Re: 9.1-RC3 LiveCD missing features

From: Devin Teske <devin.teske_at_fisglobal.com>
Date: Sat, 8 Dec 2012 15:29:23 -0800
On Dec 8, 2012, at 12:09 PM, Garrett Cooper wrote:

> On Sat, Dec 8, 2012 at 9:42 AM, Ian Lepore
> <freebsd_at_damnhippie.dyndns.org> wrote:
> 
> ...
> 
>> It shouldn't require rc.initdiskless; just the fact that rc.d/var
>> detects it can't write to /var should cause it to automatically create a
>> memory fileystem for it, and minimally populate it.  As far as I know,
>> this is automatic unless you use rc.conf knobs to disable it.
> 
> That doesn't solve other things like hostid, rc.conf, ssh/*, etc not
> being writable. I've had to use LiveCDs (post 9.0) a few times and the
> amount of hoops that I have to go through in order to get a working
> system is silly.
> 
> Plus, it would be nice if it used an mfsroot, like another PR I filed
> said (it sucks having to wait 3 minutes for the USB CD to probe before
> I can mountroot over IPMI on Supermicro machines).
> 

Never fear… as discussed at November's DevSummit…
I'm both extremely dediated-to (and designated whipping boy for) bringing mfsroot back as an option.

It's going to take awhile tho.

The roadmap essentially looks like:

1. Use bsdconfig as a dumping ground for migrating sysinstall features such as scriptability (working on that now; aim is that the scripting engine can run sysinstall scripts, achieving backward compatibility with  15+ years of past functionality).

2. When bsdconfig has the aforementioned scripting engine, the media framework (working on that right now, simultaneously with the scripting engine), the packages module (which will be pkgng based -- requires the completion of the media framework first), and has the distributions module then there will be a bsdconfig 1.0 release. Correspondingly at that time the WITH_BSDCONFIG hook will dissolve and no-longer be required (and thusly bsdconfig(8) will be available in releng distributions).

3. With the release of bsdconfig and general availability, I can then start the process of linking bsdinstall to bsdconfig.

4. bsdinstall will become i18n-ready.

5. bsdinstall will inherit the media selection dialog from bsdconfig (which is based on sysinstall's media selection dialog)

NOTE: And I'm sure other things will happen to bsdinstall under my wing, but I'm not ready to commit to more than that at this time

6. I will then calculate the dependencies of bsdinstall/bsdconfig to create an mfsroot

7. Leave the existing releng Makefile alone, but make a release/Makefile.bsdinstall that generates an mfsroot

And then we'll start down the road of testing mfsroot-based bsdinstall media.

Right now I'm at #1, working to get to #2. bsdconfig in ports is at 0.7.x. When I complete the media framework (well over 2,000 lines of code so far -- restructuring and growing daily, trying to get to a stable point of commit that reflects a working media selection dialog with ALL permutations that sysinstall supported), I'll release a 0.8.x port. Beyond that, when I finish the packages module (pkgng based of course), I'll release a 0.9.x port (all of these being just frozen snapshots of HEAD's usr.sbin/bsdconfig/ dir). Last (and completing step #2 above), I'll quickly generate the distributions module which will conclude the entirety of scope -- replicating sysinstall's "Post-Install Configuration" menu (and thus, 1.0 release and unleashing).
-- 
Devin

_____________
The information contained in this message is proprietary and/or confidential. If you are not the intended recipient, please: (i) delete the message and all copies; (ii) do not disclose, distribute or use the message in any manner; and (iii) notify the sender immediately. In addition, please be aware that any message addressed to our domain is subject to archiving and review by persons other than the intended recipient. Thank you.
Received on Sat Dec 08 2012 - 22:29:29 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:40:33 UTC