Flagger
flagger.app
What it is
Flagger is a self-hosted Kubernetes operator that automates progressive delivery for applications running on Kubernetes. It gradually shifts traffic to a new version while checking measures such as HTTP or gRPC request success and latency. Deployment approaches include canary releases, A/B testing, and blue/green deployments, routed through an ingress controller or a service mesh compatible with the Kubernetes Gateway API. Built-in checks cover HTTP request success rate and duration; teams can add custom metrics and webhooks for acceptance tests, load tests, or other validation. Release analysis can query systems including Prometheus, Datadog, CloudWatch, New Relic, Graphite, Dynatrace, InfluxDB, and Google Cloud Monitoring. If SLO analysis detects a performance drop, Flagger rolls back the release automatically. Manual controls allow approval, pausing, or resuming deployments, and notifications can go to Slack, Microsoft Teams, Discord, or Rocket. It is configured through Kubernetes custom resources, works with Kubernetes-compatible CI/CD solutions, and can be used with Flux for GitOps canary pipelines. Flagger is Apache 2.0 licensed and free.
Spec sheet
- Purpose
- Flagger automates progressive delivery for applications running on Kubernetes by shifting traffic gradually while measuring metrics and running conformance tests.Read 8 Oct 2026 · docs.flagger.app
- Product
- Flagger is a progressive delivery Kubernetes operator that automates application releases on Kubernetes.Read 8 Oct 2026 · flagger.app
- Support
- The project directs users to its documentation, the CNCF community Slack #flagger channel, Flux talks, and GitHub issues for help.Read 8 Oct 2026 · flagger.app
- License
- Flagger is licensed under Apache 2.0.Read 8 Oct 2026 · flagger.app
- Founded
- 2018Read 23 Sep 2026 · flagger.app
- Built-in metrics
- The built-in checks are HTTP request success rate and request duration.Read 8 Oct 2026 · docs.flagger.app
- Compatibility
- Flagger is configured with Kubernetes custom resources and is compatible with CI/CD solutions made for Kubernetes.Read 8 Oct 2026 · docs.flagger.app
- Configuration
- Flagger is configured using Kubernetes custom resources and works with CI/CD solutions made for Kubernetes.Read 8 Oct 2026 · docs.flagger.app
- Deployment strategies
- It supports canary releases, A/B testing, and blue/green deployments.Read 8 Oct 2026 · docs.flagger.app
- GitOps
- Flagger can be used in GitOps pipelines with tools such as Flux CD.Read 8 Oct 2026 · docs.flagger.app
- History
- Flagger was initially developed in 2018 at Weaveworks by Stefan Prodan and became a CNCF project in 2020.Read 8 Oct 2026 · flagger.app
- Manual controls
- The feature matrix lists manual gating to approve, pause, or resume deployments across supported routing integrations.Read 8 Oct 2026 · github.com
- Metrics
- Release analysis can query Prometheus, InfluxDB, Datadog, New Relic, CloudWatch, Stackdriver, or Graphite.Read 8 Oct 2026 · docs.flagger.app
- Metrics integrations
- Metric query integrations listed include Prometheus, Datadog, CloudWatch, New Relic, Graphite, Dynatrace, InfluxDB, and Google Cloud Monitoring.Read 8 Oct 2026 · flagger.app
- Notifications
- It can send deployment and canary result notifications to Slack, Microsoft Teams, Discord, and Rocket.Read 8 Oct 2026 · flagger.app
- Project history
- Flagger was initially developed at Weaveworks in 2018 by Stefan Prodan and became a CNCF project in 2020.Read 8 Oct 2026 · flagger.app
- Project status
- Flagger is a graduated Cloud Native Computing Foundation project and part of the Flux family of GitOps tools.Read 8 Oct 2026 · docs.flagger.app
- Rollback
- If SLO analysis detects a performance drop, Flagger automatically rolls back the release.Read 8 Oct 2026 · docs.flagger.app
- Routing
- Traffic routing uses an ingress controller or a service mesh compatible with the Kubernetes Gateway API.Read 8 Oct 2026 · flagger.app
- Traffic shifting
- It gradually shifts traffic to a new version while measuring metrics such as HTTP or gRPC request success rate and latency.Read 8 Oct 2026 · flagger.app
- Validation
- Built-in metric checks can be extended with custom metrics and webhooks for acceptance tests, load tests, or other validation.Read 8 Oct 2026 · flagger.app
Plans and prices
Strong points and limits
Strong points
- Supports canary, A/B, and blue/green releases
- Automatic rollback on detected performance drops
- Custom metrics and validation webhooks
- Manual approval, pause, and resume controls
Limits
- Requires a Kubernetes environment
- Self-hosted deployment
- Routing requires an ingress controller or compatible service mesh
Who it suits
Flagger suits teams running applications on Kubernetes that want automated release analysis and traffic shifting. It is a self-hosted option configured through Kubernetes custom resources.
Verdict
Flagger combines gradual traffic shifting with metric checks, manual gates, and rollback for Kubernetes releases. Its self-hosted model and Kubernetes configuration are central to its use.
Where Specifiction places it
- Progressive Delivery Software#11 of 298.2
- Kubernetes Deployment Tools#11 of 288.1
- Rollback Management#10 of 258.1
- Continuous Delivery Software#15 of 328.0