.1861 Oops enabling eth0/eth1?

Clyde E. Kunkel clydekunkel7734 at cox.net
Thu Jan 19 21:30:35 UTC 2006


Tom London wrote:
> OK. This now has me stumped....
> 
> Booting with 'vga-791' pauses for about 1.5 minutes trying to enable
> eth0, but now it just fails and goes on to eth1 which also fails. 
> (I'm running off a different network than before, so perhaps something
> about the DHCP server is different, etc.).
> 
> The system completes booting, but the network is 'dead'. I noticed
> this in /var/log/messages:
> 
> Jan 19 12:25:59 localhost kernel: e1000: eth1: e1000_probe: Intel(R)
> PRO/1000 Network Connection
> <<<<SNIP>>>>
> Jan 19 12:26:03 localhost kernel: ip_tables: (C) 2000-2006 Netfilter Core Team
> Jan 19 12:26:03 localhost kernel: Netfilter messages via NETLINK v0.30.
> Jan 19 12:26:03 localhost kernel: ip_conntrack version 2.4 (8192
> buckets, 65536 max) - 232 bytes per conntrack
> Jan 19 12:26:03 localhost kernel: e1000: eth0: e1000_watchdog_task:
> NIC Link is Up 100 Mbps Full Duplex
> Jan 19 12:26:03 localhost kernel: e1000: eth0: e1000_watchdog_task:
> NIC Link is Up 100 Mbps Full Duplex
> <<<SNIP>>>
> Jan 19 12:26:53 localhost dhclient: send_packet: Network is unreachable
> Jan 19 12:26:53 localhost dhclient: send_packet: please consult README
> file regarding broadcast address.
> 
> So I created an /etc/iftab file like:
> # tbl
> eth0 mac 00:0E:7B:4A:79:0B
> eth1 mac 00:0E:35:36:F2:34
> 
> and rebooted.  This time I get:
> Jan 19 12:42:44 localhost kernel: e1000: 0000:02:09.0: e1000_probe:
> (PCI:33MHz:32-bit) 00:0e:7b:4a:79:0b
> Jan 19 12:42:44 localhost kernel: e1000: eth0: e1000_probe: Intel(R)
> PRO/1000 Network Connection
> 
> so the device now seems matched to the proper interface. But:
> Jan 19 12:42:50 localhost kernel: e1000: eth0: e1000_watchdog_task:
> NIC Link is Up 100 Mbps Full Duplex
> 
> but no joy on eth0. However, once booted, I can use the nice
> NetworkManager applet to select the wireless network sucessfully. (Not
> the hardwired one, just the ipw2200 one).
> 
> Could there be something funny with the e1000 driver?   Something else?
> 
> tom
> 
e1000 is broken in .1861 kernel.

see bzs 178340 and 178350

-- 
Regards,

Old Fart
(my reply-to address is "munged" to defeat spambots)




More information about the fedora-test-list mailing list