Android Build Not Finishing

[tranquil idol]

Hi,

I’m reaching out because I’m experiencing an issue with an Android build job that appears to be stuck.

I triggered an Android build to generate an APK/AAB for testing. The job has been in the status “Generating Android Build” for over 33 hours (currently more than 2000 minutes) without completing.

Job ID: e8df1623-b58c-4b4d-8a2e-5cc5ec756d7e

Some additional context:

The logs section shows no activity or output.

No error message or failure status is displayed.

I attempted to test the app using Echo Go, but it remains on a blank loading screen, which suggests the build may not have completed or been published correctly.

Waiting longer or retrying has not resolved the issue.

This has been blocking testing and deployment since yesterday.

I already reached out through community-support and was advised to contact you here for more direct assistance.

It seems possible that the job is stalled or stuck in the build queue. I’d really appreciate your help with:

Checking the current state of this specific job,

Cancelling or resetting it if necessary,

Or advising on the safest way to regenerate the Android build without running into the same issue again.

Please let me know if you need any additional information from my side.
Thanks a lot for your time and support.

[thorny temple]

Short answer: there isn’t a single magic command that guarantees 100% coverage of every flow.

What works better is being very explicit and structured with the agent. A few tips that usually improve coverage a lot:

• Ask the agent to first enumerate all pages, routes, and features (auth, dashboards, CRUD flows, edge cases)
• Then instruct it to test each item one by one, not in a single pass
• Explicitly mention negative cases (invalid inputs, empty states, permission issues)
• Ask it to report what was tested vs what was skipped at the end

For example, instead of “test the whole website”, use something like:
“List all user-facing and admin-facing flows first. Then run functional tests for each flow individually, including edge cases, and summarize coverage + gaps.”

Also, if the app is non-trivial, doing this in multiple iterations (pages → flows → edge cases) tends to be much more reliable than a single comprehensive request.

Hope that helps 👍

[tranquil idol]

Thanks for the explanation — that makes sense.
I agree that being explicit and structured with the agent improves coverage, and that approach has actually worked well for me throughout development.

In this case, the main challenge I’m facing isn’t test coverage but a blocked Android build job with no logs or feedback, which prevents further testing altogether.

I’ve already reached out via the proper support channels and got no response. Still, I'll keep waiting. Appreciate you taking the time to share the guidance.

[warped oyster]

Hello @tranquil idol, You can make this process a lot easier by using GitHub actions.

• tell the AI agent to create a GitHub actions file in the appropriate folder location inside your project.
• Once that is created please click the Save button below the chat session and your code will be saved to github.
• Once it's saved to github you go to the Actions tab and start the app compile process and wait for that to finish.

If you see any errors please let me know I can help you figure those errors out to tell the AI agent to fix them by giving the AI agent the full errors and it can correct any issues.

Hope this was helpful.

Thanks,

TierOne - @warped oyster