Open zscomehuyue opened 9 months ago
I've seen that the version of EventMesh in your issue is 1.2.0. May I ask that have you ever tried the lastest version to test whether the bug still remained?
I tested 1.9.0 and found the issue, which should also be present in the master version
这情况是消费者耗时比较长把nio全部阻塞后,导致写事件没有线程处理导致的; 心跳和读取消息使用不同线程池:
心跳和读取消息使用不同线程池
Good idea.
消费者耗时比较长把nio全部阻塞
Is it because of a network failure or an application design flaw?
an application design flaw
It has been 90 days since the last activity on this issue. Apache EventMesh values the voices of the community. Please don't hesitate to share your latest insights on this matter at any time, as the community is more than willing to engage in discussions regarding the development and optimization directions of this feature.
If you feel that your issue has been resolved, please feel free to close it. Should you have any additional information to share, you are welcome to reopen this issue.
It has been 90 days since the last activity on this issue. Apache EventMesh values the voices of the community. Please don't hesitate to share your latest insights on this matter at any time, as the community is more than willing to engage in discussions regarding the development and optimization directions of this feature.
If you feel that your issue has been resolved, please feel free to close it. Should you have any additional information to share, you are welcome to reopen this issue.
Search before asking
Environment
Windows
EventMesh version
1.2.0
What happened
After the client subscribes to the message [TCP], asynchronous [thread pool with 5 threads] reads the message and sleeps for 5 minutes. The server pushes a large number of messages;
Problem: Heartbeat heartbeat, unable to send to the server, resulting in session timeout being cleared;
How to reproduce
Subscribe to a topic, and then sleep for 5 minutes after receiving the message;
Send 1000 messages to the topic
Debug logs
Are you willing to submit PR?
Code of Conduct