A second implementation of what? The code Marcus mentions also already exists, and is in k5ssl. (Part of rxk5.)
Matt ----- Original Message ----- From: "Jeffrey Altman" <[email protected]> To: [email protected] Sent: Thursday, June 11, 2009 1:15:04 AM GMT -05:00 US/Canada Eastern Subject: Re: [OpenAFS-devel] configuration files for servers Marcus Watts wrote: > Simon Wilkinson just asked for "configuration files for servers" > at the best practices workshop. > > I also talked briefly with Simon about this the night before last. > > As it happens, the k5ssl code can parse krb5.conf files today. > This may not be exactly the file format that people would like, > but it may be "close enough". > > in any of the rxk5 branches, look at > src/k5ssl/k5s_cf.c > > there isn't anything that writes these files back out. > I'm not convinced this is a good idea, but if people > really want that - it is possible. I think I > have a test program that does this, but I don't seem > to have bothered to save it with k5ssl. It is > trivial to walk the in-core configuration file contents, > and to write it out in whatever format. > src/k5ssl/t_cf.c > is sample code to do just that. > > -Marcus The Heimdal implementation is BSD licensed and can be utilized by OpenAFS. I would rather not have a third implementation that needs to be supported. There are too many undocumented *features* in the krb5 profile format. However, given the discussion that has taken place recently on this list I believe that the krb5 profile format is the one we should go forward with. Jeffrey Altman _______________________________________________ OpenAFS-devel mailing list [email protected] https://lists.openafs.org/mailman/listinfo/openafs-devel _______________________________________________ OpenAFS-devel mailing list [email protected] https://lists.openafs.org/mailman/listinfo/openafs-devel
