RFR: JDK-8066859 java/lang/ref/OOMEInReferenceHandler.java failed with java.lang.Exception: Reference Handler thread died (original) (raw)

Laurent Bourgès bourges.laurent at gmail.com
Thu May 7 18:06:48 UTC 2015

Peter,

Thanks for long and detailled answer.

I know now better why OOME should not happen. However any application may also use phantom references and the ReferenceHandler thread will call Cleaner.run () which could catch OOME from the application code implementing thunk.run (). Am I right ?

If this block also throws a new oome2 due to the first oome1 (no memory left), it will work but I would have prefered a more explicit solution and check oome1 first ...

I looked back at your patch and it is fine. Howevdr I wonder if it would be possible to avoid any allocation in the catch(Throwable) block:

You have a point and I asked myself the same question. The question is how to treat OOME thrown from thunk.run(). Current behavior is to exit() JVM for any exception (Throwable). I maintained that semantics. I only added a handler for OOME thrown in the handler of the 1st exception. I might have just exit()-ed the VM if OOME is thrown, but leaving no trace and just exiting VM would not help anyone diagnose what went wrong. So I opted for keeping the VM running for a while by delaying the handling of 1st exception to "better times". If better times never come, then the application is probably stuck anyway.

Seems very good after a 2nd look. However, it could loop for a while if no more memory left ? For example: oome1 => oome2 (catch) => throw x=> oome2 (catch) => ....

An alternative would be to catch OOME from thunk.run() and ignore it (printing it out would be ugly if VM is left to run), but that would silently ignore OOMEs thrown from thunk.run() and noone would notice that Cleaner(s) might not have clean-ed up the resources they should.

I am a bit lost but I like logging such exceptional case but if no allocation can happen, how to ensure logging such case anyway ? ...

Anyway. If none of the Cleaner.thunk's run() methods can throw any exception, then my handling of OOME is redundant and a code-path never taken. But I would still leave it there in case some new Cleaner use comes along which is not verified yet...

Agreed. It is always better to handle such exceptional case if you can at least log them...

Best regards, Laurent

More information about the core-libs-dev mailing list