[1768] in NetBSD-Development
Re: NetBSD-Athena sucks, again!
daemon@ATHENA.MIT.EDU (Nathan J. Williams)
Sat Sep 5 11:03:30 1998
To: "Charles M. Hannum" <mycroft@MIT.EDU>
Cc: netbsd-dev@MIT.EDU
From: "Nathan J. Williams" <nathanw@MIT.EDU>
In-Reply-To: Your message of "Sat, 05 Sep 1998 04:11:15 EDT."
<199809050811.EAA02558@portnoy.MIT.EDU>
Date: Sat, 05 Sep 1998 11:03:22 EDT
> In a most ridiculous set of affairs, it seems that NetBSD-Athena 8.2
> was released with a *major* known bug in the installation procedure,
> despite the bug and a fairly trivial workaround being known for
> multiple months in advance. That is, to wit, that the AFS startup
> during the installion hangs.
If you're going to flame the install into the ground, then at
least get the facts right first. The "trivial workaround" you refer to
is not, in fact, a workaround. The current install floppy, which I
presume is the one that you were using, has that change (using a
mfs-disk cache instead of a memcache). While one person observed that
it fixed the problem for them, it did not fix it for everybody.
The problem is not particularly deterministic; it happens both
with and without this "workaround"; at times, it will happen on a
machine on one attempt to install, but not happen the next time, five
minutes. At one point I thought that the bug was related to the AFS
server caching the non-working state of the machine and that changing
the IP address on the install helped, but I now think that those
observations were simply coincident with random successes and
failures.
As you are well aware, AFS is a touchy beast, and somewhat
gruesome to debug. If you would like to contribute to fixing it, I'm
all ears.
> If it weren't for the fact that probably every user who tries it is
> going to lose in this fashion, and is going to simply try Linux
> instead, I would find this comical. As it stands, it's just yet
> another example of NetBSD-Athena losing catastrophically over the
> years.
Every user is not going to lose. Many installs do not run into
this problem. I would certaintly prefer not to have the problem exist
at all, but I have been unable to find a solution or actual
workaround.
> I'm sick of this. I'm also sick of butting heads with people (largely
> Greg) every time I actually offer to contribute something to
> NetBSD-Athena or related efforts.
I'm sorry that you feel that you've been butting heads with
people. I'm actually trying to get something to work, and I think that
when we've interacted on this project, it has been productive.
- Nathan
