tmpbridge is the name of the bridge and networking not working

[gmoore777]

I was under the impression that the name of the default bridge would be "eth0"
(or "peth0" or "xenbr0"), instead I see "tmpbridge" as having been created.

In either case, I can't ssh to Dom0 from DomU, nor the reverse.
Nor can I ping any machine on the Internet from DomU.

I think xen-tools is not working correctly
on Ubuntu PrecisePangolin 12.04 with xen-tools 4.2.1-1.

I also see my Network Manager icon constantly reporting
that something has been disconnected every 30 seconds or so.

Output from `sudo ifconfig -a` on Dom0:

eth0 Link encap:Ethernet HWaddr 00:22:19:2c:b9:db
          inet addr:192.168.0.102 Bcast:192.168.0.255 Mask:255.255.255.0
          inet6 addr: fe80::222:19ff:fe2c:b9db/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
          RX packets:136340 errors:0 dropped:0 overruns:0 frame:0
          TX packets:93781 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:176393891 (176.3 MB) TX bytes:14595539 (14.5 MB)
          Interrupt:21 Memory:fe6e0000-fe700000

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:4115 errors:0 dropped:0 overruns:0 frame:0
          TX packets:4115 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:470107 (470.1 KB) TX bytes:470107 (470.1 KB)

tmpbridge Link encap:Ethernet HWaddr 92:45:75:85:57:a7
          BROADCAST MULTICAST MTU:1500 Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)

Contents of one of my VM's config file: /etc/xen/xen002.cfg:

kernel = '/boot/vmlinuz-3.2.0-24-generic'
ramdisk = '/boot/initrd.img-3.2.0-24-generic'
vcpus = '1'
memory = '1024'
root = '/dev/xvda2 ro'
disk = [
                  'phy:/dev/vg1/xen002-disk,xvda2,w',
                  'phy:/dev/vg1/xen002-swap,xvda1,w',
              ]
name = 'xen002'
vif = [ 'ip=192.168.0.201,mac=00:16:3E:25:6F:08' ]
on_poweroff = 'destroy'
on_reboot = 'restart'
on_crash = 'restart'

Contents of xend-config.sxp with commented and blank lines removed:
$ cat xend-config.sxp | grep -v "^#" | sed -e '/^$/d'

(network-script network-bridge)
(vif-script vif-bridge)
(dom0-min-mem 196)
(enable-dom0-ballooning yes)
(total_available_memory 0)
(dom0-cpus 0)
(vncpasswd '')

I also will add that I don't think the syntax of naming a bridge works either.
Meaning, If I change the network-script line to
(network-script 'network-bridge bridge=br3')
I don't see anything get created named "br3".

I have zero experience with Xen so there may be a lot of user
error here but I am willing and able to provide more tests and information.

[Axel Beckert (xtaran)]

That's likely not a bug in xen-tools as it does not fiddle with the bridges unless explicitly told to do so. Not sure which package else, though, maybe xen-utils-common?

[gmoore777 (guy-moore)]

i'm getting stuff to work.
Will post within a day with more information.

[gmoore777 (guy-moore)]

so the solution for me to get all this automajic bridging to work is to add this line to /etc/network/interfaces:
    iface eth0 inet dhcp

Such that my /etc/network/interfaces look like:
  auto lo
  iface lo inet loopback
  auto eth0
  iface eth0 inet dhcp

Then reboot.
That line seems to stop NetworkManager from managing that eth0 interface.
Which then seems to allow the /etc/xen/scripts/network-bridge script to be successful.

Such that after the reboot, this is the output of `sudo ifconfig -a`:

eth0 Link encap:Ethernet HWaddr 00:22:19:2c:b9:db
          inet addr:192.168.0.108 Bcast:192.168.0.255 Mask:255.255.255.0
          inet6 addr: fe80::222:19ff:fe2c:b9db/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
          RX packets:18370 errors:0 dropped:0 overruns:0 frame:0
          TX packets:16217 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:22075797 (22.0 MB) TX bytes:2582972 (2.5 MB)

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:7600 (7.6 KB) TX bytes:7600 (7.6 KB)

peth0 Link encap:Ethernet HWaddr 00:22:19:2c:b9:db
          inet6 addr: fe80::222:19ff:fe2c:b9db/64 Scope:Link
          UP BROADCAST RUNNING PROMISC MULTICAST MTU:1500 Metric:1
          RX packets:20195 errors:0 dropped:0 overruns:0 frame:0
          TX packets:16226 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:22774536 (22.7 MB) TX bytes:2653740 (2.6 MB)
          Interrupt:21 Memory:fe6e0000-fe700000

I now have a "peth0" interface.
And the output of `sudo brctl show` now shows:

  bridge name bridge id STP enabled interfaces
  eth0 8000.0022192cb9db no peth0

Rather than showing the broken "tmpbridge" bridge.

and once I bring up a virtual machine with: `sudo xl create /etc/xen/xen002.cfg`

I see that I have another interface successfully getting created on the dom0 machine:
vif1.0 Link encap:Ethernet HWaddr fe:ff:ff:ff:ff:ff
          inet6 addr: fe80::fcff:ffff:feff:ffff/64 Scope:Link
          UP BROADCAST RUNNING PROMISC MULTICAST MTU:1500 Metric:1
          RX packets:99 errors:0 dropped:0 overruns:0 frame:0
          TX packets:470 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:32
          RX bytes:10898 (10.8 KB) TX bytes:88627 (88.6 KB)

NetworkManager still periodically (every 60 seconds) pops up a message of:
  "Wired network. Disconnected"

so maybe I need to stop the NetworkManager service all together to stop its
pesky meddling with these newly created interfaces or at least have it stop
the irritating, and, I think, non-fatal messages.

I also now have ssh connectivity between my dom0 and domU and vice-versa.
(although, I have to use IP addresses and not the hostnames. Evidently, the DNS
 server does not know about these hosts?)

So I think this "bug"...