Firebase Outage History

The issue with Firebase Realtime Database has been resolved for all affected projects as of Monday, 2025-11-03 18:13 PST. During the incident, Firebase Cloud Functions that are triggered by Realtime Database events were not firing, which might have impacted automated workflows for customers. While the databases themselves are working, any function that is configured to trigger on certain read/write events may have been significantly delayed and may only have executed after the issue got mitigated. From preliminary analysis, the issue was triggered by an internal backwards incompatible change in a new version rollout. Our engineers mitigated the issue by deploying a fix to all affected regions. We thank you for your patience while we worked on resolving the issue.

The issue has been mitigated on our end. Customers should now be able to change resources (e.g. firebase deploy) normally. For customers who ran into partial / total failures during an earlier deployment, we recommend double-checking the current resource status and deploying again if needed.

(The service has been in a stable state since last update. We just want to give an update to the cause.) The issue was due to a bad rollout in a downstream dependency, and our engineers have mitigated the issue by fully rolling back to a previous version. We are continuing to monitor the situation to ensure the issue is fully resolved.

The issue is external and has been mitigated. Tests should start to pass again. This is the last update unless something changes.

Firebase Studio users are experiencing publishing errors when using the "Publish" button.

Arm emulator test executions have resumed, though at a reduced capacity.

A slow-rolling update to the build system of Firebase App Hosting broke customers' code from being able to see environment variables defined in YAML files. Though the rollout started Monday October 6, we have not correlated this with any customer reports before 8P PST Thursday October 9. The experiment has been rolled back and most customers were never on the experiment path.

The issue has been fixed.

Mitigations have been put in place for docker.hub and Cloud Build, and build error rates have recovered.

Fix has been fully deployed to the fleet. Normal operations should be restored shortly. Apologies again for inconvenience and thank you for your patience. Next update in four hours or when something significantly changes.

Service should now be nearly fully restored.

The fix has been rolled out

Certificate creation for new Custom Domains in Firebase Hosting is taking longer than usual. Certificates are still expected to be available within the alloted 24 hour window.

Resolved around 1:00 PM PST

The issue with app prototyper workspace creation has been fully resolved.

Crash-free metrics are now displaying normally.

The problem with Realtime Database should be resolved for the vast majority of affected users. We will continue to work towards restoring service for the remaining affected users, but no further updates will be added to this dashboard.

Resolved: Can't create Flutter or React Native workspaces.

All services have recovered.

The outage preventing users from creating new versions is now mitigated.

The outage is now mitigated. During it, users would have noticed issues when downloading their app configuration files.

RTDB Control API were unavailable during the incident

The fix has been rolled out successfully and App Prototyping agent workspaces are working normally again

We expect to resolve the problem affecting less than 1% of FCM user base by 23:59 PT, July 9th. Please note that this time frame is an estimate and may change.

We are currently investigating an issue preventing some workspaces from opening

We are currently investigating an issue preventing some workspaces from opening.