Message-ID: <gh-pgjdbc-pgjdbc-3886-c3611459558@github.com>
From: "davecramer (@davecramer)" <noreply+davecramer@github.com>
To: "pgjdbc/pgjdbc" <noreply+pgjdbc-pgjdbc@github.com>
Date: Thu, 04 Dec 2025 10:44:40 +0000
Subject: Re: [pgjdbc/pgjdbc] PR #3886: fix: avoid memory leaks in Java <= 21 caused by Thread.inheritedAccessControlContext
In-Reply-To: <gh-pgjdbc-pgjdbc-3886@github.com>
References: <gh-pgjdbc-pgjdbc-3886@github.com>
List-Id: <gh-pgjdbc-pgjdbc.github.com>
X-GitHub-Author-Login: davecramer
X-GitHub-Comment-Id: 3611459558
X-GitHub-Comment-Type: issue_comment
X-GitHub-Issue: 3886
X-GitHub-Repo: pgjdbc/pgjdbc
X-GitHub-Type: comment
X-GitHub-Url: https://github.com/pgjdbc/pgjdbc/pull/3886#issuecomment-3611459558
Content-Type: text/plain; charset=utf-8

Using claude it suggested 
```
ForkJoinPool.commonPool().execute(() -> {
    RefQueueBlocker<Object> blocker =
        new RefQueueBlocker<>(queue, threadName, threadTtl, () -> !checkEmpty());

    while (!checkEmpty()) {
        try {
            ref.onClean(true);
            ForkJoinPool.managedBlock(blocker);

            Node<?> ref = (Node<?>) blocker.drainOne();
            if (ref != null) {
                ref.onClean(true);
            }
        } catch (InterruptedException e) {
            if (checkEmpty()) {
                LOGGER.log(Level.FINE,
                    "Cleanup queue is empty, and got interrupt, will terminate the cleanup thread");
                break;
            }
            LOGGER.log(Level.FINE, "Ignoring interrupt since the cleanup queue is non-empty");
        } catch (Throwable e) {
            LOGGER.log(Level.WARNING, "Unexpected exception while executing onClean", e);
        }
    }
});
``` 
as a simplification to the try catch logic

Also had this to say

5. Missing edge cases
- What happens if ForkJoinPool.commonPool() is unavailable (custom security manager)?
- No handling for RejectedExecutionException from execute()
- The threadFactory.newThread() can return null but only logs a warning
-