[3.3] Enable server socket reuseAddress option before binding

[zrlw]

What is the purpose of the change?

try to fix #15238
sometimes the port got from NetUtils#getAvailablePort could not be bound directly because the port status is still TIME-WAIT as getAvailablePort closed it without setting SO_REUSEADDR option.

Checklist

• Make sure there is a GitHub_issue field for the change.
• Write a pull request description that is detailed enough to understand what the pull request does, how, and why.
• Write necessary unit-test to verify your logic correction. If the new feature or significant change is committed, please remember to add sample in dubbo samples project.
• Make sure gitHub actions can pass. Why the workflow is failing and how to fix it?

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 72.22222% with 5 lines in your changes missing coverage. Please review.

Project coverage is 60.78%. Comparing base (d3b1e1f) to head (cba2c7b).
Report is 15 commits behind head on 3.3.

Files with missing lines	Patch %	Lines
...n/java/org/apache/dubbo/common/utils/NetUtils.java	73.33%	2 Missing and 2 partials ⚠️
...ng/transport/netty/NettyPortUnificationServer.java	0.00%	1 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff              @@
##                3.3   #15239      +/-   ##
============================================
- Coverage     60.79%   60.78%   -0.01%     
- Complexity    10892    10894       +2     
============================================
  Files          1885     1885              
  Lines         86082    86102      +20     
  Branches      12895    12898       +3     
============================================
+ Hits          52331    52341      +10     
- Misses        28303    28316      +13     
+ Partials       5448     5445       -3     

Flag	Coverage Δ
integration-tests	33.09% <50.00%> (-0.06%)	⬇️
samples-tests	29.23% <33.33%> (+0.04%)	⬆️
unit-tests	58.92% <72.22%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[oxsean]

Is it possible to add some tests?

[zrlw]

Is it possible to add some tests?

the results of these codes vary with os and jdk version,

ServerSocket serverSocket = new ServerSocket(port);
assertFalse(serverSocket.getReuseAddress());

passed if running on win10 + compiled by jdk8,
failed if running on linux + compiled by jdk17 as serverSocket.getReuseAddress() returns true.
i doubt there should have unified solution.

[zrlw]

see SO_REUSEADDR at java.net.StandardSocketOptions,

Re-use address.
An implementation allows this socket option to be set before the socket is bound or connected. Changing the value of this socket option after the socket is bound has no effect. The default value of this socket option is system dependent.

since the default value of SO_REUSEADDR is always true on some system, I didn't think of how to write appropriate testing case.

    @Test
    void testRepeatedStatusChecking() {
        int port = NetUtils.getAvailablePort();
        for (int i = 0; i < 1000; i++) {
            assertFalse(NetUtils.isPortInUsed(port));
        }
    } 

[zrlw]

Enabling SO_REUSEADDR prior to binding the socket using bind(SocketAddress) allows the socket to be bound even though a previous connection is in a timeout state.

[AlbumenJ]

Are there any systems which cannot support such configuration or will throw exception here?

[zrlw]

if an error occurs, or the socket is closed, SocketException will be thrown here.

[zrlw]

i see. Embedded, real time system, or old Unix ( before 1980?) might not support SO_REUSEADDR. added condition check before setting it.

[AlbumenJ]

Suggested change

	} catch (IOException e) {
	} catch (Throwable ignore) {

[zrlw]

done.

[zrlw]

there are still somethings need to be revised, the port 39119 opened by NettyTransporterTest#shouldAbleToBindNetty4 does not be closed after testing.
[updated] fixed it.
https://github.com/apache/dubbo/actions/runs/13937299977/job/39007599423