Incidents | Truefoundry

Logins and signups are failing

Thu, 16 Jul 2026 10:01:00 -0000

The root cause was the dependence of an upstream provider on cloudfront which underwent a global downtime. We are completely up now

Logins and signups are failing

Thu, 16 Jul 2026 08:17:00 -0000

Logins and signups are failing for all users

Some of the requests on Truefoundry control plane UI are getting timed out

Wed, 15 Jul 2026 14:05:00 -0000

We have identified the issue and mitigated it. TrueFoundry Control Plane UI is healthy now.

Some of the requests on Truefoundry control plane UI are getting timed out

Wed, 15 Jul 2026 13:51:00 -0000

Some of the requests on Truefoundry control plane UI are getting timed out. We have the root cause and are working on fixing it

Control Plane APIs failing with intermittent 500 errors

Fri, 05 Jun 2026 15:28:00 -0000

We have identified the root case and resolved it. Few of the MCP server API was affected causing errors in Frontend to crash with 500 error page. We have mitigated the issue and all control plane APIs are up now. This does not affect any AI Gateway functionality.

Control Plane APIs failing with intermittent 500 errors

Fri, 05 Jun 2026 14:30:00 -0000

Control Plane APIs failing intermittently with 500 errors. During rollout, few APIs have started failing with 500 errors intermittently.

AI Gateway: chat/completions API with 'prompt_fqn' failures

Fri, 22 May 2026 16:42:00 -0000

We have identified the issue in the authorization of prompts in AI Gateway leading to failures in request containing prompt_fqn in chat/completion API. This was because of a bug in the gateway code that affected prompt hydration in the gateway and resulted in authorization errors. We will be adding more tests to safeguard this code path so that this doesn’t happen again in the future. We have resolved the issue completely.

AI Gateway: chat/completions API with 'prompt_fqn' failures

Fri, 22 May 2026 06:52:00 -0000

chat/completions API request with prompt_fqn in the request body started failing with error: 'An error occurred while processing prompt configuration: Prompt not found or access denied: '.

AI Gateway: Some regions experienced 401 error for inference APIs

Wed, 06 May 2026 11:30:00 -0000

We have identified the root cause and resolved it. All the regions of AI Gateway are operational now and we are actively monitoring for any issues.

AI Gateway: Some regions experienced 401 error for inference APIs

Wed, 06 May 2026 11:20:00 -0000

Some users were experiencing 401 error for their LLM Model inference via AI Gateway. Following error was observed in few of the regions. Error code: 401 - {'status': 'failure', 'message': 'Unauthorized: Invalid token: Unable to verify signature', 'error': {'message': 'Unauthorized: Invalid token: Unable to verify signature', 'type': 'Error', 'code': '401'}, 'error_origin_level': 'authentication'} Incident started at 11:20 AM UTC on 6th May 2026 and was successfully resolved at 11:30 AM UTC, 6th May 2026. Duration of incident: 10 min.

Platform is not loading and giving intermittent 500 page

Tue, 05 May 2026 15:14:00 -0000

We have identified the root cause and resolved the issue. All the APIs and Platform UI are operational. We are actively observing for any issues.

Platform is not loading and giving intermittent 500 page

Tue, 05 May 2026 15:02:00 -0000

Platform was not loading intermittently and giving a 500 error page. Users are unable to use the Platform using UI. Deployment APIs are giving 500 error intermittently and AI Gateway is completely operational.