Forest Admin Outage History

We are currently experiencing a service disruption due to an ongoing outage at our hosting provider, Heroku. You can follow their status updates here: https://status.heroku.com/incidents/2822. Our technical team is actively investigating the impact on our services and is closely monitoring the situation. We will continue to provide updates here as we learn more. Thank you for your patience and understanding.

We've mitigated the issue by deploying our application on AWS. The application should be back but custom domains are still not working. If you are on a custom domain, please add app.forestadmin.com to your CORS configuration and access https://app.forestadmin.com instead.

The team quickly investigated to understand the reasons of these errors. We have rollbacked a recent contribution that led to the current situation. All API services are now up and running.

All API endpoints respond with the 500 status code.

All the missing activity logs have been re-introduced. This is the final update for this incident.

The team is currently investigating on this regression.

All the missing activity logs have been re-introduced. This is the final update for this incident.

Outage started at 17:01 (GMT+1) Outage ended at 17:22 (GMT+1) Servers have stopped responding to requests due to incorrect configuration. After a reconfiguration, systems have been able to start up properly and we could answer to requests. https://docs.forestadmin.com/public-api

We’re currently investigating reports of a potential issue on Smart Charts and Alter values. We apologize for any inconvenience and will post another update as soon as we learn more.

We have identified the source of the issue and are rolling back one of this morning's releases to resolve it.

Some customers have reported that their collections are not visible. We are currently investigating the issue.

We identified the cause of the hidden collections issue some clients were experiencing. Those affected were using the new folder feature. They were able to generate unexpected requests that resulted in folders being created but not listed. As a result, after a refresh, those folders and their contents would disappear. We have addressed the issue on our backend, and a migration is currently underway to fix the affected projects. The issue should be resolved within a few minutes. We are still investigating how these requests were generated.

We are currently experiencing a service outage. This outage has affected our api. The engineering team is working to identify the root cause and fix the issue. We apologize for any inconvenience and we'll keep you updated on the status.

A migration of our database has been incorrectly propagated across our instances, preventing some of our customers to have access to their project.

Some collections are not showing anymore in the navigation bar on the left

Our app is currently experiencing an issue where "belongs to" data is unexpectedly displayed as a record alongside the actual records in the Related Data views

We’re currently investigating the issue and will keep you updated once we got more information.

This morning, our Technical Team triggered a release containing a database migration to prepare a new important feature. This migration aimed to update our main database. The migration locked the main database to execute schema and data transformation. For the entire time the migration was running, a transaction lock prevented users from accessing our main database, via API calls, for their own needs. All requests ended up in error. This migration used a new utility function that was introduced last week to prevent migrations from running more than once (in very rare concurrent cases) but its usage requires further work. Our Technical Team is currently working on aligning migration practices and documentation for this new utility function. This will prevent similar events in the future. Thank you for your patience during this incident.

The migration should be finalized soon. Our team is investigating why such generic migration led to the current situation.

This morning, our Technical Team triggered a release containing a database migration to prepare a new important feature. This migration aimed to update our main database. The migration locked the main database to execute schema and data transformation. For the entire time the migration was running, a transaction lock prevented users from accessing our main database, via API calls, for their own needs. All requests ended up in error. This migration used a new utility function that was introduced last week to prevent migrations from running more than once (in very rare concurrent cases) but its usage requires further work. Our Technical Team is currently working on aligning migration practices and documentation for this new utility function. This will prevent similar events in the future. Thank you for your patience during this incident.

Our hosting provider just acknowledged their incident: https://status.heroku.com/incidents/2721 They confirmed that this new issue was once again affecting the EU area.

The Technical team observes a high level of 4XX responses. We are currently investigating.

We suspect that our hosting provider is currently facing new routing issues. Other customers report similar issues right now on X: https://x.com/search?q=heroku&src=typed_query

Our hosting provider just acknowledged their incident: https://status.heroku.com/incidents/2721 They confirmed that this new issue was once again affecting the EU area.

We experienced a brief service degradation between 11:30 and 11:49 CEST where some users encountered errors when accessing the platform. The issue was caused by some of our servers reaching their connection capacity during peak usage. All services are now operating normally. We sincerely apologize for the inconvenience and appreciate your patience.

Starting at 10:00 a.m., our backend experienced downtime due to simultaneous outages affecting AWS and Heroku services worldwide. By 10:30 a.m., we successfully updated our routing policy on AWS to fully switch our backend to our multi–availability zone AWS infrastructure, restoring service stability. Thank you for your patience and understanding 🙏

Please read our official communication to the community about the recent incidents: https://community.forestadmin.com/t/october-2024-forest-admin-important-downtime-on-october-8-9/7529

Our team is investigating, we suspect our hosting platform to be the culprit.

Similar apps using the same hosting platform observe the same downs. Users of our hosting platform report similar issues on X: https://x.com/search?q=heroku&src=typed_query We are waiting for Heroku to acknowledge the issue: https://status.heroku.com

Heroku just acknowledged the issue on their end, you can follow the evolution directly following this link: https://status.heroku.com/incidents/2684

Please read our official communication to the community about the recent incidents: https://community.forestadmin.com/t/october-2024-forest-admin-important-downtime-on-october-8-9/7529

The impacted projects should now be able to use the impacted Smart Views and Smart Chart as before the incident. This regression has been introduced while working on improving the overall Smart View and Smart Chart edition experience. Editors must now opt-in the JavaScript activation, understanding the consequences in terms of security. This improvement is now entirely live and does not impact any existing Smart View and Smart Chart anymore. Thanks for your patience.

Smart Views on collections without Analytics per record charts on any environment type, Smart Charts in Analytics per record section on non-production environments, are not usable anymore. Custom JavaScript is not executed on these Views and Charts.

Our API is currently failing to authenticate users. Our technical team is actively investigating the issue.

The issue has been identified and solved. All services are up and running. We apologize for any inconvenience this may have caused and thank you for your patience. Incident report : https://forest.slite.page/p/8Iik3jo9_MXQZn/Incident-2025-07-05