Re: Access Control Problem: Status

[Patrick Tullmann <tullmann@cs.utah.edu>]

Craig Milo Rogers wrote:
> 	I've done an awful lot of banging my head agai.., er,
> debugging the access control problem.  My initial hypothesis, that the
> wrong classloader was being used, is disproven..

Actually, I'm glad it wasn't the classloader.  Those sorts of problems
are awful.  :)


> I'm guessing that the JDK1.2.2 we're using on son.isi.edu (and
> elsewhere) is generating an access violation to a public member of
> edu.utah.janos.nodeos.Node that it inherited from
> edu.utah.janos.nodeos.pj_Node.

This sounds plausible.  However, we use JDK 1.2.2 on FreeBSD 3.x here
for lots of our development and testing, so I would have thought that
we'd see this problem.  Here's the output of 'java -version':
	% java -version
	java version "1.2.2"
	Classic VM (build Linux_JDK_1.2.2_RC4, green threads, sunwjit)
	%

(Its running under Linux emulation on FreeBSD.)  Is your version
different in any respect?

You might also try "testing" the hypothesis by replacing the body of
NodeStat.time() with 'return 0'....  Another test would be to build a
simple testcase that demonstrates the problem (and it would be handy
for testing our VM :).

We would have a lot of code to fixup in that case (both in Node.java
and in Debug.java).

I will be out of town until Monday morning, so I can't look into this
any closer until then.

-Pat

----- ----- ---- ---  ---  --   -    -      -         -               -
Pat Tullmann                                       tullmann@cs.utah.edu
	   "I'd kill for a Nobel Peace Prize." -- S. Wright