PR: Implement Prometheus/OpenTracing Instrumentation in k8s Storage Shim

[dvonthenen]

Implement Prometheus/OpenTracing (with Zipkin? can abstract?) Instrumentation in higher level shim layer in Kubernetes

Prometheus Metrics:
Currently, a proposal for an implementation to implement metrics. Verified that this sits above the plugins and will require no change to individual plugins. Currently targeted for 1.8 and has an assignee. PR: kubernetes/community#809

Tracing:
There is an open issue to consider tracing support using some OpenTracing implementation. A PR was submitted that implemented Jaeger (https://github.com/jaegertracing/jaeger-kubernetes) which is a project out of Uber. The PR was dropped in favor of revamping Kubernetes' Event mechanism in the proposal (not implemented yet. target for 1.7 but was missed. now labeled next.) can be found here (need to join sig-apimachinery or sig-scale groups for access): https://docs.google.com/document/d/13BeJlrEcJhSKgsHOHWmHdJGqXrjSm_o9XxOtwcN6yNg/edit#heading=h.vsvi1hdn0y4v

Interesting side note is that in the gRPC ecosystem there is an effort under way to use gRPC as a transport for an OpenTracing implementation. This project is just getting underway and is still very new. They don't have any releases published as of yet. Information found here:
https://github.com/grpc-ecosystem/grpc-opentracing Only server side implementation for this transport seems to be https://github.com/tracer/tracer, which is listed on OpenTracing.io but is dead last in the list. Looked over source code and the implementation seems to be fairly new.

Deliverable Date: 8/8

[dvonthenen]

Updated the initial post with some information I have discovered regarding metrics and tracing in Kubernetes. Please above.

[dvonthenen]

Closing as this issue is no longer relevant.