I have set up a 4-NIC bond on my Debian Wheezy box. After a fresh boot, the system has LAN access, but cannot resolve outside host names or access outside IPs. It seems that this is because the bond0 interface is not getting a default gateway route, but instead one of the slave interfaces (usually eth3) is getting it.
Result of route -n after booting:
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
0.0.0.0 192.168.2.1 0.0.0.0 UG 0 0 0 eth3
169.254.0.0 0.0.0.0 255.255.0.0 U 1000 0 0 eth3
192.168.2.0 0.0.0.0 255.255.255.0 U 0 0 0 bond0
192.168.2.0 0.0.0.0 255.255.255.0 U 0 0 0 eth3
My immediate network problem could be alleviated by doing: route add default gw 192.168.2.1 bond0. But I at least need the computer to get outside network access after a reboot when I may not have physical access.
The supposedly slaved interface is getting its own IP address every single time. I suspect that may be the root of the problem. What is going on here? Result of ifconfig, note eth3:
bond0 Link encap:Ethernet HWaddr bc:5f:f4:be:4a:cc
inet addr:192.168.2.3 Bcast:192.168.2.255 Mask:255.255.255.0
inet6 addr: fe80::be5f:f4ff:febe:4acc/64 Scope:Link
UP BROADCAST RUNNING MASTER MULTICAST MTU:1500 Metric:1
RX packets:522992 errors:0 dropped:4871 overruns:0 frame:0
TX packets:885760 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:53459302 (50.9 MiB) TX bytes:736701670 (702.5 MiB)
eth0 Link encap:Ethernet HWaddr bc:5f:f4:be:4a:cc
UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1
RX packets:85101 errors:0 dropped:5 overruns:0 frame:0
TX packets:412890 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:8568619 (8.1 MiB) TX bytes:37311255 (35.5 MiB)
Memory:f7400000-f7480000
eth1 Link encap:Ethernet HWaddr bc:5f:f4:be:4a:cc
UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1
RX packets:1285 errors:0 dropped:12 overruns:0 frame:0
TX packets:75 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:185579 (181.2 KiB) TX bytes:16930 (16.5 KiB)
Memory:f7300000-f7380000
eth2 Link encap:Ethernet HWaddr bc:5f:f4:be:4a:cc
UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1
RX packets:2039 errors:0 dropped:11 overruns:0 frame:0
TX packets:472583 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:326122 (318.4 KiB) TX bytes:699347095 (666.9 MiB)
Memory:f7200000-f7280000
eth3 Link encap:Ethernet HWaddr bc:5f:f4:be:4a:cc
inet addr:192.168.2.135 Bcast:192.168.2.255 Mask:255.255.255.0
UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1
RX packets:434565 errors:0 dropped:410 overruns:0 frame:0
TX packets:210 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:44378800 (42.3 MiB) TX bytes:26282 (25.6 KiB)
Memory:f7100000-f7180000
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:78 errors:0 dropped:0 overruns:0 frame:0
TX packets:78 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:9310 (9.0 KiB) TX bytes:9310 (9.0 KiB)
and /etc/network/interfaces is like so:
auto lo
iface lo inet loopback
auto eth0
iface eth0 inet manual
bond-master bond0
bond-mode 4
auto eth1
iface eth1 inet manual
bond-master bond0
bond-mode 4
auto eth2
iface eth2 inet manual
bond-master bond0
bond-mode 4
auto eth3
iface eth3 inet manual
bond-master bond0
bond-mode 4
auto bond0
iface bond0 inet static
address 192.168.2.3
netmask 255.255.255.0
gateway 192.168.2.1
dns-nameservers 192.168.2.1
slaves eth0 eth1 eth2 eth3
bond-slaves none
bond-mode 4
bond-miimon 100
bond-downdelay 200
bond-updelay 200
I have tried several combinations of ports on my switch, and each time come up with roughly the same result (sometimes a different interface than eth3). This problem did arise after upgrading my motherboard; with the old board I had a 2-NIC bond, which worked fine. (Is there any possible issue with upgrading hardware on an existing installation of Wheezy?) After upgrading, I removed the old /etc/udev/rules.d/70-persistent-net.rules file, and let it regenerate, so the file now contains:
# PCI device 0x8086:/sys/devices/pci0000:00/0000:00:1c.2/0000:07:00.0 (igb)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="bc:5f:f4:be:4a:cc", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="eth*", NAME="eth0"
# PCI device 0x8086:/sys/devices/pci0000:00/0000:00:1c.4/0000:09:00.0 (igb)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="bc:5f:f4:be:4a:bd", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="eth*", NAME="eth2"
# PCI device 0x8086:/sys/devices/pci0000:00/0000:00:1c.3/0000:08:00.0 (igb)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="bc:5f:f4:be:4a:ce", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="eth*", NAME="eth1"
# PCI device 0x8086:/sys/devices/pci0000:00/0000:00:1c.5/0000:0a:00.0 (igb)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="bc:5f:f4:be:4a:bf", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="eth*", NAME="eth3"
One last thing, I did try adding the following to my /etc/network/interfaces file, hoping to get a sort of 'work-around' solution.
auto bond0
...
post-up route add default gw 192.168.2.1 bond0
pre-down route del default gw 192.168.2.1 bond0
That seems to make no difference, and after a boot, there is no default GW route for bond0 in the routing table (it is exactly as above). I would love to know what I'm doing wrong there, and at least use that as an interim fix.
