Topic: interface eth1 startup failing

[spelchek]

Hi all,

At prsent, my SME server 5.5 cannot log into telsra cable network (as in my previous post).

On looking through my messages log , and comparing with an older messages log i have found that on bootup, my eth1 interface fails at startup.

Eth1 is configured to be the ethernet card that my cable modem is connected to.

On bootup of server, it sems that eth0 interface starts fine, but the eth1 fails at the DHCP request stage.

here are extracts from the 2 different messages logs.

Firstly the 'new' log.. where eth1 does not start succesfully:

Apr 29 04:33:13 e-smith network: Bringing up interface eth1:  
Apr 29 04:33:13 e-smith kernel: rtl8139.c:v1.07 5/6/99 Donald Becker http://cesdis.gsfc.nasa.gov/linux/drivers/rtl8139.html
Apr 29 04:33:13 e-smith kernel: eth1: SMC1211TX EZCard 10/100 (RealTek RTL8139) at 0xe000, IRQ 10, 00:e0:29:9d:b8:e8.
Apr 29 04:33:13 e-smith ifup: Determining IP information for eth1...
Apr 29 04:33:14 e-smith dhcpcd[737]: broadcasting DHCP_REQUEST for 112.245.255.163
Apr 29 04:33:14 e-smith ifup: ./ifup: line 173:   737 Segmentation fault      /sbin/dhcpcd ${DHCPCDARGS} ${DEVICE}
Apr 29 04:33:14 e-smith ifup:  failed.
Apr 29 04:33:14 e-smith network: Determining IP information for eth1..../ifup: line 173:   737 Segmentation fault      /sbin/dhcpcd ${DHCPCDARGS} ${DEVICE}
Apr 29 04:33:14 e-smith network:  failed.
Apr 29 04:33:14 e-smith network: Bringing up interface eth1:  failed
Apr 29 04:33:14 e-smith network:
Apr 29 04:33:14 e-smith rc: Starting network:  succeeded

Now the relevant extract from the 'old' msgs log (where eth1 starts OK).

Apr  6 00:49:51 e-smith network: Bringing up interface eth1:  
Apr  6 00:49:51 e-smith kernel: rtl8139.c:v1.07 5/6/99 Donald Becker http://cesdis.gsfc.nasa.gov/linux/drivers/rtl8139.html
Apr  6 00:49:51 e-smith kernel: eth1: SMC1211TX EZCard 10/100 (RealTek RTL8139) at 0xe000, IRQ 10, 00:e0:29:9d:bd:cb.
Apr  6 00:49:51 e-smith ifup: Determining IP information for eth1...
Apr  6 00:49:51 e-smith dhcpcd[737]: broadcasting DHCP_DISCOVER
Apr  6 00:49:51 e-smith dhcpcd[737]: broadcasting second DHCP_DISCOVER
Apr  6 00:49:51 e-smith dhcpcd[737]: DHCP_OFFER received from  (172.18.17.81)
Apr  6 00:49:51 e-smith dhcpcd[737]: broadcasting DHCP_REQUEST for 138.130.223.63
Apr  6 00:49:51 e-smith dhcpcd[737]: DHCP_ACK received from  (172.18.17.81)
Apr  6 00:49:51 e-smith ifup: dhcpcd: your IP address = 138.130.223.63
Apr  6 00:49:51 e-smith ifup:  done.
Apr  6 00:49:51 e-smith ifup: Can't ignore signal CHLD, forcing to default.
Apr  6 00:49:52 e-smith e-smith[808]: Processing event: ip-change 138.130.223.63
Apr  6 00:49:52 e-smith e-smith[808]: Running event handler: /etc/e-smith/events/ip-change/S15set-external-ip
Apr  6 00:49:52 e-smith network: Bringing up interface eth1:  succeeded
Apr  6 00:49:52 e-smith network: ^[[60G
Apr  6 00:49:52 e-smith network:
Apr  6 00:49:52 e-smith rc: Starting network:  succeeded

In the 'failed to start' instance, note this line:

Apr 29 04:33:14 e-smith dhcpcd[737]: broadcasting DHCP_REQUEST for 112.245.255.163

The IP address listed here is NOT one I have entered..I have no idea how it got here.. in the 'successful' startup case the same line is:

Apr  6 00:49:51 e-smith dhcpcd[737]: broadcasting DHCP_REQUEST for 138.130.223.63




also note that in the second log extract (the successful startup) that these lines are in the log, but do not appear in the log for the unssuccesful eth1 startup.


Apr  6 00:49:51 e-smith dhcpcd[737]: broadcasting DHCP_DISCOVER
Apr  6 00:49:51 e-smith dhcpcd[737]: broadcasting second DHCP_DISCOVER
Apr  6 00:49:51 e-smith dhcpcd[737]: DHCP_OFFER received from  (172.18.17.81)


Can anyone shed light on what is going on here?? Im assuming that the lack of success at initialising the eth1 interface is the root of my 'unable to log into telstra cable' problems... I dont seem to be getting dhcp assigned IP to eth01.

Any help much appreciated.

[jvdbossc]

I started with your type of network cards and replaced them both.

The most recent one just locked after a few minutes an the older realtek card worked fine  

I did not manage in having a stable internet connection with them.

Second:  Are you sure, you are not mixing up the connections?? (lan-internet)

Third:  Do you need to clone mac-adresses??

Fourth:  If you get an adress, is there an old gateway like a linksys-router on your network?

[Anonymous]

Card has been faultless for 3 1/2 years.

Put a new one (same type) strait outta the box into the server to test... same result

Connect has been fine until now.

Not mixing up lan/internet address (i dont think anyway )

how does one clone a mac address?? and why would i need to??

[Anonymous]

Card has been faultless for 3 1/2 years.

Put a new one (same type) strait outta the box into the server to test... same result

Connect has been fine until now.

Not mixing up lan/internet address (i dont think anyway )

how does one clone a mac address?? and why would i need to??


Thanks for your reply

[jvdbossc]

Some providers let you only work with one mac adress, and that is the one provided by you, when you first started your connection.

If you use the same network card, or you had previously a working network, I don't think it will matter.

If you replaced your network card, you should check with your provider, if they need to change your mac.

(excuse: My network card was damaged..)

Actually I do not know how to change it on sme, (At my location, I have to wait for 2 hours)

I noticed some posts about it, a while back.