LFS 5.1

Nathan Coulson conathan at conet.dyndns.org
Wed Jan 7 00:09:28 PST 2004


> Bruce Dubbs wrote:
>
>> Note: Bug 333 looks like it can be combined with bug 325.
>>
>
> Is anyone else working on these 2 ATM?  I've not heard from conathan in
> a couple of weeks on the lists, but lfs-bootscripts-1.13-conathan-rc4
> implements all of 325 and even a simple workaround for 333.  It also
> includes the netmountfs script, but that's easily ripped out and been
> put in BLFS already anyways, tho I'd still like to see it in LFS.  There
> were a couple of maybe better suggestions on 333...personally I like
> putting in a sleep statement as opposed to the read, but I think 10 is a
> little long.

whoops, been busy.  [and haven't bothered putting my stuff back after I
rebuilt my server].  I never got much input on my work other then you [DJ]
and Zeplin anyway, and J_Man [jeremy?] wanted to just redo everything.

I'll see if I can clean up what I got, and repost it, and if I meet J_Man
[jeremy?] online, I'll see what he wants to add.

> For me, (and in Nathans's rc4..least I think he put it in there) I put a
> workaround in funcitons that allows you to set a variable for this.
> Currently if $ERRWAIT == 1, it does what was always done (easily
> changable), but I also added a very simple logging function that else
> $ERRWAIT != 1 it keeps on going. No stop, and just logs it to a file
> (/var/log/lfs-shutdown.log) with date and time.  I did this with
> headless systems in mind.  Might be kinda silly I guess, but it works
> for me.  Any other or better ideas?

I dont remember adding that, but sure.  Where would a bootscript
configuration file go?

what happens if something happens after mountfs on shutdown?

> Further...any interest on making them a little nicer looking?  I seen a
> bootdisk (the partimage system rescue cd) that does nice colors and
> allows for errors and warnings, so I mimiced it in my own
> functions...sort of.  INIT messages are separated from the script echo
> functions [func_echo(), warn_echo(), and fail_echo()] by prefixing the
> scripts echos with a green "  *  " (or yellow *** or red ***** for warn
> and fail).  No more need for the sed in sysvinit either as the INIT
> messages stick out like a sore thumb.  A 'sleep 1' and 'dmesg -n 3' in
> rcsysinit/S05dmesg, and my boot looks real clean and colorful.  :-)
>
>> I suspect the XML overhaul and glibc 2.3.3 should be postphoned to 5.2,
>> otherwise everything looks quite doable to me.
>>
>
> Same here...think 2.3.3 is probably okay, but it's awful close to
> release to try and thoroughly test the new glibc.  I've been using it
> for a while now, but not in anything other than recent LFS hosts.  How's
> it do with say an old RH-7x or soemthing along those lines?
>
> My longwinded, runon and on opinions.
>
> -- DJ



More information about the lfs-dev mailing list