Scheduled GitHub Actions workflow never triggers in a new private repository #213230
Replies: 4 comments 4 replies
|
Here is what is happening and how to fix it: Why this happens
How to Fix & Verify
Hope this helps! If this resolves the issue for you, please feel free to mark this as the accepted answer so it helps others running into the same scheduler delay. |
|
Thank you for your detailed response. There is one important constraint in our setup. Our production workflow intentionally uses only the For this reason, adding The workflow exists on the default Could you clarify two points?
We would prefer to investigate without changing the production workflow or weakening its OIDC security controls. Thank you again for your help! |
|
Hi @niltonjn2022-commits — answering your two follow-ups directly, since the OIDC constraint rules out the usual fix: 1. No, I couldn't find that in the official docs. I re-read the Two documented points that do matter for your case: your repo is private, so the often-quoted "scheduled workflows are disabled after 60 days without activity" rule doesn't apply to you — the docs limit that rule to public repositories. And your two missed runs were the first two scheduled chances ever for a brand-new repo (Nov 15, 23:37 and Nov 16, 00:07 UTC), at minutes 07 and 37. A first run that never appears is within the best-effort behaviour the docs describe, not proof the schedule was never registered. 2. Yes, a minimal schedule-only diagnostic workflow is the right isolation test, and it doesn't touch your OIDC setup. I use OIDC-based trusted publishing myself (PyPI, for a package I maintain), so I understand why you don't want to add name: schedule-probe
on:
schedule:
- cron: '11,41 * * * *'
jobs:
probe:
runs-on: ubuntu-latest
steps:
- run: echo "schedule fired at $(date -u)"Leave it enabled and give it a few cycles before judging it — one missed run proves nothing, since scheduled runs can be dropped silently under load. If the probe fires and the production workflow doesn't, the scheduler is fine and the problem is specific to the production workflow file. If the probe never fires either, it's repo/scheduler level, and that's the evidence to take to support. A schedule-triggered probe also runs with One last thing: you mentioned the workflow "was disabled again after the testing window." Every disable/re-enable resets the schedule, so while you're testing, leave both workflows enabled continuously. |
Schedules don’t backfill. A schedule trigger only fires for cron ticks that occur after the workflow file exists on the default branch and the workflow is active. A tick that passes before registration is simply lost, not queued.
There’s no endpoint listing “registered cron events”, but you can verify the prerequisites: bash state must be active (not disabled_manually or disabled_inactivity). Also confirm the file is on the default branch tip: bash Also make sure the YAML is valid and the cron is quoted: yaml Adding workflow_dispatch lets you manually run it, which proves the file parses and runners/billing work.
A registration delay is plausible for new workflows (I’ve seen first runs take a while after the first push). Account-level problems usually show up differently: billing/spending-limit issues on private repos create a run that fails to start with an annotation, so “zero runs” points to the event not firing rather than a billing block. Still, check Settings → Billing → Actions for usage/limits, and Repo → Settings → Actions → General for any org/enterprise policy overrides.
Leave the workflow enabled and use a temporary canary instead: yaml Wait 30–60 minutes. If runs with event schedule appear, registration works and your original workflow just needs to stay enabled through a few ticks. If still nothing, run workflow_dispatch to rule out YAML/runner issues, then push a small commit changing the cron line from your own account (to reset the actor), and contact GitHub Support with the repo name and timestamps. |
Uh oh!
There was an error while loading. Please reload this page.
🏷️ Discussion Type
Question
💬 Feature/Topic Area
Schedule & Cron Jobs
Discussion Details
Hello GitHub Community,
I'm experiencing an issue with a scheduled GitHub Actions workflow in a newly created private repository.
Environment:
main.github/workflows/p04-email-dispatch.yml7,37 * * * *Problem:
The workflow is recognized by GitHub Actions, but scheduled executions are not being created.
I enabled the workflow ahead of two scheduled opportunities:
Neither execution was triggered.
The Actions page shows zero workflow runs, with no pending jobs or error logs.
What I have verified:
My questions:
I would appreciate guidance on how to identify the root cause.
Thank you for your help!
All reactions