AI Skill
ecsStopTimeoutSecs config option to all ECS based services
CLI to validate flightcontrol.json
Get job execution API endpoint
Bring your own subnets now supported for preview environments
New docs:
Ability to override image repository in create deployment API
Enable encryption to instance root volumes for all ECS EC2
Better handling of install failures for EC2 builders
Improve deletion
Add createdAt and updatedAt to get services API endpoint
Improve ECS error messages
Fix log view on short screens
Fix cloudmap namespaces not being cleaned up
Fix issue with preview environment domains
Fix shell escaping issues for env vars in EC2 builders
Fix issue connecting Github account
Fix user canβt accept invite if they have no existing organization access
Fix API error being swallowed
Aside from the below, we are hard at work getting the major new version ready for early access testing!
Protected Environments
Now you can enable delete protection under environment settings in the dashboard
Also restricts actions for users with the new Restricted Developer role
New Restricted Developer RBAC role
Same as Developer, except in protected environments they cannot view env var values or connection strings, cannot edit environment config, cannot add or delete services, cannot add or delete domains, but can change service config, add & update env var values, and deploy.
Fine grained autoscaling config
Add following fields:
scaleInCooldownTimerSecs (falls back to deprecated cooldownTimerSecs if not present)
scaleOutCooldownTimerSecs (falls back to deprecated cooldownTimerSecs if not present)
scaleInProtectionCPU
scaleInProtectionMemory
scaleInProtectionRequests
Use your own VPC subnets
Previously we would create new subnets in your VPC
Now you can connect your existing subnets
Fix various ECS-EC2 deployment issues
Fix blue/green deployment issues
Fix port form error not showing on network server
Misc fixes and improvements
Blue green deployments for ECS based services in private preview
Blue-Green deployment is a deployment strategy that enables zero-downtime deployments by maintaining two identical production environments. This allows you to deploy new versions of your application, test them in a production-like environment, and then immediately switch traffic to the new version. It also makes rollbacks very fast.
Contact support if you'd like to try it
Postgres v18 support
Improve reliability of ECS EC2 clusters and deployments
Allow build cancellation in all cases
Fix Contributor role to be able to create env variables
Fix builder git clone error
Fix edge case with Lambda provisioning
Fix issue with form dialogs
Fix delete account flow
Fix IAM permissions not setting for scheduler service
Fix projects list pagination
Fix project cost calculation error
Misc fixes and improvements
Another round of exciting updates for you all! We're now working on blue/green deployments for ECS based services, in addition to the ongoing major pipelines upgrade still in progress that will unlock infinite flexibility for you.
Managed ECS-EC2 clusters now enabled for everyone
AWS has a long standing issue with the ECS agent randomly disconnecting, resulting in orphaned EC2 instances which can cause traffic or deployment degradation.
We have attempted to solve this a few ways in the past, but there were still critical edge cases falling through.
So we bit the bullet, and developed a robust, full featured ECS cluster management solution to solve this problem once and for all.
It's enabled for everyone and working very well
New "Server instances below minimum" notification
Enabled by default for all. This will notify you if your running server instances go below your configured "minimum instances" count.
If you receive this, it means one or more of your servers likely crashed (it will be auto replaced by ECS).
Add Linux Capabilities config to ECS based services
Support ElastiCache Valkey 8.1
Fix environmentType filter in GET /services API
Misc fixes
Managed ECS-EC2 clusters in preview
AWS has a long standing issue with the ECS agent randomly disconnecting, resulting in orphaned EC2 instances which can cause traffic or deployment degradation.
We have attempted to solve this a few ways in the past, but there were still critical edge cases falling through.
So we bit the bullet, and developed a robust, full featured ECS cluster management solution to solve this problem once and for all.
It's currently in private preview. To get early access before we roll it out to everyone, contact support.
Scaling API endpoint to manually scale servers up and down
You can use this to implement queue-length based autoscaling OR to scale staging and preview environments to 0 outside work hours
Ability to disable autoscaling in service config
List services API endpoint
Add download button to runtime logs
Scheduler UI improvements
Added link to ECS service AWS resource card
Fix rollback case where no services were available for rollback
Fix some deletion issues
Fix a budget UI not working
Fix some GitHub integration issues for some users
Add scale to 0 and ability to pause services by allowing minInstances and maxInstances to be set to 0
Cost budgets for preview environments
Support for Mysql v8.4 and MariaDB v10.11 and v11.4
Self-serve ability to delete your entire Flightcontrol organization and user account
Better orphaned EC2 instance management for ECS+EC2 service types
Improve select git repo authentication flow and repo loading performance
Update static site edge lambdas
Automatically update Github repo name in Flightcontrol when changed in Github
Improve a number of error messages
Fix ECS AWS console link for Network Services
Fix access denied error in an edge case for image registries
Misc fixes and improvements
Deployments now show service config and a diff of config changes

ElastiCache Valkey now supported in addition to Redis.
Valkey is AWS's fork of Redis with improved performance and reduced cost. It's a drop in replacement.
Deploy specific docker image tags or shas with Create Deployment API
New Lambda config option injectEnvVariablesInRuntime to disable our default env var injection (docs)
Improve performance of jobs execution page when you have thousands of executions
Improve some deploy error messages
Fix environment duplication to include plain text env vars and service level env vars
Don't allow empty start command for scheduler jobs
Fix "Deploy this commit" does not use correct image for image registry build type
Misc fixes
New Network Server service type is now available in public preview
Supports exposing multiple TCP, UDP, HTTP, and HTTPS ports
Let us know if you encounter any problems!
Add full deployment timestamp to a tooltip when hovering over the deployment time
Improve scheduler job's list UI and pagination
Some issues with edge Lambdas
Fix a number of issues with Lambda services
Fix issue with region selector scroll being reset
Add missing AWS console link on load balancer resource card