I am an end-user since 2004. Over time I suffered with long but successful transition to Anaconda. I think that at times you need to see things from another perspective, the guy who wants to install or reinstall one or more systems.
As an example, using Fedora, we have a DVD that was released on December 13th, 2013, and it is current as of that date. With it's use comes around 1000 updates, about equal to the 2 or more gigabytes of downloads. If one has to install this distribution onto several systems, we have a tremendous waste of downloads and time.
But...
If we have an install iso that is built from kickstart files, and it's sole role is to assemble and create a current iso within the user's Download directory, using that ISO image to do installations would result in new systems as as current as the day the kickstart iso was downloaded. Creating other systems from that iso created via ks will result in the other systems being current, with no massive downloads of updates.
The other advantage to this approach is that the system to analyze hardware, could be enhanced, so that new ISO installations are done smoothly. And one more thing, the users will be able, to add other repositories and software groups to that ks iso.
Thus
Kickstart ISO ==>Download==>Create Install ISO==>[Optionally add repositories and other software]==>
Create Local distribution DVD Image ==>Install to other systems.
If I am off topic, sorry.
Currently my approach is do 3 netinstalls for 6 multiboot systems on 3 different hardware systems, and and I am annoyed at the waste of bandwidth and time.
If I am off topic, sorry.
Currently my approach is do 3 netinstalls for 6 multiboot systems on 3 different hardware systems, and and I am annoyed at the waste of bandwidth and time.
Regards
Leslie
Leslie
Mr. Leslie Satenstein
SENT FROM MY OPEN SOURCE LINUX SYSTEM.
From: Jon Stanley <jonstanley@xxxxxxxxx>
To: Discussion of Development and Customization of the Red Hat Linux Installer <anaconda-devel-list@xxxxxxxxxx>
Sent: Monday, July 21, 2014 2:23 PM
Subject: LiveCD or equivalent questions...
I'm not sure where to post this, but I think all of the relevant
players are on this list, but I'm happy to move this elsewhere if
warranted.
At $DAYJOB we have a fairly simple requirement - a stateless image
that can be either PXE booted or booted off of local disk. The
important characteristic here is that when the machine is rebooted, it
loses all concept of local state. The authority of what state the
machine should be in lives elsewhere, so it needs to come up, be
stupid. and phone home for further instructions.
We originally thought that LiveCD was the perfect mechanism for this,
and it worked fine - until the devicemapper read/write overlay (RAM
backed in this case) got filled up, and then the whole thing goes
BOOM!
What I'm interested in is finding some way for a non-persistent
overlay to not blow up in our faces when the system has been up for
awhile and various logs, etc have been written. The use case is for a
KVM hypervisor, so we'll have long running VM's on the hypervisor and
all the attendant logs being written. The VM's are of course backed by
local disk.
Is such a thing feasible? Am I barking up the wrong tree, or is there
something that I'm not thinking of here?
_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/anaconda-devel-list
_______________________________________________ Anaconda-devel-list mailing list Anaconda-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/anaconda-devel-list
