About this project
cronstable is a cron replacement designed for both single machines and clusters. It reads schedules from YAML or classic crontab files and runs commands on cron-style schedules, with support for second-level scheduling, time zones, business-day and hashed schedules, schedule introspection, and linting.
Beyond basic scheduling, it provides failure handling with retries and exponential backoff, result verification, late-run detection, and alerts through Sentry, email, Slack-compatible webhooks, shell commands, Windows Event Log, and end-to-end encrypted push notifications to its iOS app. Durable state can preserve run history and pending retries across restarts, and workflows can be defined as directed acyclic graphs with dependencies, fan-out, sensors, and approval gates.
Observability is available through a built-in web dashboard, a terminal TUI, and an iOS app. These show job status, live logs, run history, workflow graphs, cluster health, and resource usage. An HTTP API allows reading status and controlling jobs, and metrics can be exported to Prometheus or statsd. Optional clustering uses leader election through Kubernetes Leases, etcd, a shared filesystem, or mutual TLS.
Deployment options include Docker images in multiple distribution variants, standalone binaries for Linux, macOS, Windows, FreeBSD, OpenBSD, NetBSD, and illumos, plus Homebrew, WinGet, pip, and system packages. The project targets many CPU architectures and supports restricted container environments with non-root operation and read-only root filesystems.
Comments
0 people shared their preference · Deer Point appears after 10 participants
Sign in to join the discussion.