[Bug] After running for a period of time, the provider is lost.

### Pre-check

- [X] I am sure that all the content I provide is in English.

### Search before asking

- [X] I had searched in the [issues](https://github.com/apache/dubbo/issues?q=is%3Aissue) and found no similar issues.

### Apache Dubbo Component

Java SDK (apache/dubbo)

### Dubbo Version

Dubbo version 3.2.11
JDK 21
in k8s

### Steps to reproduce this issue

start provider and consumer in k8s use tri protocol。
not using a registry center, but connecting directly using DNS.
There is no issue during startup, but after running for a period of time (about 7 or 8 hours), the client reports that it cannot connect to the provider.

### What you expected to happen

idk

### Anything else

_No response_

### Are you willing to submit a pull request to fix on your own?

- [ ] Yes I am willing to submit a pull request on my own!

### Code of Conduct

- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)

Reproduce the Java SDK 3.2.11 setup in Kubernetes with the tri protocol, direct DNS connection, and no registry center. Start by investigating why the consumer loses connectivity after approximately 7–8 hours using the attached error evidence. Done means identifying the cause and verifying that the consumer remains connected to the provider during sustained operation.

Written by the indexing model from the issue text.