configserver hang

[Jerrywgl]

版本： 2.10

configserver 服务异常，表现如下：

8080//health 接口的响应延迟上升，甚至超时(大于35s)
8080端口，有时候会出现TCP连接超时。（超时间大于10s）,甚至会出现持续30s.

而且当时请求并没有明显的上升。

[Jerrywgl]

暂时没有收集到有用的日志，有大佬遇到过吗

[nobodyiam]

可以看下 cpu/mem/gc 等指标

[dosubot]

I found some similar discussions that might be relevant to your issue:

1. 使用Kubernetes部署报socket read timeout exception，请问有可能是什么问题？？ - This discussion is open and might provide insights into socket read timeout issues in a Kubernetes deployment ^[1].

2. 启动portal报错：Env health check failed - This discussion is also open and discusses health check failures, which might be related to your problem ^[2].

Regarding the health check and TCP connection timeout issues you're experiencing with Apollo configserver version 2.10, here are some configurations you might consider:

• Connection Timeout: Configured using api.connectTimeout, default is 3000 milliseconds (3 seconds).
• Read Timeout: Configured using api.readTimeout, default is 10000 milliseconds (10 seconds).
• Connection Time To Live: Configured using api.connectionTimeToLive, default is -1, indicating no limit ^[3].

Additionally, Apollo configserver handles health checks using the ConfigServiceHealthIndicator class, which checks the health status by attempting to retrieve a page of applications. A successful check reports the health status as "up" ^[4].

You might want to review these configurations and ensure they are set appropriately for your environment. Adjusting these settings could help mitigate the timeout issues you're facing.

To continue talking to Dosu, mention @dosu.

---

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Other}

[Jerrywgl]

可以看下 cpu/mem/gc 等指标 这些都已经关注过了，都是正常的。大佬可以加个联系方式吗？我们可以专门讨论下。