Re: Incremental propagation when KDCs are clients of a different realm

2015-11-09 Thread Toby Blake
> On 7 Nov 2015, at 00:04, Benjamin Kaduk wrote: > [...] > Did you investigate putting a domain_realm mapping to KDCDOMAIN in the > alternate krb5.conf during your testing? I would expect that to allow the > krb5_sname_to_principal behavior to be changed. Ach, of course, yes, that does indeed

SPNEGO question

2015-11-09 Thread Pascal Jakobi
I am still testing kerberos pretty thoroughly. Now I am at SPNEGO. I was able to have it to work (with firefox) when calling simple URI such as http://host.domain.tld but not when calling http://host.domain.tld/test_dir. I did change the negotiate URI field in firefox configuration, but did n

Re: SPNEGO question

2015-11-09 Thread Todd Grayson
No, the path failing is something application side within your setup. The configuration of the FQDN really just the domain and tld) is all you need, that is host.domain.tld adding the path should not break things in the browser configs for example in environments where many hosts use Negotiat

Re: SPNEGO question

2015-11-09 Thread Rick van Rein
Hi Pascal, > I was able to have it to work (with firefox) when calling simple URI > such as http://host.domain.tld but not when calling > http://host.domain.tld/test_dir. That surprises me. I've been putting host.fqdn.names and .domain.names into the network.negotiate-auth.trusted-uris field in