Project

General

Profile

Rgw - Multi-region Disaster Recovery (phase 2)

Summary

For Dumpling we''ve laid out all the required infrastructure to support multiple regions and multiple zones. There's still some work that needs to be done on both the gateway side, and on the agent side.

Owners

  • Name (Affiliation)
  • Name (Affiliation)
  • Name

Interested Parties

  • Yehuda Sadeh (Inktank)
  • Name (Affiliation)
  • Name

Current Status

Multi-region support has been added in Dumpling. It provides global metadata information capabilities and region awareness at the bucket level. It is comprised of different logging mechanisms, and a sync agent to support it. Some initial work has been done on the agent to provide data syncing as well, however this work has not been complete.

Detailed Description

The agent's data syncing capabilities are similar in functionality as the metadata syncing, however, they differ a bit. The data syncing requires looking at two different logs (data log and the bucket index log).
One major issue to re-consider, is how to handle multipart objects upload syncing. The way things are designed currently, syncing will only start after object has been successfully uploaded, however, that may lead to very long delays when dealing with large objects. One approach would be to start syncing the immutable parts of the objects. However, we currently don't have any knowledge about these at all, and we only log that info after the upload completion.

Work items

Coding tasks

  1. Task 1
  2. Task 2
  3. Task 3

Build / release tasks

  1. Task 1
  2. Task 2
  3. Task 3

Documentation tasks

  1. Task 1
  2. Task 2
  3. Task 3

Deprecation tasks

  1. Task 1
  2. Task 2
  3. Task 3