Bug 21997

Summary:	after pkg_add 40upgrade-2000.08.25 telnetd, ftpd and wget all fail with coredump
Product:	Base System	Reporter:	tanbin <tanbin>
Component:	misc	Assignee:	Satoshi Asami <asami>
Status:	Closed FIXED
Severity:	Affects Only Me
Priority:	Normal
Version:	4.0-RELEASE
Hardware:	Any
OS:	Any

Description tanbin 2000-10-15 11:00:01 UTC

in order to install a webalizer upgrade, i first downloaded 40upgrade.tgz and run pkg_add 40upgrade.tgz *a few times*.
After that, wget will fail with a core dump, and i can no longer ftp or telnet from other machines to this host.
i have killall -HUP inetd a few times.
from /var/log/messages:
---snip--
Oct 15 16:24:09 wwwftp /kernel: pid 68960 (wget), uid 0: exited on signal 12 (core dumped)
Oct 15 16:54:12 wwwftp /kernel: pid 69192 (fetch), uid 0: exited on signal 12 (core dumped)
Oct 15 17:07:24 wwwftp /kernel: pid 69239 (telnetd), uid 0: exited on signal 12 (core dumped)

Fix: 

reboot??

Comment 1 Will Andrews freebsd_committer

2000-10-16 03:12:16 UTC

Responsible Changed
From-To: freebsd-bugs->asami

Over to MAINTAINER.

Comment 2 sue 2000-11-01 07:58:20 UTC

Similar here. Two machines, each running 4.0R installed from CD set.
Some ports installed from CD1 and working happily, no cvsup upgrades,
no additional software added or upgraded from compiled sources.

Downloaded 40upgrade.tgz this morning (12 hours ago) and pkg_add to
both. Then added two packages (binaries) to each machine, downloaded
from www.freebsd.org/ports
  libslang-1.4.2 (an earlier version was already installed)
  urlview
On one of the two similar machines, I also installed the mutt-1.2.5
package.

After this, trouble began. Existing tcp/ip connections seemed fine, but
new
telnet and ftp attempts failed, both in and out. netstat stopped working
(in both direcctions), so did w, arp, and mutt-generated connections to
a pop3 server. But nslookup, ping, traceroute and ps still worked fine,
and the machine which had apache installed still happily served pages to
remote visitors.

I've got enough cores here to start an orchard. Here's a sample of what
happens:
(vimutti is the affected system)

sue@vimutti$ uname -a
FreeBSD vimutti.welearn.com.au 4.0-RELEASE FreeBSD 4.0-RELEASE #0: Thu
Aug 24 19:40:20 EST 2000    
root@vimutti.welearn.com.au:/usr/src/sys/compile/VIMUTTI  i386

sue@vimutti$ telnet phoenix
phoenix: No address associated with hostname
sue@vimutti$ ftp phoenix
Bad system call (core dumped)
sue@vimutti$ telnet phoenix smtp
phoenix: No address associated with hostname
sue@vimutti$ telnet phoenix 80
phoenix: No address associated with hostname
sue@vimutti$ ssh phoenix
Bad system call
sue@vimutti$ netstat
Active Internet connections
Proto Recv-Q Send-Q  Local Address          Foreign Address       
(state)
Bad system call
sue@vimutti$
sue@vimutti$ ping -c 3 phoenix
PING phoenix.welearn.com.au (203.35.200.139): 56 data bytes
64 bytes from 203.35.200.139: icmp_seq=0 ttl=255 time=1.187 ms
64 bytes from 203.35.200.139: icmp_seq=1 ttl=255 time=1.106 ms
64 bytes from 203.35.200.139: icmp_seq=2 ttl=255 time=1.094 ms

--- phoenix.welearn.com.au ping statistics ---
3 packets transmitted, 3 packets received, 0% packet loss
round-trip min/avg/max/stddev = 1.094/1.129/1.187/0.041 ms



-- 

Regards,
        -*Sue*-

Comment 3 sue 2000-11-01 09:07:03 UTC

Fixed one machine.

I didn't think a reboot would help, so I tried to reverse the changes
to files in /usr/lib by replacing with files from CD2. This resulted
in a forced reboot of a confused system, finishing the upgrade-reversal
in single user mode, and coming up fine after that. Does it work because
I
replaced all those new files with the old ones, or the reboot, or both?
Dunno. I've got my machine back. I hope nobody else gets trapped without
the CD.

Comment 4 Satoshi Asami freebsd_committer

2000-11-12 06:48:39 UTC

State Changed
From-To: open->closed

Package removed, thanks.