[1541] in NetBSD-Development
More bugs in the new stuff
daemon@ATHENA.MIT.EDU (Nathan J. Williams)
Sat Jan 17 08:42:35 1998
To: netbsd-dev@MIT.EDU
From: "Nathan J. Williams" <nathanw@MIT.EDU>
Date: Sat, 17 Jan 1998 08:42:25 EST
Tonight I convinced snorklewacker to run from a local srvd
built from our new repository. A bunch more bugs turned up. I'm going
to be working on them, but I thought I'd mention them here in case
they ring bells with anyone.
- discuss is quite broken. The tty client sets the terminal to not
echo what the user types, and core dumps easily ("print last",
for example). M-x discuss in emacs produces edsc zombies until you
run out of processes. dsgrep also likes to core dump, though not
quite as easily.
- krb5 stuff is slow and somewhat flaky. There's some bad interactions
here between the kerberos servers and our fallback code, but I'm not
sure where the real problem is. tcpdumping while logging in via krb4
shows the machine talking to all four kerberos servers.
- emacs has some brokenness. Loading files that are under version
control produces:
"wrong type argument: sequencep, re-search-forward"
Version control bindings (C-x v and friends) error out with
"Symbol's value as variable is void: nded"
- MH is trying to use /usr/lib/sendmail to send mail. This doesn't
work. I haven't quite figured out the right way to tell MH to use
sendmail somewhere else.
- lpr doesn't print; it complains about "Lost connection to printer..."
- running /etc/athena/syncconf -a produces an ed-esque "?" as output.
- clusterinfo stuff is still not very useful
- Nathan
