Sysbench Testing TiDB with HAProxy Load Balancing: Database Connection Fails at 4500 Concurrent Connections

Translated
1

Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.

Original topic: sysbench测试tidb,haproxy负载均衡，并发到4500，数据库连接失败

| username: Hacker_a0z45bcw

【TiDB Usage Environment】Production Environment / Testing / PoC
【TiDB Version】
【Reproduction Path】Operations performed that led to the issue
【Encountered Issue: Issue Phenomenon and Impact】
【Resource Configuration】
【Attachments: Screenshots / Logs / Monitoring】
FATAL: error 2013: Lost connection to MySQL server at ‘handshake: reading initial communication packet’, system error: 11
FATAL: thread_init' function failed: /usr/local/share/sysbench/oltp_common.lua:349: connection creation failed FATAL: unable to connect to MySQL server on host '10.47.74.216', port 4000, aborting... FATAL: error 2013: Lost connection to MySQL server at 'handshake: reading initial communication packet', system error: 11 FATAL: thread_init’ function failed: /usr/local/share/sysbench/oltp_common.lua:349: connection creation failed
FATAL: Thread initialization failed!

2
| username: tidb菜鸟一只 | Original post link

What is the cluster configuration like?

3
| username: liuis | Original post link

Why do I feel that 4500 concurrency is already very high :joy: With the addition of cache, it can handle tens of thousands of QPS.

4
| username: xingzhenxiang | Original post link

4500, what kind of requirement is that?