On 03/30/2010 08:35 PM, Chris Lumens wrote:
anaconda exiting should signal to loader/init that it's time to shut
down and that will kill X for us, yeah?
Unfortunately not quite. When exit anaconda with a non-zero error code,
init does the delayed reboot thing and doesn't start killing processes
until user enters c-a-d, so debugging of the running programs is still
possible.
BTW there's a separate debate whether e.g. upon reporting exception to a
bugzilla we should still wait for c-a-d or whether that is just a
correct termination of the install process. It's bug 525615. I prefer to
keep the delayed reboot in this case because it is an exception after
all and we might still want to look at the terminals, use shell, etc.
I think I will just switch back to tty1 from the init/loader as
Chris suggests. That can't do any harm (except not being too
consistent of course).
The advantage of this approach is that it's more like what a regular
program would do - we're not putting more special "pretend anaconda's in
a completely different environment" stuff in.
Absolutely. I will send a patch later today.
Ales
_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/anaconda-devel-list
