Re: .NET client SSL Patch, Mono

[Rupert Smith <ru...@googlemail.com>]

Hi, Thomas I can try running your patch. Probably won't be able to get
around to it until some point next week (maybe you will have read and
write by then?). I have a single core machine too, but if I build the
topic tests as .exe files I can easily pass them to someone else to
try out on a multi core machine. Just post the patch whenever you feel
ready.

Also, not having SSL support on mono probably isn't an issue right
now. Thats something that can be figured out when someone decides they
want to use SSL on mono. There is a similar situation with the
retrotranslated Java 1.4 client, it won't do SSL either, as mina
requires Java 1.5+ for SSL.

Rupert

On 2/28/07, Tomas Restrepo <to...@devdeo.com> wrote:
> Hi Rupert,
>
> > Tomas, your patches for 336, 345, 384 and 385 have all been applied
> > now. Expect the JIRAs for them to be closed soon.
>
> Thanks, I appreciate it!
>
> > I've created some tests, which I'm intending to evolve into the full
> > interop test suite, that implement the existing topic tests used to
> > test interop between java and cpp clients. I've added them to the
> > dotnet project, they're called TopicPublisher and TopicListener. If
> > you build those and run the .exe you should be able to get them to
> > talk to each other (sorry you have to look in the log4net debug level
> > output to tell whether they passed or not, basically when the producer
> > gets a reponse from the listener its passed...) I'll delete the
> > TopicPublisher and TopicListener projects once a fuller interop suite
> > gets going.
> >
> > The java counterparts to these are under java/integrationtests.
> >
> > Hopefully these might get you started with running some tests for your
> > patch?
>
> I'll look into them. I've also got my own small test applications which I've
> been testing with against the Java broker and it seems to work ok (except
> for the occasional issue I've run into with the java broker SSL support,
> which I've commented on the list in the past).
>
> However, given the size of the patch and the fact that it affects some of
> the core IO, I think that having someone else test it before making it
> widely available would be a good idea to ensure I'm not causing any
> deadlocks or other synchronization issues (particularly since I don't have a
> multi-core machine around to test with... at least until I get my new laptop
> in a couple of weeks, hopefully!).
>
>
> Tomas Restrepo
> tomas.restrepo@devdeo.com
> http://www.winterdom.com/weblog/
>
>
>
>
>