs, find heartbeat, and watch its run history populate every five minutes with the last result and log output.