[24277] in Athena Bugs
Re: Linux Athena boxes hanging on update 9.2.16 to 9.2.20
daemon@ATHENA.MIT.EDU (Greg Hudson)
Wed Sep 24 17:41:21 2003
From: Greg Hudson <ghudson@MIT.EDU>
To: Tom Cavin <cavin@mit.edu>
Cc: Athena Bugs list <bugs@mit.edu>
In-Reply-To: <16242.818.880235.171197@lap1-wccf.mit.edu>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Message-Id: <1064439660.3579.32.camel@error-messages.mit.edu>
Mime-Version: 1.0
Date: Wed, 24 Sep 2003 17:41:13 -0400
So, it looks like these machines hung in "kudzu -q", which we run after
rpmupdate so that machines don't display new-hardware-detected dialogs
at boot time after a kernel update. The stack trace suggests some kind
of issue with the firewire drivers:
-bash-2.05b# strace -p 2193
wait4(2200, <unfinished ...>
(gdb) back
#0 0xffffe002 in ?? ()
#1 0x08052efc in compareDevice ()
#2 0x080530c3 in compareDevice ()
#3 0x080601e0 in firewireNewDevice ()
#4 0x08051d64 in compareDevice ()
#5 0x0804fc2a in strcpy ()
#6 0x420156a4 in __libc_start_main () from /lib/tls/libc.so.6
When the machines boot, they don't apparently display a kudzu window.
I'm not sure what happened here, or whether it will happen again the
next time the kernel updates. I'm not sure if we're triggering weird
behavior by running kudzu after the kernel has updated but before it has
booted. Andrew, any ideas?
