Charles Steinkuehler wrote:
>
> > My thought, too, is that all packages EXCEPT root.lrp could be put on
> > the network on an FTP host or such like - and thus, not only could the
> > system be updated on the fly, but if you had multiple hosts they would
> > all have their sources updated at the s
> My thought, too, is that all packages EXCEPT root.lrp could be put on
> the network on an FTP host or such like - and thus, not only could the
> system be updated on the fly, but if you had multiple hosts they would
> all have their sources updated at the same time, and so on.
See trinix...
htt
Jacques Nilo wrote:
>
> - Original Message -
> From: "arne @ loopback . org" <[EMAIL PROTECTED]>
> > I would go a step further. make a minimal busybox
> > only containing very few
> > applets(tar,msh as shell,mount,ls,cat,...)
> > And link it statically with
> > uClibc. This will result i
Matt Schalit wrote:
> I got around to replacing the ipmask-0.30 shell script
> with the 0.32 compiled version on the LEAF oxygen packages site.
>
> I don't get the logic of the -c option. It says:
> -c Display only CIDR
>
> So why would that output ip/cidr?
> # ipmask -
- Original Message -
From: "arne @ loopback . org" <[EMAIL PROTECTED]>
> I would go a step further. make a minimal busybox only containing very
few
> applets(tar,msh as shell,mount,ls,cat,...) And link it statically with
> uClibc. This will result in a quite small binary and you don't need
Jacques Nilo wrote:
> Well I think initrd.gz should contain as few things as possible. In my
> proposal it boils down basically to libc6, ld-linux, ash, busybox, sed
> and root.linuxrc.
In this case, it's like this: libc6 (500k), ash (90k), busybox (150k),
sed (64k), and others so right away
"Angelacos, Nathan" wrote:
> load_disk_package contains this line:
>
> [ "${PKG##*/}" = "${CONFIG##*/}" ] && return 0
>
> which if I read correctly, exits if the current package is the config
> package. That seems like the right thing to do (although we do load root.lrp
> twice.) Howev
