[2906] in RedHat Linux List
Re: Strange login behaviour:
daemon@ATHENA.MIT.EDU (Donnie Barnes)
Wed Nov 6 01:17:11 1996
To: redhat-list@redhat.com
In-reply-to: Your message of "Tue, 05 Nov 1996 19:11:32 EST."
<Pine.LNX.3.95.961105190554.5072A-100000@robob1.robob.com>
Date: Wed, 06 Nov 1996 01:13:55 -0500
From: Donnie Barnes <djb@redhat.com>
Resent-From: redhat-list@redhat.com
Reply-To: redhat-list@redhat.com
>> Look, RPM has a very sophisticated mechanism for deciding if it's safe
>> to leave a file in place, or if it needs to change it.
>
>I think many of us would disagree.
Okay, I'll explain in full for those who have missed it.
When RPM first installs a config file, it logs the md5 of it in the
database. It saves that as long as the package is installed.
Now, when you upgrade a package that has that config file in it,
RPM knows the md5 of the new file. It first calculates the md5 of
the existing file. If it's the *same* as it originally was, it
knows you never modified it, and it's safe to upgrade to the new
one. If it's not, we have a problem. Now, it checks the md5 of
the new file against the *original* md5. If those are the same,
we have no problem because we *know* beyond a shadow of a doubt
that the file format did *not* change between the two package
versions (if it had changed, then the new file wouldn't be the
same as the old or it wouldn't be a working config file). We
can then safely *leave* your modified config file because we know
for sure that it is valid (and we throw away the one the package
would have installed). If, however, the new file and the original
have different md5's (ie, they are not the exact same file), we
can *not* be sure the file formats for the config files are the same
and hence can *not* be sure that your config file is valid. In
this case, we install *our* config file, and back yours up as a
.rpmsave (and notify you that we did that).
You *could* argue that certain config files like sendmail.cf
should be tagged so that they are always "compatible"...that may
be a valid argument.
>> If you have complaints,
>> please send them to the rpm-list@redhat.com along with a *better* way to
>> handle things.
>
>Hey man I really am on RH's side. You should learn how to talk to your
>customers in a more polite and professional manner. NOT my job to improve
>RPM, that's for you folks. Complaints are what make the product better
>and is an oppurtunity to improve the product in order to sell more.
I didn't see anything unpolite or unprofessional there. I didn't tell
you to go change the code yourself, I just said we have a way to do
things, and if you prefer something different (or "better"), you
should tell us what that is.
Also, this is the free software community. Most folks will agree that
it's everyone's job to improve free software in any way they can, from
bug reports and suggestions, to patches and new utilities.
"Complaints" that basically say "I don't like the way this works"
don't do us any good. You'd better tell us how you'd like it to work.
Nothing unprofessional about that.
As I've said before, this is an open forum where users can batter us
all they want. We take that and come back for more. But, you'd better
be able to take short, direct, and concise statements without reading
things into them. I don't have time to sugar coat everything when I'm
trying to help users at 1:00am....
--Donnie
--
Donnie Barnes http://www.redhat.com/~djb "Bah."
djb@redhat.com http://www.turner.com/lazarusman/
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
_Things You'd NEVER Expect A Southerner To Say_ by Vic Henley:
** I hate the long version of ``Free Bird''.
--
PLEASE read the Red Hat FAQ, Tips, Errata and the MAILING LIST ARCHIVES!
________________________________________________________________________
http://www.redhat.com/RedHat-FAQ http://www.redhat.com/RedHat-Errata
http://www.redhat.com/RedHat-Tips http://www.redhat.com/mailing-lists
------------------------------------------------------------------------
To unsubscribe: mail -s unsubscribe redhat-list-request@redhat.com < /dev/null
