The java/net/httpclient/HttpClientLocalAddrTest.java has been observed failing relatively frequently in timeout - and the log shows a Carrier thread being pinned. Whether that's the root cause of the test failure is hard to say, but we should fix the code to avoid pinned threads. This is one path that had escaped my notice when I went through the HttpClient code before.
------------- Commit messages: - 8316031 Changes: https://git.openjdk.org/jdk/pull/15761/files Webrev: https://webrevs.openjdk.org/?repo=jdk&pr=15761&range=00 Issue: https://bugs.openjdk.org/browse/JDK-8316031 Stats: 7 lines in 2 files changed: 2 ins; 0 del; 5 mod Patch: https://git.openjdk.org/jdk/pull/15761.diff Fetch: git fetch https://git.openjdk.org/jdk.git pull/15761/head:pull/15761 PR: https://git.openjdk.org/jdk/pull/15761
