[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: VERY preliminary boot floppies available for testing



(cc:'d to debian-boot because of netfetch problems)

On Thu, Mar 02, 2000 at 04:34:25PM -0500, Daniel Jacobowitz wrote:
> Odd.  4c46 is LF; I'm wondering if the binaries are corrupted, or if
> the kernel's ELF support went bad.

OK.  Now I officially feel stupid.

Turns out I downloaded base from http://http.us.debian.org/something.
So I decide to try your archive like I should have and see what
happens.

Then it turns out that your webserver wouldn't accept the mangled
http requests sent by netfetch.  Fine.  I don't feel quite as stupid,
especially after I write a primitive-but-working version http proxy
on my i386 linux box in (technically) one line of perl.

Then it turns out netfetch's proxy support won't work for me.  It
seems not to notice I even specified one.  Fine.  I modify the
proxy to only forward to your box, then put its address and port in
the URL field and 'none' back in proxy.

Then it turns out netfetch changes http://foo.bar:1883/ into
http://foo.bar:0/1183/ and tries to fetch from port 0.  Okay,
whatever.  I run the proxy as root on port 80 and don't specify a
port in the URL field.

Netfetch still connects to port 0.  I figure it must have become
'confused', kill dbootstrap from tty2, and try again.  Bingo!

When it's finished, I reboot into linux from the hard drive.
Everything seems to work now, except there's a big hunk of binary
garbage in the kernel messages.  Ignoring that, I configure the
base system.  I'm up to dselect now and so far, so good.  More
later.

I don't know for sure, but this looks to me like one of those
unexpected success reports...

--
James Deikun, Techie(tm), CSI Multimedia
The opinions expressed &c.


Reply to: