[3487] in Release_7.7_team

home help back first fref pref prev next nref lref last post

Re: Minutes of 2002-09-11 release team meeting

daemon@ATHENA.MIT.EDU (t. belton)
Thu Sep 12 11:50:19 2002

Date: Thu, 12 Sep 2002 11:50:17 -0400 (EDT)
From: "t. belton" <tbelton@MIT.EDU>
To: Greg Hudson <ghudson@MIT.EDU>
cc: <release-team@MIT.EDU>
In-Reply-To: <200209120153.VAA23137@error-messages.mit.edu>
Message-ID: <Pine.GSO.4.33L.0209121136140.19203-100000@iphigenia.mit.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

>   * Users will need to clear certificates and get new ones; converting
>     from Netscape 4 apparently doesn't work.  A program "clear_certs"
>     in the infoagents locker does the erasing part; people at the
>     meeting seemed to like the name "clearcerts" better, so Todd
>     should rename it.

Actually it has, until now, been called 'erase_certs' because I wanted to
be very emphatic about the destructive nature of this script. It removes
every bit of ANY and ALL existing certs in your Mozilla profile directory
- all the *.db files and all lines about default cert choices in prefs.js.
This is what it takes to realio, trulio wipe out all traces of those
poorly-converted 4.x certs if you want to avoid problems before they
happen. Or to decisively eliminate the problems after they happen.

(Just to make it difficult, some people can apparently just delete their
4.x-converted certs and get new ones and work fine, without ever needing
this script at all. But I figured I had better provide it just to simplify
walking people through the cleanout, should they need to do the cleanout.)

At any rate, I want it having a strong name so that people don't try to
use it on a whim. The idea is that they use it only when told to do so
by help personnel. I'm not sure 'clear'-anything is emphatic enough.
Other suggestions?

- Todd

---
"I suggest, in the future, that you replace this with the words 'real,
non-artificial, unboned dead frog' if you want to avoid prosecution!"


home help back first fref pref prev next nref lref last post