Hello Google AI Studio team, I am experiencing a persistent publication problem when deploying my Google AI Studio Build app to Cloud Run. The application itself is working correctly in the existing Cloud Run service. The current working revision is receiving 100% of the traffic. However, when I publish an updated version from AI Studio, Cloud Run creates a failed revision with: “The provided source archive is corrupted.” The latest failed revision is: schulapp-klassen-sch-lerverwaltung-00005-btg Another failed revision is: schulapp-klassen-sch-lerverwaltung-00004-pcq The Cloud Run service is in region: europe-west2 After these failed publication attempts, Cloud Run also reports: “Operation failed due to insufficient quota. Violation: Quota ActiveRevisionsPerProject current use of 4 in europe-west2 would exceed limit of 4.” This means that further AI Studio publication attempts are now blocked by the revision quota. The current revisions are: - 00003-8q2 — working, 100% traffic - 00004-pcq — failed - 00005-btg — failed, “The provided source archive is corrupted” - 00002-mmd — 0% traffic - 00001-gpm — 0% traffic I also cannot delete the old revisions from my user account because my account does not have the required Cloud Run revision deletion permission. The important point is that the application itself is already running successfully on the existing revision. The failure appears to occur during the AI Studio publication process when the new Cloud Run revision is created. Could the AI Studio team please investigate: 1. Why AI Studio is creating a corrupted Cloud Run source archive. 2. Whether this is a known AI Studio Build/Publish issue. 3. Whether the failed revisions and the ActiveRevisionsPerProject quota can be cleared so that publishing can continue. I can provide the App ID, project ID, revision IDs, operation IDs, timestamps and sanitized audit logs privately if required. Thank you.
Hi @L_Weber,
Can you please share the screenshots of the steps that you are undertaking as well as the screenshot of the error?
I am unable to publish my application from Google AI Studio Build. Whenever I click the “Publish” button, the deployment fails with the following error:
Failed to update Cloud Run service: user has exceeded quota. Please try again later.
I have been completely blocked by this error for over 30 hours (since September 24), so it is not a temporary 24-hour rolling rate limit.
Project & Applet Details:
- Applet ID: e297c391-29ad-4b01-be14-b32e5416e1e5
- Account Email: [in..@gmail.com]
- Region: europe-west3
What I Have Verified:
- Code & Build Health: The application compiles cleanly with 0 errors (npm run build, lint, and TypeScript all pass). The dev server is active and running with 0 issues.
- Prior Revision: The existing previously deployed container revision (ais-pre-…) is still online, but no new revisions can be published through AI Studio.
- Workspace Size: All build caches and large artifacts have been purged; the workspace archive is lean and minimal.
- Permissions: Attempting to view the quota in Cloud Console for project 247631084367 returns “You need additional access” because this is Google’s internal managed AI Studio project.
It appears that the publishing pipeline or internal service revision lock is stuck on the backend cluster for this Applet ID.
Could a member of the Google AI Studio team please help inspect and clear the deployment lock / quota state for this Applet ID?
Thank you!
Hi @DanielK1999, There is no ongoing backend issue that’s causing this nor do we have any way to ‘clear the deployment lock’ Are you on free tier? Can you please share screenshots of the error?