Integrations
RunVouch wraps a command or takes two HTTP calls, so it works wherever a job runs. Pick the runtime; each page lists where the job runs there, how to store the key, the snippet, and what fails silently on that platform in its own terms.
Guides with their own page
Claude Code
Routines, headless claude -p, hooks plugin.
Cron and scripts
rv run for anything on a schedule.
GitHub Actions
Scheduled workflows that stop firing.
OpenClaw
Heartbeats, per-task runs, daily caps.
n8n
Two HTTP nodes; Error Workflows miss the rest.
Python and Node
Single-file clients, any script.
Schedulers
Kubernetes CronJob
A CronJob that reports Succeeded tells you the pod exited 0.
systemd timers
A timer that fires a failing service shows up in systemctl --failed and nowhere else.
GitLab CI scheduled pipelines
Pipeline schedules run as the user who created them.
Jenkins
Jenkins tells you about red builds.
Heroku Scheduler
Heroku's own documentation calls Scheduler a best-effort service: a run can be skipped without a trace.
Cloud functions
AWS Lambda + EventBridge Scheduler
CloudWatch can alarm on errors and on invocation count, if you build those alarms.
Google Cloud Scheduler
Cloud Scheduler counts an HTTP 2xx as success and retries on anything else.
Azure Functions timer trigger
A timer-triggered function that runs, catches its own exception and returns is a success in Application Insights.
Cloudflare Workers Cron Triggers
A Worker's scheduled() handler has no output you can look at later, no exit code and, on the free plan, a C.
Vercel Cron Jobs
A cron entry in vercel.json hits a route on a schedule and logs the status.
Supabase (pg_cron + Edge Functions)
The documented pattern is cron.schedule() calling net.http_post() to an Edge Function.
Platforms
Render Cron Jobs
Render shows each cron run's log and exit status in the dashboard.
Railway cron schedules
A Railway service with a cron schedule starts, runs its start command and exits.
Fly.io scheduled Machines
On Fly you either run a Machine with a schedule (fly machine run --schedule daily) or run cron (often super.
Cron inside Docker (supercronic, crond)
Whether you use supercronic, BusyBox crond or ofelia, a scheduled command inside a container has the same two silent fai.
Orchestrators
Apache Airflow
Airflow is very good at telling you a task failed.
Prefect
A Prefect deployment with a schedule creates flow runs; a worker on the right work pool executes them.
Dagster
Schedules and sensors in Dagster are executed by the dagster-daemon.
Celery beat
Celery beat is a single process that publishes tasks on a schedule.
APScheduler (in-process)
APScheduler, node-cron and node-schedule run inside your application process.
Temporal Schedules
A Temporal Schedule starts workflows on time; a worker polling the task queue executes them.
Windmill
Windmill has error handlers and a runs page.
No-code automation
Make (Integromat)
A Make scenario on a schedule runs until Make deactivates it after consecutive errors, or until someone toggles it off.
Zapier
A Zap with a Schedule by Zapier trigger runs until Zapier turns it off (repeated errors, a disconnected app, a plan limi.
Home Assistant automations
An automation with a time trigger that runs a shell_command or a conversation agent keeps working until an .
Agent frameworks
LangGraph
LangGraph gives you cycles by design.
CrewAI
A crew has max_iter per agent and max_rpm; neither is a budget.
AutoGen
Two AutoGen agents that keep agreeing to continue, or a tool that keeps failing and being retried by the assistant, look.
OpenAI Agents SDK
The Agents SDK stops a run at max_turns and raises.
Pydantic AI and smolagents
These libraries are a few hundred lines you call from a script.
Dify and Flowise
Both tools expose a workflow or chatflow over HTTP; the schedule lives outside them, in cron, n8n or a cloud scheduler.
Ollama and local LLM batch jobs
With a local model the failure modes change: the process hangs while the model loads, the GPU is taken by another job, o.