The test searches for "JShellToolProvider" in the main thread's stack trace, which is pulled from an SA heap dump. Typically the main thread is blocked in Object.wait(), so SA can determine its stack trace. However, the wait has a 100ms timeout, so the thread does periodically wake up and does a few things before waiting again. If SA tries to get the stack trace while the thread is active, it may not be able to, and this causes the test to fail. I determined this only happens about 1 in 5000-10000 runs. So as a fix I'm allowing the test to do one retry. That should make it extremely unlikely that we ever see this failure again. I also bumped up the amount of time the test waits before doing the heap dump from 2 seconds to 4 just to make absolutely sure the main thread is fully started before doing the heap dump.
------------- Commit messages: - Retry heap dump if main thread stack trace is not present. Changes: https://git.openjdk.java.net/jdk/pull/6795/files Webrev: https://webrevs.openjdk.java.net/?repo=jdk&pr=6795&range=00 Issue: https://bugs.openjdk.java.net/browse/JDK-8269556 Stats: 26 lines in 1 file changed: 19 ins; 0 del; 7 mod Patch: https://git.openjdk.java.net/jdk/pull/6795.diff Fetch: git fetch https://git.openjdk.java.net/jdk pull/6795/head:pull/6795 PR: https://git.openjdk.java.net/jdk/pull/6795
