Host machine was running F7 no problems, upgraded via yum to F8, no
problems with deps, all went smoothly, installed latest fedora-updates
versions of everything.
Rebooted host, but found I had lost contact with it over the LAN, even
though the machine is on a datacentre this not a disaster because I
have remote serial port access to this machine, so I PXE booted into
rescue mode from an F8 repo I have locally, modified grub.conf to set
xen and linux to use serial console and got logged in as root OK
Networking showed the eth0 had been renamed to peth0, the eth0 bridge
created, peth0 enslaved to eth0 (virbr0 and vnet0 bridges also created
but I don''t actually use those)
The problem is the IP address that ifcfg-eth0 assigns statically to
[p]eth0 ends up assigned assigned to NIC eth1 instead of bridge eth0
(won''t work as that NIC is connected to a different VLAN intended for
iSCSI access).
As soon as I remove the correct IP address from the wrong NIC with
ip addr del x.x.x.x/24 dev eth1
ip link set down eth1
the same IP address somehow gets re-incarnated on the eth0 bridge!
So I manually remove all addresses and shutdown all interfaces, and
restart networking
ip addr del x.x.x.x/24 dev eth0
ip link set down eth0
brctl delif eth0 peth0,
brctl delbr eth0,
virsh net-destroy virbr0
virsh net-destroy vnet0
ip link set name eth0 dev peth0 (to rename peth0 back to eth0)
service network start
this correctly brings up only eth0 and assigns correct IP address and routes
at this point "network-bridge status" seems confused about NICs, it
seems to see two copies of eth0 and not to see eth1
# /etc/xen/scripts/network-bridge status
===========================================================2: eth0:
<BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast qlen 1000
link/ether 00:04:23:d3:3c:10 brd ff:ff:ff:ff:ff:ff
inet 172.20.1.2/24 brd 172.20.1.255 scope global eth0
inet6 fe80::204:23ff:fed3:3c10/64 scope link
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast qlen
1000
link/ether 00:04:23:d3:3c:10 brd ff:ff:ff:ff:ff:ff
inet 172.20.1.2/24 brd 172.20.1.255 scope global eth0
inet6 fe80::204:23ff:fed3:3c10/64 scope link
valid_lft forever preferred_lft forever
bridge name bridge id STP enabled interfaces
172.20.1.0/24 dev eth0 proto kernel scope link src 172.20.1.2
169.254.0.0/16 dev eth0 scope link
default via 172.20.1.1 dev eth0
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
172.20.1.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0
169.254.0.0 0.0.0.0 255.255.0.0 U 0 0 0 eth0
0.0.0.0 172.20.1.1 0.0.0.0 UG 0 0 0 eth0
===========================================================
Then I then run "network-bridge start" the script properly renames
eth0 to peth0, re-creates the eth0 bridge, enslaves peth0 to eth0 and
moves the IP address, MAC address, routes over to eth0, everything is
working properly.
Yet following a reboot it goes wrong again just like before.
Any ideas what is confusing network-bridge on reboot, when it works if
manually run after networking has been shutdown?