eth0 ceases to function after reboot on FC5T3

[David Timms]

Dane Mutters wrote:
> Thanks for the suggestions, Dave.  Here's some info:
> 
>> Can you give the result of ifconfig ?
> 
> eth0      Link encap:Ethernet  HWaddr FE:FF:FF:FF:FF:FF  
>           inet addr:192.168.1.112  Bcast:192.168.1.255  Mask:255.255.255.0
I don't think I have seen any machine not show the MAC address properly; 
this one (HWaddr) does not seem legit, but I have no idea why.
...
> peth0     Link encap:Ethernet  HWaddr FE:FF:FF:FF:FF:FF  
>           inet6 addr: fe80::fcff:ffff:feff:ffff/64 Scope:Link
>           UP BROADCAST RUNNING NOARP MULTICAST  MTU:1500  Metric:1
>           RX packets:12227 errors:0 dropped:736 overruns:0 frame:0
>           TX packets:610 errors:0 dropped:0 overruns:0 carrier:0
>           collisions:0 txqueuelen:1000 
>           RX bytes:914379 (892.9 KiB)  TX bytes:44795 (43.7 KiB)
>           Interrupt:11 Base address:0x2c00 
> 
> vif0.0    Link encap:Ethernet  HWaddr FE:FF:FF:FF:FF:FF  
>           inet6 addr: fe80::fcff:ffff:feff:ffff/64 Scope:Link
>           UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
>           RX packets:255 errors:0 dropped:0 overruns:0 frame:0
>           TX packets:11682 errors:0 dropped:0 overruns:0 carrier:0
>           collisions:0 txqueuelen:0 
>           RX bytes:12811 (12.5 KiB)  TX bytes:869926 (849.5 KiB)
> 
> xenbr0    Link encap:Ethernet  HWaddr FE:FF:FF:FF:FF:FF  
>           inet6 addr: fe80::200:ff:fe00:0/64 Scope:Link
>           UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
>           RX packets:12157 errors:0 dropped:0 overruns:0 frame:0
>           TX packets:6 errors:0 dropped:0 overruns:0 carrier:0
>           collisions:0 txqueuelen:0 
>           RX bytes:725575 (708.5 KiB)  TX bytes:468 (468.0 b)
...
So you are doing the xen thing, or did you not mean to install xen ? 
Maybe try uninstalling the xen parts: again, I have no experience with this.
>> When you said currently: what method have you been using to try
>> different ethernet settings ?
> 
> I have tried using static and DHCP.  Currently I'm using static.
What I meant was are you hand editing config files or are you using 
system-config-network (System|Admin|Network) ?

# rpm -q system-config-network

What does system-config-network|eth0|edit|hardware device|Mac address show ?
Is the bind to MAC address ticked ?
Does clicking probe update / put in a real address ?

>> What is the ethernet adapter ?
> Linksys NC100 Network Everywhere Fast Ethernet 10/100 (rev. 11)
Under what OS was the card last working ?
If you dual boot, does the card work OK in the other OS ?
Is this an upgrade install ?
...
>> What is the result of lspci ?
...
> 01:09.0 Ethernet controller: Linksys NC100 Network Everywhere Fast Ethernet 
...
>> How about lsmod ?
> 
> Module                  Size  Used by
...
> tulip                  52449  0 
...
I can only guess that tulip is the correct driver for your card
If another linux used previously was it previously using the tulip module ?
Can you find out which eth chip your linksys card actually uses ?

>> I don't know, perhaps there is an error being logged in
>> /var/log/messages ? grep eth0 /var/log/messages that can shed more light
>> on the problem.
> 
> Feb 24 10:30:41 Spot kernel: eth0: ADMtek Comet rev 17 at cf076c00, 
> 00:04:5A:55:96:57, IRQ 11.
> Feb 24 10:30:45 Spot kernel: eth0: Setting full-duplex based on MII#1 link 
> partner capability of 45e1.
It seems to be detecting the link carrier, and going in full duplex OK. 
Is the netcard and switch / hub showing link activity ?

> Feb 24 02:32:18 Spot kernel: peth0: Promiscuous mode enabled.
> Feb 24 02:32:18 Spot kernel: device peth0 entered promiscuous mode
Someone else know if this is normal for xen installs ?
(this is normal when doing packet captures, but why here ?)

...
>> In any case to make a bugzilla report you would need to provide info
>> like the above; might as well provide it here first, so that there is
>> enough info for others to help/refer you to bugzilla etc. Also, perhaps
>> provide the age and manufacturer/model of the machine (or component
>> parts if hand assembled).
> 
> It's a 6-year-old Gateway with a Pentium 3 800MHz processor.
> 
> Also, I seem to have forgotten an important detail.  I get this error message 
> when disabling the device (eth0) in the network configuration tool in Gnome:
> 
> Device eth0 has MAC address FE:FF:FF:FF:FF:FF, instead of configured address 
> 00:04:5A:55:96:57. Ignoring.
Do you know/can you find out the correct address and enter it in 
sys-con-network, in the area mentioned above ?

My guess is: for some reason the tools aren't able to retrieve the 
correct MAC address. Your machine will happily send out data with a 
return MAC address that is incorrect, which the eth driver ignores 
(since not destined to itself), leaving you in network isolation.

> Hope that helps!
As you can see it's not a one step troubleshooting process ;-)

DaveT.