Lantao Liu
Lantao Liu
@qiutongs volunteered to help fix this issue. :) Thanks!
> is this one related? #5438 Yeah, that can fix the context timeout issue, but can't fix the issue that network teardown just returned an error. To make containerd work...
With this change, even though the pod network setup and teardown are moved down, our problem is still not resolved: 1) The sandbox container metadata will still be deleted from...
> 1. create container record without netns info > 2. setup netns info > 3. update container record with netns info > 4. setup network > 5. start container This...
@mikebrow In Kubernetes, we do have `PostStart` and `PreStop` hook, but we are using exec to implement it. Runc may help us implement more hooks, e.g. `PreStart` and `PostStop`.
/assign @Random-Liu
Any updates on this? /cc @stevvooe @crosbymichael I don't think `bufconn` is enough for us. We can only wrap all the grpc services now.
> I don't really understand why you can't use bufconn though. @jhump Because we may also want to get rid of the serialization/de-serialization, :) With https://godoc.org/github.com/fullstorydev/grpchan/inprocgrpc, can I make the...
@jhump Nice. I'll look into it a bit more. Our use case is that `containerd` has a official grpc api. And `containerd` has a plugin for Kubernetes support is called...