GitHub Pages deployment stuck with HTTP 400 – in-progress deployment lock

[xuan-weather]

🏷️ Discussion Type

Question

💬 Feature/Topic Area

Workflow Deployment

Discussion Details

Hello GitHub Community,

I’m experiencing a persistent GitHub Pages deployment issue in my repository:

https://github.com/xuan-weather/xuan-weather

The deployment repeatedly fails with HTTP 400 and the following error:

Deployment request failed due to in progress deployment. Please cancel 10d6071ed60da1758774694e52cf79df15f82646 first or wait for it to complete.

However, when I tried to cancel that deployment, the API previously reported that the deployment was already finished.

A previous GitHub Pages deployment succeeded, but subsequent deployments continue to fail because GitHub reports an in-progress deployment.

I have already checked my workflow configuration and removed the custom cancellation step, but the issue persists.

Could someone help investigate this inconsistent GitHub Pages deployment state and explain how to clear the deployment lock?

Latest Request ID: A818:BED2E:4238F1C:DB3FEE4:6AE6FEF6

Thank you for your help.

[DeddCode31251]

Hi there, Deadcode here.

This happens when a GitHub Pages deployment gets unexpectedly interrupted, leaving a stale "in-progress" lock on the backend API while the front-end Actions UI registers it as completed or canceled. Because the lock is attached directly to that blocking deployment SHA, standard workflow cancellations won't clear it.

You can force-cancel the stuck deployment state and reset the pipeline by trying these steps:

1. Force-Cancel via GitHub CLI

Because the standard UI cancellation fails, you need to target the Pages deployment endpoint directly. Run this command using the GitHub CLI:

gh api -X POST repos/xuan-weather/xuan-weather/pages/deployments/10d6071ed60da1758774694e52cf79df15f82646/cancel

(If the workflow run itself is hanging in a queued state, you can also force-terminate it by finding the workflow's Run ID from the Actions URL and running gh api -X POST repos/xuan-weather/xuan-weather/actions/runs/<RUN_ID>/force-cancel).

2. Reset the Environment State

Once you run the cancellation command, force the environment settings to refresh:

1. Go to your repository's Settings -> Pages.
2. Under Build and deployment, temporarily switch the Source dropdown from "GitHub Actions" to "Deploy from a branch".
3. Click Save, wait a moment, then toggle it right back to "GitHub Actions" and click Save again.

3. Kick off a Fresh Run

Finally, push an empty commit to clear out the remaining queue and trigger a pristine deployment run:

git commit --allow-empty -m "flush deployment lock"
git push origin main

If the API still throws an HTTP 400 error after attempting a direct CLI cancellation, the backend environment is completely wedged on GitHub's side. In that case, you will need to open a quick ticket with GitHub Support. Give them your Latest Request ID (A818:BED2E:4238F1C:DB3FEE4:6AE6FEF6) so an engineer can manually flush the stuck database lock.