> > That does look a bit like magic - and as is 99.9% the case when > inexplicable magic happens, memory is getting corrupt or overwritten. You > are obviously prudent about this, but is there no clue when running under > Valgrind or similar for OSX? Nothing out of the ordinary or warnings? > Memory corruption seems unlikely: I would expect more errors to follow once some piece of memory is corrupted. Valgrind is unfortunately not very useful in this case as its OS X support is limited and it doesn't like JIT very much.
(I'm unfortunately not strong on the JAVA VM, so another approach might be > needed to achieve the checking...) > After further testing I'm more and more leaning towards blaming the JVM here. In particular I've seen some weird behaviors related to captured variables, which could conceivably explain why the wrong value was bound to the UPDATE but both the log line and the INSERT used the correct one. I haven't been able to reproduce the issue after ripping out lambdas from the java snippet in my previous email. I'll report back if the issue resurfaces at some point in the future.
