kernel 2.6.34 shutdown

[GoinEasy9]

Ever since the kernel has upgraded to 2.6.34.0.10 I've had a lockup on shutdown. This didn't occur with 2.6.34.0.09, and doesn't occur when I slid in a disk into the same machine with F13 running. I took out rhgb quiet from the kernel line and see that there is an eth0 dump, then the line:
nv_stop_tx: TransmitterStatus remained busy.
This is the last line before the shutdown locks up. It also occurs with shutdown -h now. Googling I can only find occurrences of this when network cards lock up and the corresponding networks fail, and the old reasons were bugs in the forcedeth driver. Well, this is on shutdown, and, since I have no problems with my network or network card using other versions of Fedora or other distros in the same machine, I am trying to find out what was different between the present kernel and the last kernel that worked.

The scroll is too fast to see what happens before the eth0 dump, so, does anyone know of a log or a way of vectoring the shutdown scroll to a log so I can see what is going on. Messages, dmesg and the other logs don't seem to have the info I'm looking for.

Thanks in Advance.

Edit: I'm up to kernel-2.6.34-0.28.rc3.git3.fc14 now, it doesn't seem it's going to get fixed unless I can report something.

[JEO]

Have you tried turning off the interface manually like:

service NetworkManager stop
ifconfig eth0 down

tail -n 50 /var/log/messages

[GoinEasy9]

Thanks for the idea JEO

Apr 11 16:13:12 localhost NetworkManager: <WARN> nm_signal_handler(): Caught signal 15, shutting down normally.
Apr 11 16:13:12 localhost NetworkManager: <info> exiting (success)
Apr 11 16:13:29 localhost avahi-daemon[1175]: Interface eth0.IPv4 no longer relevant for mDNS.
Apr 11 16:13:29 localhost avahi-daemon[1175]: Leaving mDNS multicast group on interface eth0.IPv4 with address 192.168.1.102.
Apr 11 16:13:29 localhost dhclient[1170]: receive_packet failed on eth0: Network is down
Apr 11 16:13:29 localhost avahi-daemon[1175]: Withdrawing address record for fe80::226:18ff:fe32:33d9 on eth0.
Apr 11 16:13:29 localhost avahi-daemon[1175]: Withdrawing address record for 192.168.1.102 on eth0.

As I suspected. networking shutdown and restarted normally. It didn't lock up the way it does on shutdown. The last lines from messages during shutdown last night really don't show anything either, at least not to my eyes:

Apr 10 22:17:30 localhost auditd[1031]: The audit daemon is exiting.
Apr 10 22:17:30 localhost kernel: type=1305 audit(1270952250.801:16): audit_pid=0 old=1031 auid=4294967295 ses=4294967295 res=1
Apr 10 22:17:31 localhost cpuspeed: Disabling ondemand cpu frequency scaling governor
Apr 10 22:17:31 localhost kernel: Kernel logging (proc) stopped.
Apr 10 22:17:31 localhost rsyslogd: [origin software="rsyslogd" swVersion="4.4.2" x-pid="1058" x-info="http://www.rsyslog.com"] exiting on signal 15.

Still trying to find a way to log the shutdown scroll.

[SlowJet]

http://koji.fedoraproject.org/koji/packageinfo?packageID=8

You've got to be kidding. 33-49 is almost production and 34-xx is not even oven pre heated yet.

$ rpm -qa kernel\*
kernel-PAE-2.6.33.2-43.fc13.i686
kernel-PAE-2.6.33.2-46.fc13.i686
kernel-PAE-devel-2.6.33.2-49.fc13.i686
kernel-PAE-2.6.33.2-49.fc13.i686
kernel-headers-2.6.33.2-49.fc13.i686
kernel-PAE-2.6.33.2-47.fc13.i686


SJ

[JEO]

2.6.34 are the standard .fc14 rawhide kernels. 2.6.33 are the .fc13 ones.

[SlowJet]

Yeah, I didn't check the forum to see it was rawhide.
But e9 is always yelling at the sky about the moon. :)

SJ

[GoinEasy9]

New kernel in tonight, 2.6.34-0.49.rc7.git0.fc14.i686.PAE, solved my shutdown problems. \o/ yay.

[kyryder]

2.6.34-0.49.rc7.git0.fc14.i686.PAE also working well here so far.

Ky