Re: moving kernels between machines

2007-01-06 Thread Marcos Laufer
Wow , that openbsdbinpatch looks pretty good! I 've just downloaded it and
the idea of making binary patches in order to easily copy them to other
similar
machines is excellent.

It would be great if openbsdbinpatch could be configured like TEPATCHE
specifying
the folder where to get the patches from, instead of specifiyng patch by
patch in
the Makefile, so it could be used to patch any OpenBSD version,
not just the last one.
Perhaps someone with some programming skills could do it.
BTW , TEPATCHE is an Automatic Patcher for OpenBSD, it can be
downloaded from http://www.gwolf.org/soft/tepatche/index.html

Regards,

Marcos Laufer

- Original Message - 
From: Tasmanian Devil [EMAIL PROTECTED]
To: OpenBSD misc@openbsd.org
Sent: Friday, January 05, 2007 5:42 AM
Subject: Re: moving kernels between machines


 - Machine A, a single i386 box without enough disk space to unpack the
 source tree

http://openbsdbinpatch.sourceforge.net/ :-)

Tas.



Re: moving kernels between machines

2007-01-05 Thread Tasmanian Devil

- Machine A, a single i386 box without enough disk space to unpack the
source tree


http://openbsdbinpatch.sourceforge.net/ :-)

Tas.



Re: moving kernels between machines

2007-01-05 Thread David Newman
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 1/5/07 12:42 AM, Tasmanian Devil wrote:
 - Machine A, a single i386 box without enough disk space to unpack the
 source tree
 
 http://openbsdbinpatch.sourceforge.net/ :-)

Thanks much for this, and also for Nick Holland's excellent suggestion
about keeping updated with -release.

Paranoid question: How does a user know which binpatches to trust?

dn
iD8DBQFFnpCayPxGVjntI4IRAugUAJ4gN67UZGnyRgxfuvtk9BFvix2q5wCaA+gD
2Oe+vTwBToPP7J6IxDyf7vA=
=LtR5
-END PGP SIGNATURE-



Re: moving kernels between machines

2007-01-05 Thread Tasmanian Devil

Paranoid question: How does a user know which binpatches to trust?


I'd say you should at least trust the ones you've build yourself on
your Machine B. ;-)

Tas.



moving kernels between machines

2007-01-04 Thread David Newman

I have two machines:

- Machine A, a single i386 box without enough disk space to unpack the 
source tree


- Machine B, a two-CPU i386 box running bsd.mp with plenty of disk

My questions:

1. For purposes of applying kernel security patches, can I compile a 
patched kernel on Machine B and just transfer it over to Machine A and 
reboot?


2. If the answer to (1) is yes, what if anything do I need to do with 
userland on Machine A? For example, how would I apply patch 001 for 4.0, 
which is just for httpd?


many thanks

dn



Re: moving kernels between machines

2007-01-04 Thread Nick Holland
David Newman wrote:
 I have two machines:
 
 - Machine A, a single i386 box without enough disk space to unpack the 
 source tree
 
 - Machine B, a two-CPU i386 box running bsd.mp with plenty of disk
 
 My questions:
 
 1. For purposes of applying kernel security patches, can I compile a 
 patched kernel on Machine B and just transfer it over to Machine A and 
 reboot?

Of course... :)

 2. If the answer to (1) is yes, what if anything do I need to do with 
 userland on Machine A? For example, how would I apply patch 001 for 4.0, 
 which is just for httpd?

IF you really know exactly what files are altered, build 'em on your
fast, big machine and copy them over to your small machine, making sure
you get permissions and such correct.

IF you do not know for sure which files are altered, I'd suggest just
making your life simple, and follow stable, make a release, and install
that on the small machine (and any others).  When staying with stable,
the process is trivial: unpack all .tgz files (don't forget the 'p'
option!!), install the kernel, reboot.  If your big, fast machine has
some time when no one would notice, you might even want to set it up to
periodically make a -stable release for you (yes, the official
instructions say reboot between building the kernel and the userland,
but since the API doesn't change in -stable, you can almost always get
away without the reboot.  In fact, on my -stable build machine
(actually, a VMware session) at work, I only reboot the thing to make
sure the build is good before installing it on a critical machine.

AGAIN, if you know exactly what subset of things need to be patched,
(for example, httpd), you could just stop and start that one service,
but usually, by the time you have figured that all out, you could have
just rebooted.

See:
  http://www.openbsd.org/stable.html
  http://www.openbsd.org/faq/faq5.html , especially sections 5.1, 5.4
  http://www.openbsd.org/faq/faq4.html#AddFileSet

Nick.
(who runs ONLY enough -release/-stable to verify the upgradeXX.html
instructions are valid)