RE: Communications recovery and related problem

Mike,

>
> I'll take a peek in the code and see what I can come up with that might
> cause the change in the references.  As to not always finding the
> references that may be intentional.  That is, to get around a "hangish"
> difficulty when starting up Designer will not attempt to initiate a host
> connection unless one has already been started.  So if you have ended all
> your host connections (by ending codeedit and the comm daemon for
> instance) then Designer will load your source as if the host connection is
> not there.  The alternative was to hang for the TCP/IP timeout waiting for
> a connection to be made in the times when you are disconnected from the
> network.  Yes, in theory a better solution could have been done but there
> are other reasons as well for this one.  So, to get around this you could
> go to the open dialogue in Designer, expand the entry for the iSeries of
> your choice, close the open dialogue then open your file.  This only has
> to be done once, unless you end communications of course...

Great!  I'll do that.

>
> The open/save timer dialogue that doesn't close is one of those things
> that _never_ happens when one is trying to make it happen, particularly
> with the debugger. It is on my list of things to fix but it is way, way,
> way down there.  Hitting Cancel is OK as long as the file is loaded/saved.
>

Good to know.

>
> >exact and reproducible?!  I'm sure those bugs have been found already.  I
> >would definately do this if I could reproduce it myself!  It usually
> happens
> >when a deadline is near, and/or after I've made alot of changes, and
> always
> >when I least expect it.  If that's any help!
>
> Nope, they are not all found.  Another fella here just found one today,
> cause whilst loading a certain incantation of bad DDS.  I should add that
> putting in the code to detect the nearness of deadlines wasn't that
> difficult as someone is always near a deadline but
> CrashWhenYouLeastExpectIt() was really tricky to write, as well as debug
> and test because it wouldn't always work.  This was probably because
> during debug and testing we were, well, always _expecting_ it.  It is good
> to hear that it is working out there in the "real" world :-)

I knew it!!!! :)

Phil


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com