Home

yacron2 Wiki

yacron2 is a cron replacement built on asyncio that runs natively on Linux, macOS, and Windows. Its "crontab" is written in YAML (classic Vixie crontabs are accepted as-is too), so jobs, schedules, and behavior are all declared in configuration: it reports job failures by email, Sentry, shell command, or HTTP webhook (Slack/Discord/Teams/ntfy-compatible); retries failing jobs with exponential backoff; emits job metrics to statsd and serves them natively to Prometheus; and can expose an optional HTTP control API and a built-in web dashboard to watch, run, cancel, and tail jobs live. When one instance is not enough, opt-in clustering and leader election lets several replicas run one config without double-running jobs, coordinated by mTLS gossip or fenced through a Kubernetes or etcd lease. It runs in the foreground, logs to stdout/stderr, and supports arbitrary timezones, which suits Docker, Kubernetes, and 12-factor deployments. yacron2 is a fork of gjcarneiro/yacron (by Gustavo Carneiro), continuing development from version 0.19.

Contents

Getting Started

• Installation: Install via Docker, pip, pipx, Homebrew, or the self-contained binary.
• Command-Line Reference: The yacron2 command, its flags, and config file/directory loading.
• Production and Container Deployment: Running hardened under non-root, read-only-root-filesystem Kubernetes/Docker.
• Running on Windows: Installing and running yacron2 natively on Windows: config path, default shell, Ctrl-C shutdown, and unsupported features.

Configuration

• Configuration Reference: The full YAML schema: top-level sections and per-job options.
• Classic Crontabs: Running plain Vixie-style crontab files as-is, how entries map onto yacron2's standard defaults, and the documented deviations from cron.
• Schedules and Timezones: Crontab strings, schedule objects, @reboot, UTC vs. local, and arbitrary timezones.
• Commands and Environment: Shell vs. argv commands, environment variables, env files, and per-job user/group.
• Output Capturing: Capturing stdout/stderr and customizing stream prefixes.
• Includes, Defaults, and Multi-File Config: Sharing settings via defaults, the include directive, and multi-file config directories.
• Logging Configuration: Customizing yacron2's own logging via the logging section.

Job Behavior

• Concurrency and Timeouts: concurrencyPolicy, executionTimeout, and killTimeout.
• Failure Detection and Retries: failsWhen rules, retry with exponential backoff, and onPermanentFailure.
• Clustering and Leader Election: the cluster section: mTLS peer attestation, quorum-gated leader election, per-job clusterPolicy, and a pluggable backend (gossip / kubernetes / etcd) for best-effort or fenced exactly-once election.

Integrations

• Reporting (Mail, Sentry, Shell, Webhook): onFailure/onSuccess reporting via email, Sentry, shell, and webhook (Slack-compatible), with jinja2 templating.
• Metrics with statsd: Emitting start/stop/success/duration metrics over UDP to statsd.
• Metrics with Prometheus: The /metrics endpoint on the web API, with job, scheduler, and cluster metrics in Prometheus or OpenMetrics format.
• HTTP Control API: The optional REST interface for status and on-demand job starts.
• Web Dashboard: The built-in browser dashboard: live status, live log tailing, run history, and timezone-aware schedule previews.

Reference and Development

• Architecture and Internals: How the asyncio scheduler, jobs, and reporters fit together.
• Contributing and Releasing: Development setup, the test/lint/type-check workflow, and the release process.
• Migration from yacron: Moving from gjcarneiro/yacron to yacron2.
• Troubleshooting and FAQ: Common problems, errors, and answers.