[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-users] Fwd: Strange network problem



I have a similar setup: Gentoo 64, Xen-4, A790GXM-A motherboard, AMD Phenom(tm) II X4 940 Processor, two NICs (RTL8111/8168B) with the LAN based one in a bridge mode...

I tried the novell & pv-op dom0 kernels, with out much luck.  By using the disable TX checksumming tip (http://wiki.xensource.com/xenwiki/XenFaq#head-4ce9767df34fe1c9cf4f85f7e07cb10110eae9b7) I can get the network on the domu's to work for a limited time.  But, under heavy load the domu network just stops working!

For now I'm using the 2.6.18.hg (2.6.18.8) kernel which has no network problems, until I can try another motherboard (different NIC's)  to see if it is hardware/driver problem...

Czar
 
On Tue, 04 May 2010 18:03 +0200, "Benjamin Schweikert" <b.schweikert@xxxxxxxxxxxxxx> wrote:
Problem still not solved, or any idea whats wrong.

here are some msgs:
device vif1.0 entered promiscuous mode
  alloc irq_desc for 1246 on node 0
  alloc kstat_irqs on node 0
brI: port 2(vif1.0) entering learning state
device vif1.1 entered promiscuous mode
brE: port 2(vif1.1) entering learning state
physdev match: using --physdev-out in the OUTPUT, FORWARD and POSTROUTING chains for non-bridged traffic is not supported anymore.
physdev match: using --physdev-out in the OUTPUT, FORWARD and POSTROUTING chains for non-bridged traffic is not supported anymore.
physdev match: using --physdev-out in the OUTPUT, FORWARD and POSTROUTING chains for non-bridged traffic is not supported anymore.
physdev match: using --physdev-out in the OUTPUT, FORWARD and POSTROUTING chains for non-bridged traffic is not supported anymore.
brI: port 2(vif1.0) entering forwarding state
brE: port 2(vif1.1) entering forwarding state
blkback: ring-ref 8, event-channel 10, protocol 2 (x86_32-abi)
  alloc irq_desc for 1245 on node 0
  alloc kstat_irqs on node 0
blkback: ring-ref 9, event-channel 11, protocol 2 (x86_32-abi)
  alloc irq_desc for 1244 on node 0
  alloc kstat_irqs on node 0
blkback: ring-ref 10, event-channel 12, protocol 2 (x86_32-abi)
  alloc irq_desc for 1243 on node 0
  alloc kstat_irqs on node 0
blkback: ring-ref 11, event-channel 13, protocol 2 (x86_32-abi)
  alloc irq_desc for 1242 on node 0
  alloc kstat_irqs on node 0
  alloc irq_desc for 1241 on node 0
  alloc kstat_irqs on node 0
  alloc irq_desc for 1240 on node 0
  alloc kstat_irqs on node 0
nfsd: last server has exited, flushing export cache
NFSD: Using /var/lib/nfs/v4recovery as the NFSv4 state recovery directory
NFSD: starting 90-second grace period
brE: port 2(vif1.1) entering disabled state
brE: port 2(vif1.1) entering disabled state
brI: port 2(vif1.0) entering disabled state
brI: port 2(vif1.0) entering disabled state
physdev match: using --physdev-out in the OUTPUT, FORWARD and POSTROUTING chains for non-bridged traffic is not supported anymore.
physdev match: using --physdev-out in the OUTPUT, FORWARD and POSTROUTING chains for non-bridged traffic is not supported anymore.
physdev match: using --physdev-out in the OUTPUT, FORWARD and POSTROUTING chains for non-bridged traffic is not supported anymore.
physdev match: using --physdev-out in the OUTPUT, FORWARD and POSTROUTING chains for non-bridged traffic is not supported anymore.
device vif2.0 entered promiscuous mode
brI: port 2(vif2.0) entering learning state
physdev match: using --physdev-out in the OUTPUT, FORWARD and POSTROUTING chains for non-bridged traffic is not supported anymore.
device vif2.1 entered promiscuous mode
brE: port 2(vif2.1) entering learning state
physdev match: using --physdev-out in the OUTPUT, FORWARD and POSTROUTING chains for non-bridged traffic is not supported anymore.
physdev match: using --physdev-out in the OUTPUT, FORWARD and POSTROUTING chains for non-bridged traffic is not supported anymore.
physdev match: using --physdev-out in the OUTPUT, FORWARD and POSTROUTING chains for non-bridged traffic is not supported anymore.
brI: port 2(vif2.0) entering forwarding state
brE: port 2(vif2.1) entering forwarding state
blkback: ring-ref 8, event-channel 10, protocol 2 (x86_32-abi)
blkback: ring-ref 9, event-channel 11, protocol 2 (x86_32-abi)
blkback: ring-ref 10, event-channel 12, protocol 2 (x86_32-abi)
blkback: ring-ref 11, event-channel 13, protocol 2 (x86_32-abi)



and when i start xend i get the following:

blktapctrl_linux.c:92: couldnt find device number for blktap0
blktapctrl.c:859:couldnt open blktap interface
blktapctrl.c:922:unable to start blktapctrl


any ideas?

Ben

-------- Original-Nachricht --------
Betreff: Strange network problem
Datum: Mon, 03 May 2010 08:22:02 +0200
Von: Benjamin Schweikert <b.schweikert@xxxxxxxxxxxxxx>
An: xen-users@xxxxxxxxxxxxxxxxxxx


Hi,
I am using the following setup:
Dom0:
Gentoo 64 Bit with a 2.6.33.2 Pv ops kernel
cpu: amd athlon II 605e
bridged network.

As domu I want to use ipfire which has a pv xen kernel based on the 
opensuse xen patches (like andy lyon does).
If I use an old dom0 xen kernel with xen 3.4.x everything is working. 
But if I use a new dom0 kernel (different xen versions doesn't change 
anything)
networking is not really working. From domu and dom0 ping does work, but 
i am unable to sync the portage tree of the dom0. From my lan i am unable
to ping. But I am able to surf from the lan (for example icq, 
jabber,.imap,.. doesnt connect, too). If i start the domu as a hvm 
domain everything is working. so i think, the bridges are working. Any 
ideas whats going wrong?

Ben
_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-user


-- 
  Thomas J. Czarnecki
  czar_tj@xxxxxxxxxxx
-- 
http://www.fastmail.fm - Choose from over 50 domains or use your own
_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users

 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.