GitHub Actions Scheduled Workflow Monitoring

The problem

GitHub Actions scheduled workflows fail quietly. The `schedule` trigger is best-effort, not guaranteed: GitHub delays cron-triggered runs under load and disables schedules entirely after 60 days of repository inactivity, so a "nightly" workflow can simply stop firing with no error and no notification. Even when a run does start, a green checkmark only tells you the workflow completed — it says nothing about whether the run was hours late, ran far longer than usual, or quietly produced no output. And if the workflow never triggers at all, there is no failed run to alert on, because there is no run. Teams discover the gap only when the backup, migration, or sync that the workflow was supposed to perform turns out to be days stale.

How Pakyas helps

Pakyas uses execution-signal monitoring: your workflow proves it ran by sending a ping, and Pakyas compares those signals against the schedule you expect. If a scheduled run is skipped or GitHub silently disables the cron, no ping arrives and Pakyas marks the check Missing. If the run starts hours after its expected window, it is flagged Late. If it runs far longer than its normal duration, it is Overrunning. And if the job sends a failure signal or a non-zero exit, it is recorded as an Error — distinct from a run that simply never happened. Because the signal comes from inside the job rather than from GitHub's status, Pakyas catches the exact failure mode GitHub Actions hides: the scheduled workflow that never fired. Alerts route to email, Slack, Telegram, or webhooks the moment a check leaves On Schedule.

Set it up

1. 1

   Create a check and grab its ping URL

   # In the Pakyas dashboard, create a check for your scheduled workflow
   # and set its expected schedule (e.g. cron 0 3 * * *).
   # Copy the check's ping URL — it looks like:
   #   https://ping.pakyas.com/<your-check-id>
   # where <your-check-id> is the check's public_id (a UUID).

   The ping URL has no /ping/ prefix. Success is the bare URL; append /start to signal a run beginning and /fail to signal failure.

2. 2

   Store the ping URL as a repository secret

   # GitHub repo → Settings → Secrets and variables → Actions → New repository secret
   # Name:  PAKYAS_PING_URL
   # Value: https://ping.pakyas.com/<your-check-id>

   Storing the full URL as a secret keeps the check ID out of your workflow YAML and logs.

3. 3

   Wrap the scheduled job with start, success, and fail pings

   name: Nightly Backup
   
   on:
     schedule:
       - cron: '0 3 * * *'   # 03:00 UTC daily
     workflow_dispatch:        # allow manual runs
   
   jobs:
     backup:
       runs-on: ubuntu-latest
       env:
         PAKYAS_PING_URL: ${{ secrets.PAKYAS_PING_URL }}
       steps:
         - uses: actions/checkout@v4
   
         - name: Signal start to Pakyas
           run: curl -fsS "$PAKYAS_PING_URL/start"
   
         - name: Run backup
           run: ./scripts/backup.sh
   
         - name: Signal success to Pakyas
           if: success()
           run: curl -fsS "$PAKYAS_PING_URL"
   
         - name: Signal failure to Pakyas
           if: failure()
           run: curl -fsS "$PAKYAS_PING_URL/fail"

   The if: success() / if: failure() pair guarantees Pakyas always receives a terminal signal. If the scheduled run never fires at all, no ping arrives and Pakyas reports the check as Missing.

Pricing

Pakyas has four tiers: Free ($0, up to 10 checks), Developer ($9/mo), Pro ($29/mo), and Business ($99/mo). Most teams can start monitoring their scheduled GitHub Actions workflows on the free tier and upgrade as the number of monitored jobs grows.

See the full breakdown on the pricing page.