Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 04 Mar 1997 08:45:46 -0800
From:      fred@lightside.net (Fred Condo)
To:        freebsd-bugs@FreeBSD.ORG
Subject:   2.1.7 upgrade bugs
Message-ID:  <1354657313-15376408@dcs-chico.com>

next in thread | raw e-mail | index | archive | help
(A copy of this message has also been posted to the following newsgroups:
comp.unix.bsd.freebsd.misc)

I recently upgraded 3 systems to 2.1.7. One was previously running
2.1.6.1, one was running 2.1.5, and one was running 2.1.0. The following
problems showed up after the installation on all three:

1. The installer stomps on /etc/resolv.conf, and doesn't properly
resurrect the real one. That meant all three machines were banging on the
slow nameserver I happened to use for installation purposes, instead of
querying themselves as they were supposed to. Name lookups were slow and
caused abysmal performance (especially on the mail server), until I
discovered the error. This is definitely a bug; I carefully compared the
contents of /etc and /etc/upgrade.

2. The on-screen message erroneously refers to /etc/update instead of
/etc/upgrade.

3. Even after fixing /etc/resolv.conf, sendmail performance was abysmal:
it took 30 seconds or more between making a connection on port 25 and
seeing the sendmail banner. Compiling and installing sendmail from the
Allman source fixed it. Something seems to be rotten in the sendmail build
in FreeBSD.

The following problem only occurred on the machine where I also did a make
world to allow for long file names:

4. I don't know if this is a bug, but it feels like one. Make world
compiles and installs a new kernel, but both ps and top kept failing with
kvm size mismatches until I manually compiled and installed a new kernel
after make world.

The following problem only occurred only on the machine that started at 2.1.0:

5. Until a second reboot, vi kept complaining about an out-of-date shared
library.
-- 
http://www.lightside.net/~fred/ + net access + http://www.lightside.net/
"Attempts to control the use of encryption technology are wrong in
principle, unworkable in practice, and damaging to the long term economic
value of the information networks." - UK Labour Party



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1354657313-15376408>