On Wed, Oct 13, 2010 at 9:20 PM, James Laska <jlaska@xxxxxxxxxx> wrote: > I was attempting just that earlier this week following instructions at > https://fedoraproject.org/wiki/How_to_build_a_Rawhide_ISO_image_for_testing > > I had success creating the ISO files, but they failed to boot due to > 'unable to find /init'. I haven't had time to dig deeper to investigate > what I did wrong. Those instructions would seem to pull everything from the rawhide repo for the build? What I do is use the appropriate .cfg file to define the way in which creation of the chrooted area for a development/f14 build will be set up when you start the mock creation command - and then as a second step go into the chrooted shell where the pungi command is then run using a kickstart file to pull in the same repo for packages and define which packages will be merged into the iso during the build. It is possible to add additional repo definitions in the .ks file and include any number of additional packages from any of the repos - however in this process anaconda would come from the main repo unless the updates-testing repo was also defined ( which case all packages in updates-testing would be pulled in as more up to date than those in development/f14/ ), but I don't know enough about the way pungi works to know if it is possible to tell the build to pull in a single additional package and if so how to put the appropriate lines into the .ks file that the build uses.... maybe this is beyond what the system was designed to do? -- mike c -- test mailing list test@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe: https://admin.fedoraproject.org/mailman/listinfo/test
