Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 11 Dec 2002 21:09:02 +0100
From:      Gerhard Sittig <Gerhard.Sittig@gmx.net>
To:        current@FreeBSD.ORG
Subject:   Re: HEADS UP: you need to install a new kernel before an installworld.
Message-ID:  <20021211210901.Q4807@shell.gsinet.sittig.org>
In-Reply-To: <20021030.011458.37440476.imp@bsdimp.com>; from imp@bsdimp.com on Wed, Oct 30, 2002 at 01:14:58AM -0700
References:  <20021028225555.5630B2A88D@canning.wemm.org> <20021028154058.Y3316-100000@master.gorean.org> <20021030.011458.37440476.imp@bsdimp.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Oct 30, 2002 at 01:14 -0700, M. Warner Losh wrote:
> 
> UPDATING has the closest thing to a comprehensive guide.  As far as I
> can tell, it is definitive in its list of potential issues, but if I'm
> wrong, let me know.
> 
> I'm just glad I don't have to document all the things that mergemaster
> does.

[ late reply, I know -- catching up with the -current ML traffic ]

May I remind you of the docs/40851 PR ([PATCH] "mergemaster -p" in
UPDATING's "COMMON ITEMS" section)?  From the PR's description:

  | UPDATING in -STABLE does not mention the needed "mergemaster -p"
  | step in its COMMON ITEMS section.  UPDATING in -CURRENT seems to
  | have the "mergemaster -p" invocation in the wrong place.

And if I may add:  One should run the mergemaster script which
lives in the source tree since the installed one (the one found
in $PATH when a short command is given) might not be up to date
or even can be inappropriate or wrong when operating on a new
source tree.

Calling the script from the source tree will work as long as it
is self contained (does not need other source which is not
installed yet but merely uses system commands from the environment
it is running in).  Installing the script before calling it (using
$PATH) might be more robust for the future.  In this case the
"make install; mergemaster [opt]" sequence should be included in
the COMMON ITEMS section (yes, I know it's among the YYYYMMDD
entries, but this might not be enough when this command is
considered part of the general and regular sequence of updating
steps).

The manpage says about the -p option:

  | .It Fl p
  | Pre-buildworld mode.
  | Compares only files known to be essential to the success of
  | {build|install}world,
  | including
  | .Pa /etc/make.conf .

that's why I suggest moving the "mergemaster -p" invocation up
before the buildworld step when we talk about robustness of the
updating sequence.  We might have gotten away with what we did
up to now, but I wouldn't count on it to always succeed ...


I just checked against my local CVS repo (revs 1.73.2.76 and 1.228
of UPDATING) and the PR's issues still apply.


virtually yours   82D1 9B9C 01DC 4FB4 D7B4  61BE 3F49 4F77 72DE DA76
Gerhard Sittig   true | mail -s "get gpg key" Gerhard.Sittig@gmx.net
-- 
     If you don't understand or are scared by any of the above
             ask your parents or an adult to help you.

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20021211210901.Q4807>