LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Linux - Networking (https://www.linuxquestions.org/questions/linux-networking-3/)
-   -   resolvconf fails to add nameservers from static interface (https://www.linuxquestions.org/questions/linux-networking-3/resolvconf-fails-to-add-nameservers-from-static-interface-4175411311/)

decibelcooper 06-13-2012 07:29 PM

resolvconf fails to add nameservers from static interface
 
Hi all.

I have a server running Ubuntu 12.04 Server that acts as, among other things, a dhcp server and gateway for a local network. It has one ethernet interface (eth0) that is on the local network, and another (eth1) that is connected to a WAN. Both interfaces are defined as static in /etc/network/interfaces

Quote:

iface lo inet loopback
auto lo

# The primary network interface
iface eth0 inet static
address 192.168.0.1
netmask 255.255.255.0
auto eth0

iface eth1 inet static
address 129.219.144.14
netmask 255.255.255.192
gateway 129.219.144.1
dns-nameservers 8.8.8.8 8.8.4.4
up ethtool --offload tx off rx off
auto eth1
It has worked flawlessly in this manner for a little while, and then apparently some automatic security update has occured and has broken something with the eth1 interface, causing resolvconf to not add the nameservers for the statically defined eth1. The interface still works, and the server still works fine as a gateway, but when administering the server, domains cannot be resolved because /etc/resolv.conf contains only a header.

When calling "sudo ifdown eth1", I get

Quote:

ifdown: interface eth1 not configured
When calling "sudo ifup eth1", I get

Quote:

RTNETLINK answers: File exists
Failed to bring up eth1.
For "ifconfig", I get

Quote:

eth0 Link encap:Ethernet HWaddr 00:30:67:a5:0a:65
inet addr:192.168.0.1 Bcast:192.168.0.255 Mask:255.255.255.0
inet6 addr: fe80::230:67ff:fea5:a65/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:5839 errors:0 dropped:0 overruns:0 frame:0
TX packets:6228 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:1071957 (1.0 MB) TX bytes:4594693 (4.5 MB)
Interrupt:44 Base address:0xe000

eth1 Link encap:Ethernet HWaddr d8:9d:b9:00:08:de
inet addr:129.219.144.14 Bcast:129.219.144.63 Mask:255.255.255.192
inet6 addr: fe80::da9d:b9ff:fe00:8de/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:24416 errors:0 dropped:36 overruns:0 frame:0
TX packets:5133 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:6431777 (6.4 MB) TX bytes:1002944 (1.0 MB)
Interrupt:45

lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:100 errors:0 dropped:0 overruns:0 frame:0
TX packets:100 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:7608 (7.6 KB) TX bytes:7608 (7.6 KB)

So the interface is somehow successfully being brought up on boot, even though manually calling ifdown and ifup fails.

Now, I could bandaid the setup by putting my nameservers in the resolvconf header file, but I would rather get to the bottom of this. My question is then, how can I go about tracking down the cause of this problem, and/or how can I fix it?

Thanks in advance.

decibelcooper 06-13-2012 08:03 PM

Wow. This one will probably go in the record books for fastest solved problem on LQ. I found the problem when reading my own post. In /etc/network/interfaces, I noticed that I have "up ethtool --offload tx off rx off". I was missing the interface name ("eth1") in that command. I fixed this, and all is well. Although, I have no idea why this failed command caused such problems.

So I tried to blame it on some automatic update, but I obviously caused this problem, when I noticed that having the cpu calculate checksums on eth1 results in less errors.

Thanks anyway, esteemed LQ users ;).


All times are GMT -5. The time now is 12:23 PM.