OpenUru.org

As Christian said earlier, I don't think you'd get as far as submitting for user authentication if you had the wrong keys. If you're using "good" credentials then I'm sort of inclined to think that your client is constructing the wrong digest to send to the server, but I'm not sure how.

Murf wrote:I suppose the obvious question, is where is the source code for the current, 1.912 build of the URU Live again client?

That source code is not available right now. It will turn up again (debinkified) at https://bitbucket.org/OpenUru_org/cwe and http://foundry.openuru.org/hg/CWE/ when we’re done sorting things out. Hopefully still this weekend.

https://foundry.openuru.org/hg/CWE-ou-NOBINK/ is an early attempt at debinkification and should not be used anymore. https://bitbucket.org/OpenUru_org/cwe-ou has the correct commits, but no Bink stubs yet, and is not quite ready yet to receive pull requests etc.

To debug your authentication failure, the Wireshark plugin may be helpful.

Also, I recommend not testing custom clients against MOULa, but setting up your own server for that. The MOULa server is flaky enough as it is and I’d personally avoid doing anything to it that has a remote potential for upsetting it even more (but maybe I’m just being paranoid).

Your linguistic inventions are astounding, Christian ... 'debinkified' or 'debinkification' ...

Those seem worth an entry in the 'Urban Dictionary' or sth similar ... When they keep playing this 'game'
with more companies/institutions/persons etc. this might prove to be quite useful as they come with and
contain a lot of context and meaning ...

What an awesome (arduous and painstaking) job you're all doing here, esp. you, Christian ...

Right along with the unicodiness!

Thanks for all that information everybody, I will just wait till the new debinked source shows up and have a try with that.
Murf.

Hi all,
Well I solved my authentication error problem, it helps to not change the build number of your client if you expect to happily connect to the URU Live server!
Beyond that I now face the 'wrong version of PhysX SDK" issue, and I am unable to find a version 2.6.0 required for the build.
The archives at http://www.nvidia.com/object/physx_archives.html#SDK only go back to v2.6.4, so if anyone has an active link for v2.6.0 can they please post it here.
Thanks,
Murf.

PhysX 2.6.0 is no longer available. However, if you compile with Skoader's physics pull request, then 2.6.4 should work just as well as the older version.

oh thanks!

Murf wrote:Well I solved my authentication error problem, it helps to not change the build number of your client if you expect to happily connect to the URU Live server!

Yes, at least the leading '1' part of '1.912' needs to be retained - it's used as a kind of API version mark to ensure that the client and server are compatible (I recall that there's actually some mention of that in the source round about where that value is set).

Both the branch and build ids must match the server's value, or you will be kicked off.