clusterpedia icon indicating copy to clipboard operation
clusterpedia copied to clipboard

Add message queue to implement multi-cluster watch interface When using the default storage layer

Open wuyingjun-lucky opened this issue 1 year ago • 3 comments

What would you like to be added?

Add message queue to implement multi-cluster watch interface When using the default storage layer

Why is this needed?

Improve Clusterpedia Query capability (dynamic query)

wuyingjun-lucky avatar Nov 17 '23 07:11 wuyingjun-lucky

Hi @wuyingjun-lucky, Thanks for opening an issue! We will look into it as soon as possible.

Details

Instructions for interacting with me using comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the gh-ci-bot repository.

clusterpedia-bot avatar Nov 17 '23 07:11 clusterpedia-bot

image

hi, this is our solution for the multi-cluster watch feature.

zhouhaoA1 avatar Nov 21 '23 01:11 zhouhaoA1

  1. A message queue is used for transporting events from ClusterSynchroManager to ApiServer, before transporting, events must be persisted to the storage layer, and rv of events must be converted to multi-clusters-rv.
  2. EventCache is a buffer for newest events whose number is limit to 100 by default.
  3. we have defined a series of universal inferfaces for message queue middleware. And we have introduced RabbitMQ by implement the interfaces.

zhouhaoA1 avatar Nov 21 '23 01:11 zhouhaoA1