On Tue, 2006-09-19 at 06:31 -0500, Jasper Hartline wrote: > I am working with some prototype installer Anaconda/Kadischi discs.. > but to make that happen, Anaconda has been modified just a tiny bit. > > I have a patch here that deals with a make dev/root node that if it > recieves an OSError it will pass > instead of fail, which is required inside the LiveCD environment. I'm > not absolutely sure this is the > right way to handle this scenario however. Any thoughts? The sentiment is reasonable -- I went for checking if it exists and only creating it if not. That way, we won't miss exceptions for other reasons (read-only filesystem due to hardware failure, other mysterious bogons) > I also have a patch that makes Anaconda not fail, when trying to invoke > syslogd when it is already > running in teh LiveCD environment, and isn't in the same path as > Anaconda would be looking for > in a normal case scenario. Please look at the patches, and see if we can > do something proper here > to allow Anaconda to be spawned from a working LiveCD system. Similarly, checking for existence and only running then rather than catching all exceptions globally Jeremy
