[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: Installing core system -- unresolved dependencies

Sergei Mikhailov wrote:


As an experiment, I am trying to install the core system in a separate directory, call it /newroot. I first create a new rpm database and then run the rpm against that database with --root specified as /newroot. As my package list I pass all the rpms specified in the group-id "core" of the comp.xml. However, this install fails with failed dependencies. Some packages that are part of the core system need packages that are not included in that core. My question is: is the set of packages in the core system not "self contained"? And what would be the correct way to install an absolute core system?

I'm wondering too why the 'core' group has unresolved dependencies - I would have thought that this group was as you say 'self-contained'. Adding the dependencies returned by getfullcomps.py soon adds up to a not-as-compact 'core' installation.

How does Anaconda install the absolute minimum set? Which file in the source does it?

Not 100% sure - but I think installclass.py should give you some idea of how packages are grouped for installation ...


Before you criticize someone, walk a mile in their shoes ...
That way when you do criticize them, you're a mile away and you have their shoes!

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]