> > In my event callback function I'm trying to null-out the ca_next
> portion
> > of the nh_auth portion of the nua_handle_t that's passed in but my
> code
> > sense isn't finding the nh_auth param of the struct. Am I allowed
to
> > access that structure or is it private?
>
> The structure is p
2009/1/9 Andrew Rechenberg Lists :
>> OK, parser was just configured to reject multiple WWW-Authenticate
>> headers, probably copy&paste error from HTTP. I've pushed a patch to
>> darcs...
>
> Thanks. I applied that patch and now I'm seeing both headers, but my
> app is misbehaving now :(
>
> In m
>
> 2009/1/8 Pekka Pessi :
> > There is nothing wrong here, but it looks like parsing two
> > WWW-Authenticate headers just fails. *Blush*. I'll have a peek in
the
> > parer
>
> OK, parser was just configured to reject multiple WWW-Authenticate
> headers, probably copy&paste error from HTTP.
2009/1/8 Pekka Pessi :
> There is nothing wrong here, but it looks like parsing two
> WWW-Authenticate headers just fails. *Blush*. I'll have a peek in the
> parer
OK, parser was just configured to reject multiple WWW-Authenticate
headers, probably copy&paste error from HTTP. I've pushed a pat
2009/1/8 Andrew Rechenberg Lists :
>> Perhaps there is a problem parsing the Kerberos header? Could you send
>> a trace containing such a WWW-Authenticate header.
> Here's a snipped trace (also available at
> http://dpaste.com/hold/106771/)
> SIP/2.0 401 Unauthorized
> Date: Wed, 07 Jan 2009 2
>
> Perhaps there is a problem parsing the Kerberos header? Could you send
> a trace containing such a WWW-Authenticate header.
>
Here's a snipped trace (also available at
http://dpaste.com/hold/106771/)
tport_recv_event(0x1044830)
tport_tls_recv(0x1044830): tls_read() returned -1
tport_tls_set
2009/1/8 Andrew Rechenberg Lists :
>> Yes, if there are several WWW-Authenticate challenges they are linked
>> via au_next.
>
> For some reason the NTLM WWW-Authenticate header is the only one that it
> finds. The Kerberos header is not in the au_next. Any reason why this
> would be the case?
Pe
> >
> > Could I add the "Supported" and "ms-keep-alive" portions of that
> header?
>
> Sure, SIPTAG_SUPPORTED_STR("gruu-10, adhoclist"),
> SIPTAG_HEADER_STR("ms-keep-alive: uac;hop-hop=yes")
>
Great!
>
> You need 3 request-responses for authentication? Or should we include
> something in the
2009/1/5 Andrew Rechenberg Lists :
>> > There are some other things that need to get added to the header for
>> > OCS. Can I just use TAGs to add those? Can I add those tags in
>> > nua_authenticate?
>>
>> That depends on the things to add? You can add SIP headers with
>> nua_authenticate(), but
> > There are some other things that need to get added to the header for
> > OCS. Can I just use TAGs to add those? Can I add those tags in
> > nua_authenticate?
>
> That depends on the things to add? You can add SIP headers with
> nua_authenticate(), but the extra tags are not passed to auc*
>
2009/1/5 Andrew Rechenberg Lists :
>> As far as I can recall the NTLM code is just copied from the digest
>> code but there is actually no real functionality there.. It should
>> probably be deactivated, or better, completed...
>
> I've started working on completing the NTLM code. The Pidgin SIPE
>
> 2008/12/31 Andrew Rechenberg Lists
> > I'm trying to send an NTLM Authorization header to a Microsoft
Office
> > Communications Server 2007 and my code is getting a EXC_BAD_ACCESS
in
> > ca_credentials when using the macro AUTH_CLIENT_IS_EXTENDED. This
> > function seems to be called indirec
2008/12/31 Andrew Rechenberg Lists :
> I'm trying to send an NTLM Authorization header to a Microsoft Office
> Communications Server 2007 and my code is getting a EXC_BAD_ACCESS in
> ca_credentials when using the macro AUTH_CLIENT_IS_EXTENDED. This
> function seems to be called indirectly when I u
13 matches
Mail list logo
