Bridged networking on openSuSE 11.4 with Xen ...

English Network/Internet
1

Newbie Question:

How do I setup networking on openSuSE 11.4 on an HP Proliant DL385 G7 with 4 NICs to host Windows 2008 R2 on Xen?

I’ve installed openSuSE 11.4 with all the patches (and most of the server patterns - Mail and News, LAMP, LDAP, Samba, etc.), and I’ve installed Windows 2008 R2 in a Xen virtual machine. I’m having problems configuring the 4 NICs

eth0, eth1, eth2, eth3 are “bound” without an ip_address as br0

eth0 connects to my ISP - currently through a private LAN - It connects directly to the router with static IP 192.168.0.105

Eventually, one of the other NICs will serve ip_addresses via DHCP to a separate LAN (for use in an office setting), and eth0 will be set with a static IP from the ISP That means, eth0 will be static to the ISP, and the other NICs will attach to a switch serving private ip_addresses in the 192.168.0.xxx range.

Currently, I have br0 unconfigured and eth0 static.

I have configured eth0 in the External Zone and br0 in the Internal Zone in the firewall, and all the correct ports are opened (afaik) I have enabled masquerading. Hostname, Domain (workgroup), DNS Server addresses and IPv4 Gateway are configured.

Windows 2008 R2 (Guest VM in Xen) “sees” the other machines on the network, and “browses” the internet, although it will not download patches except intermittently. I have not tried connecting to it from the LAN Eventually, I wish to run Windows Terminal Server.

openSuSE cannot “browse” the internet, though it initially did.

What am I missing? I “think” I need routing or NAT, or I may have my bridge setup incorrectly, although I’ve tried almost every combination.

Google says Xen should be setup with either NAT or a Bridged Network.

Output of ifconfig:

br0 Link encap:Ethernet HWaddr 3C:4A:92:76:B8:F2
inet6 addr: fe80::3e4a:92ff:fe76:b8f2/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:953 errors:0 dropped:0 overruns:0 frame:0
TX packets:4 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:119849 (117.0 Kb) TX bytes:368 (368.0 b)

eth0 Link encap:Ethernet HWaddr 3C:4A:92:76:B8:F2
inet addr:192.168.0.105 Bcast:0.0.0.0 Mask:255.255.255.255
inet6 addr: fe80::3e4a:92ff:fe76:b8f2/64 Scope:Link
UP BROADCAST RUNNING PROMISC MULTICAST MTU:1500 Metric:1
RX packets:9259 errors:0 dropped:0 overruns:0 frame:0
TX packets:588 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:2996219 (2.8 Mb) TX bytes:75044 (73.2 Kb)
Interrupt:44 Memory:f6000000-f6012800

eth1 Link encap:Ethernet HWaddr 3C:4A:92:76:B8:F4
UP BROADCAST PROMISC 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:1000
RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)
Interrupt:45 Memory:f4000000-f4012800

eth2 Link encap:Ethernet HWaddr 3C:4A:92:76:B8:F6
UP BROADCAST PROMISC 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:1000
RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)
Interrupt:51 Memory:fa000000-fa012800

eth3 Link encap:Ethernet HWaddr 3C:4A:92:76:B8:F8
UP BROADCAST PROMISC 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:1000
RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)
Interrupt:48 Memory:f8000000-f8012800

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:295901 errors:0 dropped:0 overruns:0 frame:0
TX packets:295901 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:57617543 (54.9 Mb) TX bytes:57617543 (54.9 Mb)

tap1.0 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:1442 errors:0 dropped:0 overruns:0 frame:0
TX packets:2486 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:500
RX bytes:171592 (167.5 Kb) TX bytes:698967 (682.5 Kb)

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:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:1848 overruns:0 carrier:0
collisions:0 txqueuelen:32
RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)

Thanks in advance.

2

have you tried bridging eth{0,1,2,3} and setting the br0 IP to what eth0’s IP was?

as a side note, I have a reservation on your set up, you shouldn’t put all your eggs in the same basket!
imho, you should add a routing server in charge of access control and nat’ing to outside.
and the “big” server should only do virtualization.

3

For starters, you might consider your multi-server setup as “not typical” and might need some Linux Network TCP/IP tuning. Recently I aggregated and re-worded a ton of information on the Internet about TCP/IP settings and Congestion Control algorithms.

Without more information and analysis of your network traffic, it’s not possible to know for sure what your networking problem(s) are, but lack of proper TCP/IP configuration could definitely cause the symptoms you describe, particularly if your networking links aren’t optimum quality (ie no wireless, low noise, short hops or dedicated links) and small files.

https://sites.google.com/site/4techsecrets/optimize-and-fix-your-network-connection

Try following the instructions if you have sufficient RAM to increase your TCP/IP buffer sizes, turn on various TCP/IP options and if necessary modify or trial various Congestion Control algorithms.

HTH,
Tony

4

Not sure if anyone is interested in the outcome of this issue, but I’ll add a few brief comments to close the thread.

Essentially, Broadcom NICs do not do well under virtualization, and I had 4 whether bridged or not.

I abandoned Xen because of the confusion with bridging the DOM and host, and KVM experienced a flooding of messages in virtmanager -

virtmanager floods /var/log/manages

I installed VMware ESXi 5.0, and it manages network traffic wonderfully. It emulates an IBM e1000 NIC instead of installing drivers for the Broadcom NICs

I have openSuSE 11.4 x64 installed as a VM Guest and two instances of Windows Server 2008 R2 My troubles now lie with Windows 2008 R2 and Samba.

Using ESXi as a hypervisor host simplifies virtual machine creation, and it’s considered “best practices” in virtualizing server instances. It sure simplified things for me.

Thanks for the suggestions. I followed them at the time, and it led me to other choices that were not evident when I initially posted the questions.