similar to: About xen bridge

Hi Is there anyone who has successfully resolved the low throughput problem in Guest communication. I am using Xen3.4.1 with PV os kernel 2.6.30-rc6-tip On fedora11. While running netperf benchmark on guests throughput results are very low DomU to DomU    0.29 Mbps Why is it so? What would be the problem. Is there any issue with fc11 platform? Regards, Fasiha Ashraf Add whatever you love

This patch adds an ACM hook into the network scripts (/etc/xen/scripts). It adds iptables rules that enforce mandatory access control on network packets exchanged between virtual interfaces. If ACM is active, this patch sets the default FORWARD policy in Dom0 to DROP and adds iptables ACCEPT rules between vifs that belong to domains that are permitted to share (determined by using the

> > When you run xend it runs network-bridge which creates the bridge, > > renames eth0->peth0 and veth0->eth0 and brings the new eth0 back up. > > kewl, thx Ian. i see network-route doesnt do this. i think > thats a good thing. do you forsee that changing? I expect that in most routed configurations people use their own network script rather than using the

Hi there, And thank you for your info. Most useful. The problem is that the time is WRONG on one of your machines. Let me explain how. > > C:\Documents and Settings\rr> echo. |time > > The current time is: 16:14:53.89 > Enter the new time: Your PC thinks it is 16:14 this is fine. > > ok TIMEZONE = (GMT +5.00) iSLAMABAD KARACHI, TASKHENT So your PC is in a

This is my install profile F11.install name="FC11-G1S2" memory=500 disk = [''phy:/dev/sda3,0,w'' ] vif = [ ''bridge=eth0'' ] vfb = [ ''type=vnc,vncunused=1''] kernel = "/etc/xen/vm/vmlinuz.1" ramdisk = "/etc/xen/vm/initrd.img.1" vcpus=1 on_reboot = ''restart'' on_crash = ''restart''

Hi, Boris! i have installed and run virt-manager successfully. After i boot into my xen paravirt kernel I got, [root@Server1-FC11 init.d]# xm info host                   : Server1-FC11 release                : 2.6.30-rc6-tip version                : #1 SMP Thu Oct 22 12:48:59 PKST 2009 machine                : x86_64 nr_cpus                : 8 nr_nodes               : 1 cores_per_socket       :

Objet : Xend start and after lose network Hi everybody : My configuration : Dapper 6.06 LTS with all updates Xen 3.0.2.2 My problem : After making ''xend start'', I lose network connections between my server and local network or internet. Nota : domU cas normally access to network I read about a similar problem in the bugtracker (N. 62) but it seems that this is normally

Hi! Maybe, I found a bug of xen. My system: - domain0 - gentoo - xen-devel-3.0, kernel 2.6.12.5-r1 - domain1 - debian - kernel 2.6.12.5-r1 (2 interface: vif1.1 = eth0 (0.0.0.0), vif1.2 = eth1 (10.0.1.1 + gw 10.0.1.2)) Bridge: xen-br0 (config as 10.0.1.2) include (vif1.2, vif0.0) xen-br1 (config as 0.0.0.0) include (vif1.1, peth0) On server domain1 I use a pppoe server. If a send a pppoe request

I have set-up xen on my Intel quad core server. Now running different experiment to measure network throughput in virtualized environment. these are some of the results: Netperf-4.5 results for inter-domain communication. Sr.No.              Client               Server               Time(sec)        Throughput(Mbps)    1                   Guest-1              Dom0                   

Hi, I am using Xen 4.1.1 with Linux 3.0.1 kernel. Previously (Xen 3.4.4 and linux kernel 2.6.18) Xen by default created veth0 <--> vif0.0 pair in dom0. With Xen 4.1.1 these pairs are missing in dom0. Using command ''ip link add type veth'' I can create veth0 and veth1 but vif0.0 is still missing in dom0. (When new domU created I can see vif1.x created) Could you please help

Hi, I am using Xen 4.1.1 with Linux 3.0.1 kernel. Previously (Xen 3.4.4 and linux kernel 2.6.18) Xen by default created veth0 <--> vif0.0 pair in dom0. With Xen 4.1.1 these pairs are missing in dom0. Using command ''ip link add type veth'' I can create veth0 and veth1 but vif0.0 is still missing in dom0. (When new domU created I can see vif1.x created) Could you please help

Hey folks! I want to add a CentOS 5.2 domU to an existing server (Fedora 8) which was setup by another company. The Dom0 does not provide any other services but a BIND-DHCP-server (they deleted dnsmasq and installed bind). Don''t ask my why they did that. My new DomU is connected to eth0, which is a bridge. BIND also listens on eth0 to serve other machines. I start the install of centos

Hello everybody, I have an odd problem with iptables using a Xen bridge setup. I don''t know if it would be better to post to netfilter Mailing-List. But I hope someone here know how to solve it. If it''s OT here, please let me know. I''ll try to do a little bit ASCII-Graphics to explain the topo better: _________ ________

So, there exists registers which are reserved denoted by the Machine Register Info :: getReserved field. I am wondering if these registers are reserved or preserved? I have come across a case where a reserved register was used. Namely it's in "llvm\test\CodeGen\X86\2009-04-27-CoalescerAssert.ll" where the reserved register ESP (54) in block "bb98.fragment" on line 854. Is

Hi, Can someone interpret how HVM guest communicate with domain0, as i know, H VM can not use event channel directly, is the platform-pci offer a way to solve this? Please try to explain concretely, for example : 1.how does HVM send a event to domain0, how does domain0 catch this event, and assoicate it with a handler 2.the opposite direction, how does domain0 send event to HVM guest? Does it

(if this post gets line-feed-mangled please read http://www.dl.reneschmidt.de/shorewallxenpost.txt - that''s an unmangled version, thank you) Hello, first I would like to thank the Mr. Eastep and contributors for this great piece of software and superb documentation. I have a SOHO server (Debian testing) that I''m using for several purposes so I''ve set up a Xen

Hi everyone! I have the following setup: domain0 |-domU1 |-domU2 |-domU3 I am running this with libvirt on a XEN server. Right now the server is connected to our office network, but later it will be in a hosted environment. I need domU1 and domU3 to have a public IP, and all of the domUs and the dom0 should have an internal IP address to communicate with each other. So this is what I would like

Hi, If one sets xend to use network-bridge and there are no bridges already present then it seems that xend will clone eth0 to peth0, create xenbr0 and add peth0 and vif0.0 as ports on that bridge. If on the other hand xenbr0 is created in /etc/network/interfaces then xend will not do any of that peth0 stuff, nor will it add vif0.0 to xenbr0, yet (barring some changes in iptables rules) things

Dear, "Xenoppix-x86-64(Xen3.0.2+KNOPPIX-x86_64-5.0.1) CD size" is released. http://unit.aist.go.jp/itri/knoppix/xen/index-en.html Xenoppix-x86-64 exploits latest CPU (Intel Core2 Duo and AMD Athlon64). * Xenoppix-x86-64 boots both X86 and X86_64 linux kernel on HVM of Xen. * Xenoppix-x86-64 looks like KNOPPIX but all packages are replaced for X86_64. * Xenoppix-x86-64 also

Hi, I''ve set up identical (what I think) machines using SLES 10 x86_64, but a new one of those is saying frequently in /var/log/messages: kernel: bond0: received packet with own address as source address A short summary of ifconfig looks like this: bond0 Link encap:Ethernet HWaddr 00:14:4F:2A:12:9C inet addr:132.199.176.78 Bcast:132.199.255.255 Mask:255.255.0.0 eth0