Compiling 1.2.5 on True64 v5.x

2002-07-08 Thread Peter Losher
Has anyone encountered an issue compiling Kerberos 1.2.5 on a Tru64 {5.0 or 5.1) box using Compaq's C compiler? Any pointers would be welcome. Thanks - Peter (Error message below) -=- making all in util/pty... if cmp /tmp/krb5/dist/src/util/pty/libpty.h ../../include/libpty.h >/dev/null 2>&

Re: /var/tmp/krb5_RC*

2002-07-08 Thread Tom Yu
> "ajk" == Andrew J Korty <[EMAIL PROTECTED]> writes: ajk> What might cause an MIT krb5kdc process to dump files in /var/tmp ajk> about 20k in size every 40 minutes or so? The filenames begin ajk> with "krb5_RC" (so I assume they have to do with the replay ajk> cache), followed by the krb5kd

/var/tmp/krb5_RC*

2002-07-08 Thread Andrew J. Korty
What might cause an MIT krb5kdc process to dump files in /var/tmp about 20k in size every 40 minutes or so? The filenames begin with "krb5_RC" (so I assume they have to do with the replay cache), followed by the krb5kdc process id, followed by a three-letter uniquifier. The files persist across

Re: How does gsscred work?

2002-07-08 Thread carcassone_fr
So one should not expect KDC output when running "gsscred -a"? [EMAIL PROTECTED] wrote in message news:<[EMAIL PROTECTED]>... > The gsscred command makes GSS/Kerberos names out of usernames. > That's all it does. > > So what could there be for the KDC to do for gsscred? Principal > name validat

Re: Kerberos authentication for Web Services

2002-07-08 Thread Nicolas Williams
On Mon, Jul 08, 2002 at 03:20:36PM -0400, Ken Hornstein wrote: > I believe Sam Hartman already pointed out that generating a new network > protocol to communicate with the KDC is a Really Bad Idea. In general, > that part of Kerberos is supposed to be invisible to you. You could > do that (I bel

Re: Kerberos authentication for Web Services

2002-07-08 Thread Ken Hornstein
>I am interested in building a system (similar to Microsoft's .Net My >Services) that is a family of web services that clients authenticate >against using Kerberos. The idea is to have clients hit the KDC via >SOAP calls over SSL and get the ticket. Then they ask the KDC for a >ticket to communica

Re: Kerberos authentication for Web Services

2002-07-08 Thread David Magda
[EMAIL PROTECTED] (jeremy redburn) writes: [...] > But this doesn't seem to fit into the idea of how Kerberos > authentication works. Is anyone doing Kerberos authentication via SOAP > calls? What do people recommend for an authentication mechanism for a > family of web services? Try the follow

Re: Kerberos authentication for Web Services

2002-07-08 Thread Frank Balluffi
I am not working on Kerberos authentication via SOAP, but it is my understanding that IBM's and Microsoft's WS-Security adds XML Signature and XML Encryption to SOAP and that authentication is implicit in each message, and that you will need to explicitly send KRB_AP_REQ and KRB_AP_REP message

gss-api

2002-07-08 Thread Rick
A little trouble understanding kerberos GSS-API. My KDC is Windows 2k host1 is solaris 7 host2 is NT 4 with leash32 2.1.1.1 On my 2k KDC I defined the following principals. ktpass -princ [EMAIL PROTECTED] -mapuser user1 -pass password -out 1.keytab ktpass -princ [EMAIL PROTECTED] -mapuser us

Kerberos authentication for Web Services

2002-07-08 Thread jeremy redburn
I am interested in building a system (similar to Microsoft's .Net My Services) that is a family of web services that clients authenticate against using Kerberos. The idea is to have clients hit the KDC via SOAP calls over SSL and get the ticket. Then they ask the KDC for a ticket to communicate wi

RE: How does gsscred work?

2002-07-08 Thread Nicolas . Williams
The gsscred command makes GSS/Kerberos names out of usernames. That's all it does. So what could there be for the KDC to do for gsscred? Principal name validation? A principal name might be valid one moment, and later not (e.g., a user might be fired/layed off or quit soon after you create a gss