oehme commented on PR #1281: URL: https://github.com/apache/maven-mvnd/pull/1281#issuecomment-2705073602
Turns out the real issue was in the `tryLock` method, which just silently returned `null` when the file couldn't be locked. Also today I learned that `try-with-resources` will happily accept `null`, so this was never spotted. The end effect was that there was no concurrency control, all processes would access the file without coordination. I've added a process-based test and fixed the tryLock method so that it does the retry loop until it gets a non-null lock or the timer runs out. In fact, I could revert all the other changes, the tryLock turned out to be the only true issue. To me, DataInputStream/DataOutputStream seems more straightforward than the memory mapped buffer. Whatever you prefer. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: issues-unsubscr...@maven.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
