Re: init bug / crash recovery

Matthias Urlichs (smurf@smurf.noris.de)
Sat, 16 Mar 1996 00:17:21 +0100

In linux.dev.kernel, article <199603151528.KAA09740@sunspot.tiac.net>,
Robert L Krawitz <rlk@tiac.net> writes:
>=20
> Rescue disks fail themselves, etc... I guess I'm of the two belts,
> two suspenders, and still double check everything school of system
> management. Dynamic linking is all well and good for large user-mode
> programs, but stuff that's critical for booting should be as
> absolutely bullet-proof as possible.
>=20
The point where a bunch of statically-linked binaries is larger than th=
e
same bunch, dynamically linked, plus a stripped libc.so, is reached pre=
tty
fast.

A single boot/root/rescue floppy can hold a kernel and a compressed dis=
k
image with 3 MB of RAM disk. With the bzimage stuff, you can even go to=
=20
your regular system after an emergency repair, without another reboot (=
yes,
sometimes that is important).

In fact, I'll go and use that code for regular booting. There's nothing
like having to build just _one_ kernel for a heap of machines with wild=
ly
disparate SCSI adapters... (OK, two kernels -- the math emulator for th=
e
386 I use as a test machine isn't a module, unfortunately...)

--=20
Yippies, hippies, yahoos, Black Panthers, lions and tigers alike -- I
would swap the whole damn zoo for the kind of young Americans I saw in
Vietnam.
-- Spiro Agnew
--=20
Matthias Urlichs \ XLink-POP N=FCrnberg | EMail: urlichs@smurf.=
noris.de
Schleiermacherstra=DFe 12 \ Unix+Linux+Mac | Phone: ...please use =
email.
90491 N=FCrnberg (Germany) \ Consulting+Networking+Programming+etc'i=
ng 42
PGP: 1B 89 E2 1C 43 EA 80 44 15 D2 29 CF C6 C7 E0 DE=20
Click <A HREF=3D"http://smurf.noris.de/~smurf/finger">here</A>.