Hello Campers!
Well, after some four day's of work I've got a rudimentary installation of O-X working on a 6.0.1-01 server.
Now, before you all get terribly excited it's important to realize that the installation procedure I have developed is really dirty - not at all the kind of thing that one would entrust to a working server. However, I get the O-X logon screen - a step in the right direction at least. Over the next couple of days I'll be looking at the configuration side of things to get things working as opposed to just sitting there presenting a logon screen!
Normally I would simply publish such work and let folk do what they want to do with it. However, judging from the correspondance concerning O-X in these fora, I think that a more structured approach would yield best results.
In the first instance I think that an experienced e-smith developer should have a look at what I've done because there are places where I've removed e-smith distribution RPMs and replaced the code with the results of 'make install' steps. While (possibly) not breaking any integrities, it would be far better if new RPMs were developed so that the installation of O-X could be reversed.
Secondly, there's so much code that needs to be downloaded and shoe-horned into the system that it would be better if the collection were made once and then put together into an FTP site easily available to users of SME-server. By this means the natural development of component parts would not have any effect on this installation unless the installer particularly wished to try new, updated components.
Finally, I guess that a 'proper' installation of O-X would become a popular add-on to SME server and so it would benefit everyone if the SME developers had a full say on how the O-X install is presented.
So, with these thoughts in mind, any volunteers to do some testing?? If so, I'll clean up my installtion script and send it to volunteers. I'm sure that some targetted input can quickly get this rather inexpertly written script transformed to something useful in a matter of a couple of weeks.
In the meantime I'm going to apply my newly developed script to a newly loaded 6.0.1-01 server and iron out any faults that I've introduced during the experimental build process. I guess this will take a day or so and so I estimate that the first 'working' version of the script will be available for test volunteers by Monday 21 February, 2005
Regards,
Dave