similar to: can hotplug vcpus to running Windows 10 guest, but not unplug

----- On Feb 15, 2020, at 12:47 AM, Marc Roos M.Roos@f1-outsourcing.eu wrote: > Would you mind sharing your xml? I have strange high host load on idle > windows guest/domain <domain type='kvm'> <name>pathway</name> <uuid>8235e5ae-0756-4286-5407-9fa02d372046</uuid> <description>Pathway Studio Dietrich</description> <memory

Hi guys, despite reading hours and hours in the internet i'm still struggling with "memory", "currentmemory" and "maxMemory". Maybe you can help me to sort it out. My idea is that a guest has an initial value of memory (which "memory" seems to be) when booting. We have some Windows 10 guests which calculate some stuff and i would like to increase

On Fri, Feb 14, 2020 at 16:05:00 +0100, Lentes, Bernd wrote: > Hi, > > i'm playing a bit around with vcpus. > My guest is Windows 10 1903. > This is the excerpt from the config: > ... > <vcpu placement='static' current='2'>4</vcpu> > <vcpus> > <vcpu id='0' enabled='yes' hotpluggable='no'/> >

----- On Feb 14, 2020, at 4:13 PM, Peter Krempa pkrempa@redhat.com wrote: > > Sounds like qemu doesn't support unplug of vcpus. Which version of qemu > do you use? ha-idg-2:~ # rpm -qa|grep qemu qemu-seabios-1.11.0-5.18.1.noarch qemu-ovmf-x86_64-2017+git1510945757.b2662641d5-3.16.1.noarch qemu-block-curl-2.11.2-5.18.1.x86_64 qemu-sgabios-8-5.18.1.noarch

Hi, i wanted to benchmark a windows guest, compare standard driver and virtio driver. I installed the domain first with an IDE disk. I followed https://pve.proxmox.com/wiki/Paravirtualized_Block_Drivers_for_Windows to install the virtIO driver. In the device manager my VirtIO disk is recognized as a VirtIO SCSI disk from RedHat which seems ok for me. But the driver is, following the device

On Tue, Feb 11, 2020 at 18:48:43 +0100, Lentes, Bernd wrote: > Hi guys, Hi Bernd, > > despite reading hours and hours in the internet i'm still struggling with > "memory", "currentmemory" and "maxMemory". to briefly summarize what those three knobs do: 1) memory - this is the initial memory size for a VM. qemu grants this amount of memory to the

Hi, i'm cuurently a bit confused if a guest does have a valid snapshot or not. This is the xml: ... <disk type='file' device='disk'> <driver name='qemu' type='qcow2'/> <source file='/var/lib/libvirt/images/crispor_1604/crispor_1604.sn'/> <backingStore type='file' index='1'> <format

Hi, i have a Linux domain (Ubuntu 14.04) where we like to be able to change the amount of usable memory. We have a balloon device and statistics are switched in 5 sec. rhythm. The domain shows very quickly changes (following top) when we change the memory size with setmem, but the host does not show changes in the use of memory for the respective domain. Is that expected behavior ? Host is SLES

----- On Feb 7, 2020, at 3:43 PM, Peter Krempa pkrempa@redhat.com wrote: > On Fri, Feb 07, 2020 at 15:25:22 +0100, Lentes, Bernd wrote: ... > > Libvirt is probably lacking the metadata for the snapshot. That is not a > problem though, because since libvirt doesn't support deletion of > external snapshots anyways currently you'd need to use the below > approach anyways.

Hello. I'm struggeling a little bit with the setvcpus command. I'm trying now to get this running for a linux VM. Here is what i have: <vcpu placement='auto' current='2'>4</vcpu> Now the VM is starting with 2 CPUs. When i try to increase the number of CPUs i get: setvcpus dev --guest 4 error: invalid argument: requested vcpu count is greater than the count

----- On Jun 13, 2019, at 9:56 AM, Peter Krempa pkrempa@redhat.com wrote: > > Thanks for comming back to me with the information. > > Unfortunately this is not a full debug log but I can try to tell you > what I see here: I configured libvirtd that way: ha-idg-1:~ # grep -Ev '^$|#' /etc/libvirt/libvirtd.conf log_level = 1 log_filters="1:qemu 3:remote 4:event

Hi, i have a two-node cluster running on SLES 12 with pacemaker. The cluster refused to start the domains on one node. So i took some of the domains out of the cluster and tried to start it manually. This is what happened: virsh # define /mnt/share/vm_documents-oo.xml error: Failed to define domain from /mnt/share/vm_documents-oo.xml error: cannot open /dev/null: Operation not permitted Same

Am 09.08.13 15:09, schrieb Eric Blake: > On 08/09/2013 06:26 AM, Thomas Stein wrote: >> Hello. >> >> Does someone know in what state setvcpu command is? >> >> I get: >> >> virsh # setvcpus gentookvm 4 >> error: internal error: cannot change vcpu count of this domain >> >> and >> >> error : qemuDomainHotplugVcpus:4087 :

On 4/13/2015 1:18 PM, Kashyap Chamarthy wrote: > On Mon, Apr 13, 2015 at 12:14:12PM +0530, Jatin Davey wrote: >> Hi All >> >> I am new to KVM world. I have been using VMware for about a year. >> >> In VMware world we can reserve cpu resources for a VM. Say for instance if >> we need 4 vCPUs with a total of 8000 GHz then we can mention this >>

Hi, i knwo that virsh has its own history, but are somewhere the respective timestamps logged ? Bernd -- Bernd Lentes Systemadministration Institut für Entwicklungsgenetik Gebäude 35.34 - Raum 208 HelmholtzZentrum münchen bernd.lentes@helmholtz-muenchen.de phone: +49 89 3187 1241 phone: +49 89 3187 3827 fax: +49 89 3187 2294 http://www.helmholtz-muenchen.de/idg Perfekt ist wer

Hi ML, i'm thinking about using balloon memory for our domains. We have about 15 domains running concurrently, and i think it might be nice if a domain requires more RAM it grabs it, and if it don't need it anymore, it releases it. But i have no experience with it. So i have some questions: - is live migration possible with balloon ? - is it stable ? - the domain needs an appropriate

Hello. Just a quick question. Is "setvcpus $windowsVM --guest 4" supposed to work? I have qemu-guest-agent (1.6.something) running inside a windows7 vm but all i get is: virsh # setvcpus windows7 --guest 4 error: invalid argument: requested vcpu count is greater than the count of enabled vcpus in the domain: 4 > 1 The xml file looks like this: <vcpu placement='static'

Am 05.11.14 11:48, schrieb Martin Kletzander: > On Wed, Nov 05, 2014 at 11:35:37AM +0100, Thomas Stein wrote: >>Hello. >> >>I'm struggeling a little bit with the setvcpus command. I'm trying now >>to get this running for a linux VM. Here is what i have: >> >> <vcpu placement='auto' current='2'>4</vcpu> >> >>Now

----- On Jun 5, 2019, at 4:49 PM, Peter Krempa pkrempa@redhat.com wrote: > On Wed, Jun 05, 2019 at 13:33:49 +0200, Lentes, Bernd wrote: >> Hi Peter, >> >> thanks for your help. >> >> ----- On Jun 5, 2019, at 9:27 AM, Peter Krempa pkrempa@redhat.com wrote: > > [...] > >> >> > >> > So that's interresting. Usually assertion

? 2020/6/5 ??6:05, David Hildenbrand ??: >> I guess I know what's happening here. In case we only have DMA memory >> when booting, we don't reserve swiotlb buffers. Once we hotplug memory >> and online ZONE_NORMAL, we don't have any swiotlb DMA bounce buffers to >> map such PFNs (total 0 (slots), used 0 (slots)). >> >> Can you try with