Hi,
On 12/03/2009 05:38 PM, Chris Lumens wrote:
For a number of reasons, we currently have our customers boot an
install initrd that has a file named /tmp/updates/anaconda present.
(It's my understanding that if present it's used instead of the
normal anaconda binary. Has that changed?)
No, that still works.
In our case this file is
a shell script that calls the original anaconda, and performs some
additional install work after anaconda exits.
So will the change you are describing prevent this script from
running anaconda?
Yes, it would since the parent of the anaconda process is some program
other than /sbin/loader.
Perhaps there's an alternate way to do what you need to do, though.
You're already using a special initrd with your own updates. Can you
add your additional work to a new python file and add it as another step
in the dispatcher? Is what you're doing simple enough to just add
somewhere else in anaconda?
Have you tested your current setup with Fedora or RHEL6 lately? I'm
guessing it probably doesn't work.
What release will this change likely appear in?
That's a good question. I'm undecided if this is RHEL6-worthy or not.
If not, it's time for fun with git branches. It's certainly nothing
that's going to appear in RHEL3/4/5.
I think this definitely is RHEL-6 worthy, we are going to maintain RHEL-6
for ages, so having a cleaner code base to start with is very good.
Also if these options will be present in RHEL-6 people may try and use them.
Regards,
Hans
_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/anaconda-devel-list
