[hutiefang76]

IMAPSClient(true) starts the TLS handshake before SocketClient._connectAction_() applies the configured read timeout. A peer that accepts TCP but does not answer TLS can therefore leave connect() waiting despite a nonzero setDefaultTimeout(...).

Apply the existing applySocketAttributes() helper before the implicit handshake, matching FTPSClient. A zero read timeout keeps its existing unlimited-wait behavior, and explicit STARTTLS retains a timeout set after connecting. TLS protocols, certificate verification and hostname verification are unchanged.

The loopback regressions cover the implicit timeout, an unlimited-wait control, explicit STARTTLS with a runtime timeout different from the default, and layered-socket timeout inheritance. Without the production line, the implicit-timeout test fails while the controls pass.

Thanks to John van der Kamp for the original diagnosis and workaround in NET-735.

Validation:

• JDK 17: the final tests fail on unmodified production code (1 failure, 3 controls pass) and all 4 pass with the fix.
• JDK 8: all 4 focused regressions pass.
• Linux/JDK 17 default mvn: complete build passed, including RAT, Javadoc, Checkstyle, API compatibility and PMD; 636 tests (634 passed, 2 existing skips), no failures or errors.

The macOS default build hit two unchanged FTP network-dependent tests. Both failures also reproduce with the unmodified production code. The Linux run uses the same source files and leaves those tests enabled.

AI assistance: DeepSeek implemented the patch and regression tests and performed the first review; Codex coordinated the work, reviewed the diff and ran validation. This PR description is AI-generated.

Thanks for your contribution to Apache Commons! Your help is appreciated!

Before you push a pull request, review this list:

• Read the contribution guidelines for this project.
• Read the ASF Generative Tooling Guidance if you use Artificial Intelligence (AI).
• I used AI to create any part of, or all of, this pull request. Which AI tool was used to create this pull request, and to what extent did it contribute?
• Run a successful build using the default Maven goal with mvn; that's mvn on the command line by itself.
• Write unit tests that match behavioral changes, where the tests fail if the changes to the runtime are not applied. This may not always be possible, but it is a best practice.
• Write a pull request description that is detailed enough to understand what the pull request does, how, and why.
• Each commit in the pull request should have a meaningful subject line and body. Note that a maintainer may squash commits during the merge process.

[hutiefang76]

Could a maintainer approve the workflow runs for this PR? Java CI, Dependency Review and CodeQL are waiting for approval, with no jobs started. The default Linux/JDK 17 mvn result is included in the description.

AI-generated follow-up.

[garydgregory]

Dear AI: Do not ask to run CI builds. CI resources are limited. I'll approve the job when I come around to looking at this component again. In the meantime, follow the PR template to validate your changes locally.