-
Feature
-
Resolution: Done
-
Major
-
None
SmallRye Stork provides service discovery and client-side load balancing features. This task is about the integration of Stork into Quarkus. It mainly targets the REST Client and gRPC client for now. The graphQL client could be added if time allows.
The inclusion should only support static discovery, DNS, Consul (community only), and Kubernetes.
Details from the first comment (which is usually hidden as there are many comments):
Details from Stork dev-qe information exchange meeting:
- Beta1 atm, .Final in few days / for Quarkus 2.7, metadata topic + API final touchaes
- Missing guide, missing Quickstart
- Aurea working on blog post for Kubernetes use-case with Stork
- http://smallrye.io/smallrye-stork for current component level docs
- Service Discovery results are cached
- Cache invalidation time is 5 minutes atm
- Monitoring of service discovery / loadbalancing is future task
- Custom load balancing strategy can access request details using context-propagation
- Developing service discovery for "all" the providers is not the goal of Stork
- Same for load balancing strategies, only the base ones to be part of the Stork code base, customers can implement ones based on their needs
Areas for RHBQ / TP level of this RFE:
Service Discovery
- Kubernetes Service Discovery
- Custom service discovery mechanism (API to be marked as finalized with 1.0.0.Final)
Load-Balancing
- Round-Robin Load Balancing
- Custom load balancer mechanism (API to be marked as finalized with 1.0.0.Final)
- is related to
-
QUARKUS-1949 Stork / k8s extensions is listing all endpoints of a given namespaces
-
- Closed
-