- Users - libvirt List Archives

changing memory size with virsh setmem - results only visible in domain, not on host
by Lentes, Bernd 07 Mar '20

07 Mar '20

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 12 SP4, libvirt is 4.0.0-8.15.2 Bernd -- Bernd Lentes Systemadministration Institute for Metabolism and Cell Death (MCD) Building 35.34 - Raum 208 HelmholtzZentrum münchen bernd.lentes(a)helmholtz-muenchen.de phone: +49 89 3187 1241 phone: +49 89 3187 3827 fax: +49 89 3187 2294 http://www.helmholtz-muenchen.de/mcd Perfekt ist wer keine Fehler macht Also sind Tote perfekt Helmholtz Zentrum MÃ¼nchen Helmholtz Zentrum Muenchen Deutsches Forschungszentrum fuer Gesundheit und Umwelt (GmbH) Ingolstaedter Landstr. 1 85764 Neuherberg www.helmholtz-muenchen.de Aufsichtsratsvorsitzende: MinDir.in Prof. Dr. Veronika von Messling Geschaeftsfuehrung: Prof. Dr. med. Dr. h.c. Matthias Tschoep, Kerstin Guenther Registergericht: Amtsgericht Muenchen HRB 6466 USt-IdNr: DE 129521671

2 1

KVM on CentOS 7 - sluggish performance with Win10 Guest
by Benjammin2068 06 Mar '20

06 Mar '20

Hey all, I've been running a Win10 Pro (64bit) guest for some time now.... and originally it was pretty snappy and I had no complaints. Somewhere in about the last 3-6 months I've been noticing sluggish performance -- but not overall. The system is mostly remote so I do a lot of my admin via RDP or the local console with the KVM Manager. I recently did a complete fresh install of Win10 Pro (1909. Previous install was 1607) and both show the same problem. The ONLY real thing running on this VM is Milestone's surveillance system (standalone host). Where I see the "sluggish" is: Connecting to milestone mobile server (either through web browser or client). for the small number of cameras, the connection spool up takes like 15 seconds which is noticeably longer than other systems I have set up with more load/tasks in terms of cameras. This used to be speedy and fine. The other place I see sluggish operation is on the desktop doing some typical tasks and seeing "not responding" in the window of the task that goes away - but should it ever appear to begin with? The system is a Supermicro motherboard with Dual Intel(R) Xeon(R) CPU E5-2640 v4 @ 2.40GHz w/64GB The Guest is provided with 20 cores in a config of 1 CPU, 10 cores - 2 threads per core. There are dedicated NICs (dual NIC card) attached to the system via macvtap (and are not used by CentOS in any way except to pass through to this Guest) I'm using QCOW2 for the base OS. The storage volume for video are raw devices. Once a process is running, performance seems fine... it seems to be starting processes or something. I'm not sure -- and I'm not as intimate with KVM/libvirt as I'd like to be yet. So I'm wondering if there's any performance tweaks I should know about. I've read about using .img vs .qcow2 already. I'm using virt-io drivers to some extent.... but is there anything else I'm missing? Thanks in advance, -Ben

1 0

upgrading from fedora29 to fedora 31
by Thierry Parmentelat 05 Mar '20

05 Mar '20

Hi there I am using libvirt+lxc on fedora I just upgraded a couple boxes from f29 to f31 i am running into rather puzzling glitches on one of the boxes, and am having a hard time trying to figure what exactly is going wrong symptom being that on the faulty box, my 3 containers startup correctly, but all of them stop after some random time shortly after (approx. in the 5-10 minutes) in some cases I even lose one of the containers (its name no longer shows up in `virsh list —all` and I need to reboot the box to reinstate it is there any known issue with libvirt/lxc on fedora 31 ? ***** here’s the little that I’ve been able to figure out so far by myself * I have turned on debugging of libvirtd as per the docs log_filters="3:remote 4:event 3:util.json 3:rpc 1:*" log_outputs="1:file:/var/log/libvirt/libvirtd.log” and am digging in that log file, as well as the system journal, and the containers journals, to try and find some correlation; * I have seen an error message saying this > virNetSocketReadWire:1850 : End of file while reading data: Input/output error and that could be related to the problem; ***** this is really annoying and any hint would be hugely appreciated — thanks in advance

2 1

Virtio-disk with driver from Microsoft from 2006 ?
by Lentes, Bernd 05 Mar '20

05 Mar '20

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 manager, a driver from Microsft from 21/06/2006, version 10.0.18362.1. That's strange in my eyes. This driver is outdated and not from RedHat or Fedora, what i expected. Also updating the driver and pointing to the CD, even the respective folder, didn't work. It says the driver is the most recent. I used the virtIO-ISO 0.1.173, Windows 10 64bit Edition 1903. Bernd -- Bernd Lentes Systemadministration Institute for Metabolism and Cell Death (MCD) Building 35.34 - Raum 208 HelmholtzZentrum münchen bernd.lentes(a)helmholtz-muenchen.de phone: +49 89 3187 1241 phone: +49 89 3187 3827 fax: +49 89 3187 2294 http://www.helmholtz-muenchen.de/mcd Perfekt ist wer keine Fehler macht Also sind Tote perfekt Helmholtz Zentrum MÃ¼nchen Helmholtz Zentrum Muenchen Deutsches Forschungszentrum fuer Gesundheit und Umwelt (GmbH) Ingolstaedter Landstr. 1 85764 Neuherberg www.helmholtz-muenchen.de Aufsichtsratsvorsitzende: MinDir.in Prof. Dr. Veronika von Messling Geschaeftsfuehrung: Prof. Dr. med. Dr. h.c. Matthias Tschoep, Kerstin Guenther Registergericht: Amtsgericht Muenchen HRB 6466 USt-IdNr: DE 129521671

1 1

Guests automatically stuck on pause
by JT Edwards 05 Mar '20

05 Mar '20

I have forced off the guest and started them again only to get this error Error unpausing domain: internal error: unable to execute QEMU command 'cont': Resetting the Virtual Machine is required Traceback (most recent call last): File "/usr/share/virt-manager/virtManager/asyncjob.py", line 75, in cb_wrapper callback(asyncjob, *args, **kwargs) File "/usr/share/virt-manager/virtManager/asyncjob.py", line 111, in tmpcb callback(*args, **kwargs) File "/usr/share/virt-manager/virtManager/object/libvirtobject.py", line 66, in newfn ret = fn(self, *args, **kwargs) File "/usr/share/virt-manager/virtManager/object/domain.py", line 1311, in resume self._backend.resume() File "/usr/lib/python3/dist-packages/libvirt.py", line 2174, in resume if ret == -1: raise libvirtError ('virDomainResume() failed', dom=self) libvirt.libvirtError: internal error: unable to execute QEMU command 'cont': Resetting the Virtual Machine is required Ubuntu: 5.4.0-14-generic #17-Ubuntu SMP Thu Feb 6 22:47:59 UTC 2020 x86_64 x86_64 x86_64 GNU/Linux libvirt: libvirtd (libvirt) 6.0.0 Thanks in advance JT

1 0

kvm presenting wrong CPU Topology for cache
by Satish Patel 01 Mar '20

01 Mar '20

Folks, I am having major performance issue with my Erlang application running on openstack KVM hypervisor and after so many test i found something wrong with my KVM guest CPU Topology This is KVM host - http://paste.openstack.org/show/790120/ This is KVM guest - http://paste.openstack.org/show/790121/ If you carefully observe output of both host and guest you can see guest machine threads has own cache that is very strange L2 L#0 (4096KB) + Core L#0 L1d L#0 (32KB) + L1i L#0 (32KB) + PU L#0 (P#0) L1d L#1 (32KB) + L1i L#1 (32KB) + PU L#1 (P#1) I believe because of that my erlang doesn't understand topology and going crazy.. I have Ali Cloud and AWS and when i compare with them they are showing correct CPU Topology the way physical machine showing, something is wrong with my KVM look like. I am running qemu-kvm-2.12 on centos 7.6 and i have tune my KVM at my best level, like CPU vining, NUMA and cpu host-passthrough. Thanks in advance for your help.

1 1

libvirtError: Timed out during operation: cannot acquire state change lock (held by monitor=remoteDispatchDomainMigratePerform3Params)
by Blake Anderson 28 Feb '20

28 Feb '20

Hi everyone, I have a question that you may be able to help me with. I had a live block migration of a qemu-kvm guest fail (initiated via nova), in which the guest remained running on the source, but if I try to re-initiate the live migration it returns libvirtError: Timed out during operation: cannot acquire state change lock (held by monitor=remoteDispatchDomainMigratePerform3Params). Looking at blockjob --info I see there is a block copy that has been stuck at 23 % for a few hours now. Is it safe to issue a —abort to the blockjob without impacting the vm and will that resolve the lock? Source: libvirt-4.5.0-23.el7_7.5.x86_64 qemu-kvm-ev-2.10.0-21.el7_5.7.1.x86_64 Destination: libvirt-4.5.0-23.el7_7.5.x86_64 qemu-kvm-ev-2.10.0-21.el7_5.7.1.x86_64 Thanks, Blake

2 2

Windows guest stalls during reboot
by Benjammin2068 27 Feb '20

27 Feb '20

Hey all, Quick question. I have a (RHEL/CentOS v7) Windows 10 Pro (64bit) guest that shutdown just fine (i.e. KVM shows the host as shut down), but when Windows is set to reboot, KVM Manager shows the guest screen as shut down, but KVM Manager shows the guest as still running and the system never reboots. If I do 'virsh list' -- there's no guest running. If I do 'virch start MyWin10Guest' -- that starts the guest back up and things are like they were before. Am I missing a new setting someplace? I can't seem to type in the right thing in Google to help me figure this one out. Thanks a bunch, -Ben

1 0

pci device vga added not showing
by Barry Cisna 23 Feb '20

23 Feb '20

Hello All, Trying to get virgl working from linux host to windows guest in virt-manager i have added the host graphics card Matrox G200ew to the hardware settings as a pci device for Windows 10 guest. When i go to spice server settings, and select Opengl checked there are no available pci devices listed in the drop down box select. Am trying to get some 3d working on Windows guest and seems from reading this is almost impossible short of doing gpu passthrough with a second GPU in the host. Am probably missing something. Thank You.

1 0

build problem with libxl
by Christoph 21 Feb '20

21 Feb '20

Hi all since 5.9.0 configure doesnt find the libxl libs/headers... It works With 5.8.0, with 5.9.0 not... not even if I set --with-libxl What can be the problem? -- ------ Greetz

3 7