similar to: Dom0 scheduling on SMP systems

Hi, I saw a nice mail about bridging in Xen at http://article.gmane.org/gmane.comp.emulators.xen.user/9538. Can I ask you a quick question? I''ve set up bridge-networking for Xen 3.0and networking was being done properly for Dom0 and DomUs. However, I noticed that Dom0 receives a lot of network interrupts or network packets even when they were not actually meant for either Dom0 or DomU.

Hi, I had a quick question. I''ve few machines available in a cluster. Now, if I want to install Xen on them, do I individually install in all of them (one by one); or can I have on single image available over a network/SAN and can I boot Xen from there? Are there any other smarter ways to do that? Thanks, Amitayu _______________________________________________ Xen-users mailing list

The Patch I made for SMP dom0 is also applicable to SMP domU. The issue is same for dom0 & domu as they both are compiled from the same code. The dom0 fix I sent earlier also fixes the domU SMP issue. Thanks & Regards, Nitin ------------------------------------------------------------------------ ----------- Open Source Technology Center, Intel Corp >-----Original Message-----

While running xen-unstable (2005-01-13 nightly) with dom0 and dom1 with CONFIG_SMP=y I got 271k of output... sshd was running in dom0 and I remotely initiated an scp from dom0 to another remote host (not dom1) Ill only post the first and last few "scheduling while atomic" entries, let me know if someone wants the whole trace. scheduling while atomic [schedule+1682/1696]

Keir, Ian, With PCI mmconfig option on, and with the PCI express enabled BIOS, the dom0 kernel reads the PCI config from fix-mapped PCI mmconfig space. The PCI mmconfig space is of 256MB size, and it''s access is implemented differently on i386 & x86_64. On x86_64 the whole 256MB is mapped in the Kernel virtual address space. On i386 it will consume too much of the kernels

Is this PAE or non-PAE? Please can you try forceing emulation mode by toggling the "#if 0" in arch/x86/mm.c ptwr_do_page_fault The other thing to try is modifying set_pte_pfn_ma to call xen_l1_update rather than set_pte. You could try set_pte_at too. This will help narrow down the issue. Thanks, Ian > Keir, Ian, > With PCI mmconfig option on, and with the PCI express >

Hi listmembers, not a really urgent question, but i''m just curious about it: Is it advised to use an irqbalanced on dom0 when running domU''s pinned to particular cores? as an example, i''ve got a dual quadcore xen system running with domU pinned to cores 1-3 (CPU#0) domU pinned to cores 4-5 (CPU#1) domU pinned to cores 6-7 (CPU#1) so dom0 should have 100% time on

Hopefully I summed up the jist of the problem in the subject line. ;) I have a GFS cluster with ten Xen 3.0 dom0s sharing an iSCSI LUN. There are on average 8 domUs running on each Xen server. The dom0 on each server is hard-coded to 2 CPUs and 2GB of RAM with no ballooning, and has 2GB of partition-based swap. When creating a new domU on any of the Xen servers, just after the completion of

Hi, How can we know about the minimum time-quantum that a Xen scheduler allocates for a DomU? How big or less is that compared to the length of one jiffy? Please note that I''m not asking for a slice value within a period. In order to schedule a DomU for a total time of slice value within a period, what is the minimum time that a DomU would be scheduled by Hypervisor each time within a

Hi, I''m curious about how the SMP support in Xen works. Is it possible to have a SMP aware 2 vcpu domain running on both physical processors at the same time? Can Xen schedule two domains at the same time, one running on each processor? How about on a hyperthreaded kernel? If one domain has a run away process (eg. fork loop) or bad device driver, will it affect the running of other

["Standard" CentOS 5.5 Xen Virtualization installation.] I'm setting up a new box on AMD Opteron platform. I'm doing some preliminary baselines (no domUs running) with an inline Watt/VA meter. It appears to enable full power management Xen dom0 must have all cpus by not limiting in kernel statement nor in xendconfig.sxp. In previous installations I cut them back to 1 or 2.

Hello, I''m using a server with 10Gbps network interfaces and 4 CPUs, running several domUs. The problem is that in this setup, with high network load, dom0 turns out to be the bottleneck, using only a single CPU which is saturated and 100%. So the network speed is bounded to much less than 10Gbps. How could I make dom0 use more CPUs in parallel ? I checked my VCPU setup; i have 4 VCPUs

Hi, I have a question about the memory layout of the Xen hypervisor, dom0 and domU. *The goal I want to achieve is: I want to show which RAM area is used by hypervisor, which RAM area is used by dom0, which RAM area is used by each domU (PV guest domain). I hope to know the above information when xen is booted as well as when xen is running some domUs. I already know that we can dedicate a fix

This may not be the best place to ask, but I was prompted by a question about guest cores on KVM. We currently use VMWare Server (v1.0) on CentOS5. It supports up to two virtual CPUs, but not very well, as I understand it. VMWare Server 2.0 might do better at supporting the same maximum of 2 CPUs, but if my research is correct, they both use what is called "strict co-scheduling".

Hi. My domU date is about 75 seconds (and climbing?) behind domU. Tue Feb 21 23:11:31 UTC 2006 - dom0 Tue Feb 21 23:10:16 UTC 2006 - domu Tue Feb 21 18:10:17 EST 2006 - domu domUs'' /proc/sys/xen/independent_wallclock was 0 Set that to 1: echo 1 > /proc/sys/xen/independent_wallclock Still out by about 75 seconds: Tue Feb 21 23:14:32 UTC 2006 - dom0 Tue Feb 21 23:13:17 UTC 2006 -

Situation: Running a simple UDP client/server program where the client on one domU on one computer sends echo packets to another domU on another computer, server sends echoes back. They do this on a specified port (will use any port between 5000-6000). This program works on non-Xen machines in various environments, Linux and Solaris. Program just hangs on the domUs. I believe I need help with

I wanted to post this in case there is a real bug here. The system this is on is running Debian Etch with the Xen packages from Debian Unstable (currently 3.0.2+hg9697-1). I am running all packaged software so this is probably slightly out of date. This is on an HP DL320 G4 with a Pentium D 930 processor. I tried unloading and reloading the qla2xxx module, but rmmod reported it in use, however

Hi list, We, an OSS advocacy group, setup a Xen 3.1 machine composed of : . a 64 bits dom0 running Debian stable amd64 . 2 hvm domUs running OpenBSD amd64 . 2 hvm domUs running NetBSD i386 This machine is to be hosted and reachable from the Internet, but it will only have one public IP. Naturally, our first tought was to port-forward using iptables / netfilter. We didn''t really

> I wanted to post this in case there is a real bug here. > > The system this is on is running Debian Etch with the Xen packages from > Debian Unstable (currently 3.0.2+hg9697-1). I am running all packaged > software so this is probably slightly out of date. It would be interesting to see if this can be repro''ed on a tree built from a recent xen-unstable.hg.

I am currently running a Redhat Rel5 Dom0 with a solaris 11-66 xen domU as well as a Redhat REL4 domU. I am experiancing problems similary to those described the the Aug 2006 Release notes regarding ... Solaris domUs on non-Solaris dom0s 32-bit Solaris domUs should run on any 3.0.2-3 or later dom0. The majority of our testing has been on CentOS dom0s. 64-bit solaris dom0 and domU require updates