Re: Weird JavaMail problem: it just keeps getting weirder

Well, after a day in the wonderful world of truth, justice, and Fess Parker

(i.e., jury duty), I'm back running experiments on JavaMail, and it just
keeps 
getting weirder.

It seems that the debug output is absolutely
identical between the old 
production system (where it was always working
fine) and the V4R5 box (where it 
was blowing up). I tried adding a second
"Transport.send," turning the debug 
flag off after the first one. It worked
properly. Then I tried a version in 
which I was back to one
"Transport.send," but with the debug flag on the 
"session" EXPLICITLY turned
off. That worked fine, too. Now, I'm about to try 
it with my own
"System.out" diagnostics pulled out (but the explicit 
"setDebug(false)" left
in); hopefully that will work, too.

Dunno what it could have been, but it
sure is weird. Sort of like the 
occasional cases I've seen where a perfectly
valid MI source member blows up 
QPRCRTPGM, but the same statements in
slightly different (but functionally 
equivalent) order compiles
fine.

--
J.Lampert