apache / apache/fluss

[KV] Support hot standby mode for KVTablet replicas to accelerate failover recovery when the leader changes

Open
#1,456 0 comments 1 reaction 0 assignees View on GitHub
Dominant language
Java
Stars
2.1k
Forks
625
Avg merge
3d 14h
Merged PRs (30d)
97

Description

### Search before asking

- [x] I searched in the [issues](https://github.com/alibaba/fluss/issues) and found nothing similar.

### Motivation

Currently, only the leader replica maintains an active KVTablet for serving, with followers replicating logs only. When leader switching occurs, it can result in failover recovery times of up to ten minutes to half an hour when the KV snapshot is large.

### Solution

Replicas support hot standby mode for KVTablet. At any time, the system will try to maintain one leader replica, one hot standby replica, and the rest as normal replicas.

### Anything else?

_No response_

### Willingness to contribute

- [x] I'm willing to submit a PR!

Contributor guide

No contributing guide indexed for this repository

Research direction

Start by tracing the KVTablet replica and leader-switching paths described in the issue, including how followers replicate logs and how recovery is triggered after a leader change. Define completion as maintaining one leader, one hot standby, and remaining normal replicas while reducing failover recovery time for large KV snapshots.

Written by the indexing model from the issue text.

Assessment

Tech stack
java
Domain
distributed-systems
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
30/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.