Hello, > > Ok, I see. But then, how should the repo file look like if you create a repo like this? > createrepo --split -o dir1 dir2 > > As the documentation for --split says, it is used for "split > media". Ie. where you have more data than will fit on a single CD/DVD, > at which point anaconda has some magic which makes this work. > If you are just accessing this from the network you'll want to create > a single repo. ... if you are doing both, I think you want to run it > twice (probably via. some other tool for the media creation bit -- > maybe someone who does CD install disks can answer?) Thanks for your answer, James. I understand now. But then, what would be the repo file I should use if I create a repository in this way? createrepo -i list1 -o dir1 main_dir createrepo -i list2 -o dir2 main_dir The thing is that I'm wondering whether it makes sense to do this in yum so that main_dir contains all the packages of my release but in dir1 and dir2 I have different views of the repository which actually contain some of the packages of main_dir (as specified in list1 and list2). Does this make sense? How should I define my repo file? If I do this: [repo] name=my repo baseurl=http://whatever/dir1, http://whatever/main_dir I get some error messages as explained in my original mail. But if I don't specify main_dir, the installation fails since yum is not able to find the physical packages. Thanks a lot for your help, Maria _______________________________________________ Yum mailing list Yum@xxxxxxxxxxxxxxxxx http://lists.baseurl.org/mailman/listinfo/yum
