apache / apache/pulsar

when ServerCnx has no consumers and producers, recycle the connection

Open
#16,050 1 comment 0 reactions 0 assignees View on GitHub
Stale type/feature
Dominant language
Java
Stars
15.3k
Forks
3.8k
Avg merge
1d 14h
Merged PRs (30d)
160

Description

**Is your feature request related to a problem? Please describe.**

now, we sometimes meet this issue

our users use an old version of the client, for example, backlog exceeded.

the client will close the producer, but they won't close the connection, then the TCP connection will increase very quickly.

we can think of another scene: if the user creates a pulsar client, then create a producer,then close producer, then create a pulsar client.

**Describe the solution you'd like**
A clear and concise description of what you want to happen.

we want to use a brokerDeleteInactiveConnectionEnabled config, when we close producer and close consumer when will judge

if the current Servercnx has no consumer and producer, we will register a timer (such as 60s), if there is no consumer and producer after 60s,we will close cnx.

**Describe alternatives you've considered**
A clear and concise description of any alternative solutions or features you've considered.

**Additional context**
Add any other context or screenshots about the feature request here.

Contributor guide

Open the contributing guide

Research direction

Start by tracing ServerCnx handling when producers and consumers are closed, then locate the brokerDeleteInactiveConnectionEnabled configuration and any existing connection timers. Done means an enabled broker closes a ServerCnx only after it has had no consumers or producers for the configured delay, while active connections remain open.

Written by the indexing model from the issue text.

Assessment

Tech stack
java
Domain
distributed-systems
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.