Richard Monson-Haefel wrote:
> > Interesting ... I was just reading the RMI/IIOP forum and it appears that
> > propagation of transaction and security contexts are not
> > supported in the Sun-IBM
> > version of RMI over IIOP, which I believe is the official
> > version. In fact, it
> > appears that there is no way for the current implementation to
> > track specific
> > references (they are arbitrary to the server object), which seems
> > very bizarre.
and:
> IMHO, as long as the vendors follow the RMI/IIOP specification they can
> implement it anyway the want, which means they can opt to support transaction
> and security context passing even know the reference implementation provided by
> Sun/IBM does not.
The key is that there is no "official" version. The fact that Sun (with
partners) is producing a "reference implementation" in no way obligates
anyone to make use of that reference implementation. Some vendors will use
what they've produced, some will extend it, some will ignore it in favor of
their own implementations.
The only official thing is the specification, to which all implementations
should conform.
Marc San Soucie
GemStone Systems, Inc.
Beaverton, Oregon
[EMAIL PROTECTED]
===========================================================================
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff EJB-INTEREST". For general help, send email to
[EMAIL PROTECTED] and include in the body of the message "help".
