[3073] in testers
Re: sun4 8.1.0: update_ws, mkserv
daemon@ATHENA.MIT.EDU (Albert Dvornik)
Fri Apr 25 23:41:36 1997
To: Jonathon Weiss <jweiss@MIT.EDU>, Greg Hudson <ghudson@MIT.EDU>
Cc: testers@MIT.EDU, mkserv@MIT.EDU
From: Albert Dvornik <bert@MIT.EDU>
In-Reply-To: Albert Dvornik's message of Fri, 25 Apr 1997 16:25:56 EDT
Date: Fri, 25 Apr 1997 23:41:12 EDT
"Houston, we have a problem."
So, while I thought that the update setup might give mkserv different
std{in,out,err} depending on whether it's auto-updating or not, I
decided it would be a good idea to test it and see once and for all.
This turned out to be a little more tricky than I anticipated (due in
part to problems mentioned in my next piece of e-mail), but doable, as
long as the machine is coerced into believing that 8.1 is in fact not
a test release, really, honest. O=)
The upshot of all this is that when a machine auto-updates on reboot,
previously correctly implemented mkserv scripts will get a valid TTY
as their stdin and ask questions during "mkserv update", and AFAIK
there is no way for them to know better. (I didn't test the
reactivate-caused update case, but I sort of suspect that it has the
same failure mode.)
I'm not sure what the best way to fix this is, but I'm sure you'll
come up with something... =)
--bert
