aws-cloud-map-mcs-controller-for-k8s
aws-cloud-map-mcs-controller-for-k8s copied to clipboard
K8s controller implementing Multi-Cluster Services API based on AWS Cloud Map.
The current implementation syncs the not ready IPs also to cloud-map from the endpointslices. This ends up publishing not ready IPs to cloud-map. Can we make it a flag to...
Bumps [k8s.io/client-go](https://github.com/kubernetes/client-go) from 0.24.2 to 0.25.0. Commits 2666bd2 Update dependencies to v0.25.0 tag b3e4a40 Merge remote-tracking branch 'origin/master' into release-1.25 c2f61ae Update removal warnings to 1.26 54e42ab update-gofmt ef26118 Revert...
Bumps [k8s.io/apimachinery](https://github.com/kubernetes/apimachinery) from 0.24.3 to 0.25.0. Commits 117bd9b Merge pull request #111113 from mimowo/retriable-pod-failures-job-controller 74deb3d Merge pull request #111696 from liggitt/go119mod fef5499 Update go.mod to go1.19 41606c6 Merge pull request...
Instead of assuming AWS Cloud Map for service discovery, we should make the interface pluggable to support other providers.
Prometheus
Propose realistic testing scenario for number of clusters, services, endpoints and run the controller with that setup. Document the output on GitHub doc/readme to give users confidence about controller capabilities....
See #105 To better support cross account scenarios, we can add the ability to assume a cross-account role to the controller. Implementation wise, it would be a [credential provider](https://pkg.go.dev/github.com/aws/aws-sdk-go-v2/credentials/stscreds) that...
Right now, we are using the default configuration for reconciles. We can optimize the reconciler to improve performance. https://pkg.go.dev/sigs.k8s.io/controller-runtime/pkg/controller#Options
Current strategy for error handling is to log the error. This story is figure out ways to extend error handling capabilities, and make the reconciler fault-tolerant.