- Users - libvirt List Archives

no/empty NUMA cells on domain XML
by Polina Agranat 01 Jul '20

01 Jul '20

Hi , I'm looking for a possibility to simulate a VM (to be used as a host) reporting no or empty NUMA <cells> in 'virsh capabilities'. 'virsh capabilities' usually reports single NUMA in the VMs, like <numa> <cell id='0' cpus='0-63' memory='16777216' unit='KiB'/> </numa> I appreciate if someone could advise Thank you Polina

2 2

image of OS X how to boot
by Jerry Geis 30 Jun '20

30 Jun '20

I have an image of OS X that I made with from the physical disk using dd. When I boot with libvirt-manager and environment I get an error. no bootable option or device was found. Using CentOS 8 host - I tried fdisk image.img and I no longer see any options there to toggle bootable partition option. How can I boot OSX ? Thanks, Jerry

2 1

plug pre-created tap devices to libvirt guests
by Miguel Duarte de Mora Barroso 30 Jun '20

30 Jun '20

Hi all, I'm aware that it is possible to plug pre-created macvtap devices to libvirt guests - tracked in RFE [0]. My interpretation of the wording in [1] and [2] is that it is also possible to plug pre-created tap devices into libvirt guests - that would be a requirement to allow kubevirt to run with less capabilities in the pods that encapsulate the VMs. I took a look at the libvirt code ([3] & [4]), and, from my limited understanding, I got the impression that plugging existing interfaces via `managed='no' ` is only possible for macvtap interfaces. Would you be able to shed some light into this ? Is it possible on libvirt-5.6.0 to plug pre-created tap devices to libvirt guests ? [0] - https://bugzilla.redhat.com/show_bug.cgi?id=1723367 [1] - https://bugzilla.redhat.com/show_bug.cgi?id=1723367#c2 [2] - https://bugzilla.redhat.com/show_bug.cgi?id=1723367#c3 [3] - https://github.com/libvirt/libvirt/blob/master/src/qemu/qemu_interface.c#L4… [4] - https://github.com/libvirt/libvirt/blob/master/src/qemu/qemu_interface.c#L4…

3 7

Do we need "amd_iommu=on" for AMD system anymore?
by Yalan Zhang 30 Jun '20

30 Jun '20

Hi, Since long time ago, to enable the SR-IOV VF pci passthrough function, I'm always adding "amd_iommu=on" into kernel cmdline on AMD system. But recently I found even I do not do this action, IOMMU is still enabled by kernel on AMD system. After searching, I found there is no such setting any more, refer to https://github.com/torvalds/linux/blob/master/Documentation/admin-guide/ker… There are only 3 possible value as below: amd_iommu= fullflush off force_isolation Could anyone can help to confirm the changes? Thank you! And another question, it is said that the "iommu=pt" option is to improves IO performance for devices in the host, it is not a must for VF PCI passthrough, right? I'm not sure about the user cases. [Reference] 1. Reference about adding "amd_iommu=on", and it may be outdated: http://dpdk-guide.gitlab.io/dpdk-guide/setup/iommu.html 2. On AMD system without adding "amd_iommu=on" in the kernel cmdline, the iommu is enabled: # cat /proc/cmdline BOOT_IMAGE=(hd0,msdos1)/vmlinuz-4.18.0-193.el8.x86_64 root=/dev/mapper/rhel_hp--dl385g10--16-root ro crashkernel=auto resume=/dev/mapper/rhel_hp--dl385g10--16-swap rd.lvm.lv=rhel_hp-dl385g10-16/root rd.lvm.lv=rhel_hp-dl385g10-16/swap console=ttyS0,115200n81 # dmesg | grep -i iommu [ 3.712029] iommu: Default domain type: Passthrough [ 6.736019] pci 0000:00:00.2: IOMMU performance counters supported ... [ 6.780040] pci 0000:e0:00.2: IOMMU performance counters supported [ 6.786740] pci 0000:00:01.0: Adding to iommu group 0 [ 6.791876] pci 0000:00:01.1: Adding to iommu group 1 [ 6.797015] pci 0000:00:01.2: Adding to iommu group 2 [ 6.802150] pci 0000:00:01.4: Adding to iommu group 3 ... [ 7.866463] pci 0000:e0:00.2: Found IOMMU cap 0x40 [ 7.920222] perf/amd_iommu: Detected AMD IOMMU #0 (2 banks, 4 counters/bank). [ 7.927428] perf/amd_iommu: Detected AMD IOMMU #1 (2 banks, 4 counters/bank) ... ------- Best Regards, Yalan Zhang IRC: yalzhang

2 1

Virtual Bridge "Network" for Sandbox
by Paul O'Rorke 29 Jun '20

29 Jun '20

Hi all, I couldn't find any documentation on this, hopefully someone can point me in the right direction. I recently set up a sand-boxed environment for our developers. There are domain controller(s), workstations and servers in there. The whole thing is running on a single host using a "Virtual Network" defined in virt-manager on that host. Now I find I want to add more guests and there are not enough resources on this one host. Can I somehow make this Virtual Network available to two hosts? I do not want to move to a bridged network and have to physically join the two hosts with a discrete link when they are already on the same subnet at the host level. Is that possible? -- *Paul O'Rorke*

3 6

get contents of a storage volume in bytes
by Shashwat shagun 28 Jun '20

28 Jun '20

Hi, i have two machines with libvirt installed. I want to move a storage volume from the 1st machine to 2nd machine with code. I'm thinking of reading a storage volume into a byte array/stream then uploading through my own/custom handler (written in go) to the 2nd machine's custom handler which will then be writing the byte stream to libvirt storage volume. Is there a libvirt function to read storage volume into byte stream? Shashwat. shashwatshagun2581(a)gmail.com

3 2

RBD volume pools and locks
by Guy Godfroy 26 Jun '20

26 Jun '20

Hello, One of my libvirt cluster is using a RBD volume pool. I wish that I could configure libvirt lock manager to use RBD locks. Is that possible? I read here https://libvirt.org/kbase/locking.html that libvirt lock manager is shipped with 2 active plugins. Is there any other plugins that exist? Thanks. Guy Godfroy

1 0

virsh edit does not work when <initiator> and <auth> is used in config
by Fourhundred Thecat 25 Jun '20

25 Jun '20

Hello, I am having problem when using: "virsh edit <vm_name>" my VM has network iscsi disk defined: <disk type='network' device='disk'> <driver name='qemu' type='raw'/> <source protocol='iscsi' name='iqn.1992-08.com.netapp:5481.60080e50001ff2000000000051aee24d/0'> <host name='10.1.212.52' port='3260'/> <initiator> <iqn name='iqn.2013-01.bla.bla:01:test'/> </initiator> <auth username='myname'> <secret type='iscsi' usage='libvirtiscsi'/> </auth> </source> ... </disk> when I defined thje VM the first time, as always, libvirt reorders the lines in the XML config file as it likes. One of the reordering it did, was to put the "<initiator>" block above the "<auth>" block. But once I want to edit, "virsh edit <vm_name>", whatever change I make, even unrelated to iscsi disk, it reports error: error: XML document failed to validate against schema: Unable to validate doc against /usr/share/libvirt/schemas/domain.rng Extra element devices in interleave Element domain failed to validate content after long trial and error, I found I can finally save it when I reorder the "<initiator>" and "<auth>" blocks, so that "<auth>" is above "<initiator>". Once i save it, libvirt then reorders it back to the order, which will generate erro next time I attemt to edit. Anyway, this seems like a bug, and especially evil one. How could I get rid of this behaviour? thanks,

2 1

volume pool for LXC
by Guy Godfroy 23 Jun '20

23 Jun '20

Hello, Is it possible to use volume pools for LXC ? For instance LVM pools. It seems to be supporte by LXC bare commands. From the man page of lxc-create : -B, --bdev backingstore 'backingstore' is one of 'dir', 'lvm', 'loop', 'btrfs', 'zfs', 'rbd', or 'best'. The default is 'dir', meaning that the container root filesystem will be a directory under /var/lib/lxc/container/rootfs. This backing store type allows the optional --dir ROOTFS to be specified, meaning that the container rootfs should be placed under the specified path, rather than the default. (The 'none' backingstore type is an alias for 'dir'.) If 'btrfs' is specified, then the target filesystem must be btrfs, and the container rootfs will be created as a new subvolume. This allows snapshotted clones to be created, but also causes rsync --one-filesystem to treat it as a separate filesystem. If backingstore is 'lvm', then an lvm block device will be used and the following further options are available: --lvname lvname1 will create an LV named lvname1 rather than the default, which is the container name. --vgname vgname1 will create the LV in volume group vgname1 rather than the default, lxc. --thinpool thinpool1 will create the LV as a thin-provisioned volume in the pool named thinpool1 rather than the default, lxc. --fstype FSTYPE will create an FSTYPE filesystem on the LV, rather than the default, which is ext4. --fssize SIZE will create a LV (and filesystem) of size SIZE rather than the default, which is 1G. If backingstore is 'loop', you can use --fstype FSTYPE and --fssize SIZE as 'lvm'. The default values for these options are the same as 'lvm'. If backingstore is 'rbd', then you will need to have a valid configuration in ceph.conf and a ceph.client.admin.keyring defined. You can specify the following options : --rbdname RBDNAME will create a blockdevice named RBDNAME rather than the default, which is the container name. --rbdpool POOL will create the blockdevice in the pool named POOL, rather than the default, which is 'lxc'. If backingstore is 'best', then lxc will try, in order, btrfs, zfs, lvm, and finally a directory backing store.

1 0

qemu hook: event for source host too
by Guy Godfroy 23 Jun '20

23 Jun '20

Hello, this is my first time posting on this mailing list. I wanted to suggest a addition to the qemu hook. I will explain it through my own use case. I use a shared LVM storage as a volume pool between my nodes. I use lvmlockd in sanlock mode to protect both LVM metadata corruption and concurrent volume mounting. When I run a VM on a node, I activate the desired LV with exclusive lock (lvchange -aey). When I stop the VM, I deactivate the LV, effectively releasing the exclusive lock (lvchange -an). When I migrate a VM (both live and offline), the LV has to be activated on both source and target nodes, so I have to use a shared lock (lvchange -asy). That's why I need a hook event on the source host too (as far as I know after my tests, the migration event is only triggered on the target host). Is such a feature a possibility? Thanks for your attention. Guy Godfroy

2 13

Missing vnet interface in migration
by Ales Musil 22 Jun '20

22 Jun '20

Hi, in oVirt we have come across issue with a missing vnet interface (Original bug report [0]). This happened once in a huge batch of migrations e.g 2 failed out of 1200 and we came to conclusion that this is happening somewhere out of our control as libvirt reports: 2020-05-17 02:34:11,530+0000 ERROR (migsrc/623f6cc1) [virt.vm] (vmId='623f6cc1-d49f-43a4-8170-b5fec01f2897') operation failed: binding 'vnet54' is already being removed (migration:278) Unfortunately we don't have libvirt logs and producing them might not be possible. Is there anything else we can do to tackle down this issue? Thank you. Regards, Ales [0] https://bugzilla.redhat.com/1837395 -- Ales Musil Software Engineer - RHV Network Red Hat EMEA <https://www.redhat.com> amusil(a)redhat.com IM: amusil <https://red.ht/sig>

1 0

Feature request? Auto start vm upon next shutdown
by Marc Roos 22 Jun '20

22 Jun '20

Sometimes when you change the configuration, this configuration change will only be available after a shutdown. Not to have to monitor when the vm shuts down, to start it again. It could be nice to have libvirt start it one time. Something like: 1. change the network interface of a running guest 2. at the prompt of the virt-manager "Some changes may require a guest shutdown to take effect" introduce check box 'Auto start vm upon next shutdown' 3. shutdown the guest 4. after the guest completed shutdown, libvirt is automatically starting it again.

3 4

NVDIMM sizes and DIMM hot plug
by Milan Zamazal 19 Jun '20

19 Jun '20

Hi, I've found out that NVDIMM size and label size matter for regular (non-NV) DIMM hot plug. If the NVDIMM is not aligned correctly, the guest OS will not accept the hot plugged memory and will complain with messages such as Block size [0x8000000] unaligned hotplug range: start 0x225000000, size 0x10000000 The start address above is also reported within <memory> element of the hot plugged memory in the domain XML: <address type='dimm' slot='1' base='0x225000000'/> Apparently, in order to make memory hot plug working in the guest OS, the inserted memory must be aligned to the platform memory alignment (128 MB on x86_64). I'd like to clarify, how libvirt makes the DIMM address above. How is the NVDIMM memory range determined? According to my experiments, it seems the NVDIMM specified <size> is taken, NVDIMM <label> size is subtracted from it and the resulting value is reduced to the nearest multiple of NVDIMM <alignsize>. Is this observation correct? Is it guaranteed to be stable in future versions? I need to determine the right NVDIMM size to make the subsequent memory modules correctly aligned and then I can't change the NVDIMM size, to not damage data stored in the NVDIMM. Additionally, when adjusting maxMemory due to NVDIMM presence, should I increase it by the specified NVDIMM <size> or a different value? Thank you, Milan

4 5

Guidance for First Timer
by Sanyam Rajpal 18 Jun '20

18 Jun '20

Hey everyone, I am a first-time contributor and I was interested in this project. I wanted to make some open source contributions and learn. I am not looking for getting enrolled in the program. Sanyam

2 1

Is it possible to configure libvirt's MAC generation?
by Ian Pilcher 16 Jun '20

16 Jun '20

Is it possible to configure libvirt to generate "predictable" MAC addresses for virtual NICS? (A configuration which accomplishes this by limiting the pool of available addresses would be acceptable for my use case.) Read on for why I want to do this ... I have a somewhat unusual use case. I am working with the OpenShift bare metal "IPI" installation process, which is documented here: https://openshift-kni.github.io/baremetal-deploy/4.4/Deployment.html At a high level, the installation process goes like this: 1. Install RHEL 8 on the "provisioner" node. 2. Configure a couple of bridges on the provisioner node, which will be used by the "bootstrap VM" (see below). 3. Create a configuration file & do various other things. 4. Run the installer a. Installer creates a "bootstrap VM" on the provisioning node, which does the actual work of kicking off the OpenShift installation. The bootstrap VM runs Red Hat Enterprise Linux CoreOS (RHCOS). b. Bootstrap VM requests an IP address via DHCP. <====== ****** Step 4b is problematic in this environment. Because of routing and firewall rules, I really need the bootstrap VM to have a predictable IP address. It is possible to configure a static IP address on the bootstrap VM via the CoreOS ignition file, and we've successfully done this, but ... RHCOS requests an IP address via DHCP *before* it processes the ignition file. If this fails, it appears to simply give up. So we need to have our DHCP server provide an address to the bootstrap VM, even if it isn't ever used. And of course, we're not allowed to configure the DHCP server to respond to any old MAC address that comes along; we're only allowed to respond to known MAC addresses. Hopefully this all makes sense, and hopefully it explains why being able to configure libvirt's MAC pool/sequence would allow things to work in this environment. Thanks! -- ======================================================================== In Soviet Russia, Google searches you! ========================================================================

6 6

Reintroduce modern CPU in model selection
by Gionatan Danti 15 Jun '20

15 Jun '20

Hi list, in virt-manager ver. 2.2.1 (fully upgraded CentOS 8.1), the CPU model list only shows ancient CPU (the most recent is Nehalem-IBRS). On the other hand, in virt-manager 1.5.x (fully upgraded CentOS 7.8) we have a rich selection of CPU (as recent as Icelake). Why was the list in newer virt-manager so much trimmed? Is it possible to enlarge it? Thanks. -- Danti Gionatan Supporto Tecnico Assyoma S.r.l. - www.assyoma.it [1] email: g.danti(a)assyoma.it - info(a)assyoma.it GPG public key ID: FF5F32A8

1 1

Live migration to another volume at the same host
by Paul van der Vlis 12 Jun '20

12 Jun '20

Hello, I am looking for a way to migrate a VM live to another location. The qcow2 images are now on /data , but I want to move some of them to a volume mounted on /data2 what's faster. Is this possible without downtime? With regards, Paul -- Paul van der Vlis Linux systeembeheer Groningen https://www.vandervlis.nl/

2 2

Migrate to a bigger disk possible?
by Paul van der Vlis 11 Jun '20

11 Jun '20

Hello, I have a qcow2 disk what needs to become increased. I don't have space on the location where it is now to have it two time, so I want to live-migrate it to another host. On the other host I have to create a qcow2 disk before I can migrate. What would happen when I would create there a bigger qcow2 disk then the original one? Can I resize the filesystem after the migration? With regards, Paul van der Vlis -- Paul van der Vlis Linux systeembeheer Groningen https://www.vandervlis.nl/

3 9

Fwd: Hello
by K. Kahurani 11 Jun '20

11 Jun '20

---------- Forwarded message --------- From: K. Kahurani <k.kahurani(a)gmail.com> Date: Sun, May 31, 2020 at 2:26 PM Subject: Hello To: <eblake(a)redhat.com> Hello, Hopefully this finds you well. This is not a bug report but more or less an inquiry as based on the information acquired from Jim Fehlig a while back, it is possible you would have some important information regarding this. It is currently not possible for me to compile libvirt right from the word go while configuring with the error [1]. From the look of it and a bit of searching on the internet, it does look like the package portablexdr is obsolete. Could it be that this is a known issue? Could it be there already is a workaround this? Do you suppose this should go into the mailing list? Sincerely, David. [1] checking for WIRESHARK_DISSECTOR... no checking for xdrmem_create in -lportablexdr... no checking for library containing xdrmem_create... no configure: error: Cannot find a XDR library

2 2

Disable virtlockd
by Felix Queißner 09 Jun '20

09 Jun '20

Hello! Is it possible to disable the virtlockd daemon or VM file locking? I start qemu with a -snapshot option which prevents and changes to the disk image anyways. Using <readonly /> is not supported for IDE disks. Another option would be to not require locking on the NFS share, but i have no idea how. Can someone help me with that? Regards Felix Queißner

2 2

No outbound connectivity from guest VM(fedora 32)
by Justin Stephenson 09 Jun '20

09 Jun '20

Hi, I recently installed a fresh install of Fedora 32 and I am having trouble with my virtual machine networking, I can ssh and connect into my guest VMs from my host, but the guest VMs cannot ping out to the internet. I am using the "default" NAT virtual network, the interesting thing is I have made no configuration changes on my host or in the guest VMs, simply created and installed two VMs(Fedora and RHEL8) in Fedora where the VMs are having the same issue. I am happy to provide any logs or command output if that would help. -Justin

3 3

Permission to disk set wrong when restoring from memory snapshot?
by Liran Rotenberg 09 Jun '20

09 Jun '20

Hi all, Passing on Bug 1840609 <https://bugzilla.redhat.com/show_bug.cgi?id=1840609> - Wake up from hibernation failed:internal error: unable to execute QEMU command 'cont': Failed to get "write" lock. In Ovirt/RHV there is a specific flow that prevents the VM from starting on the first host. The result is: 2020-06-09T12:12:58.111610Z qemu-kvm: Failed to get "write" lock Is another process using the image [/rhev/data-center/3b67fb92-906b-11ea-bb36-482ae35a5f83/4fd23357-6047-46c9-aa81-ba6a12a9e8bd/images/0191384a-3e0a-472f-a889-d95622cb6916/7f553f44-db08-480e-8c86-cbdeccedfafe]? 2020-06-09T12:12:58.668140Z qemu-kvm: terminating on signal 15 from pid 177876 (<unknown process>) We have a rerun mechanism. Therefore the VM starts again on another host if available. On the second try the VM does succeed to run. After looking in the debug logs (attached in the bug), I have seen this on the first host: 2020-06-09 12:12:46.288+0000: 177879: debug : qemuSetupImageCgroupInternal:139 : Not updating cgroups for disk path '<null>', type: file 2020-06-09 12:12:46.288+0000: 177879: debug : qemuSetupImagePathCgroup:75 : Allow path /rhev/data-center/3b67fb92-906b-11ea-bb36-482ae35a5f83/4fd23357-6047-46c9-aa81-ba6a12a9e8bd/images/0191 384a-3e0a-472f-a889-d95622cb6916/7f553f44-db08-480e-8c86-cbdeccedfafe, perms: rw *2020-06-09 12:12:46.288+0000: 177879: debug : qemuSetupImagePathCgroup:75 : Allow path /rhev/data-center/mnt/yellow-vdsb.qa.lab.tlv.redhat.com:_Compute__NFS_GE_compute-ge-4_nfs__0/4fd23357-6047-46c9-aa81-ba6a12a9e8bd/images/0191384a-3e0a-472f-a889-d95622cb6916/7f553f44-db08-480e-8c86-cbdeccedfafe, perms: r* And immediately after retry on the second host: 2020-06-09 12:13:01.839+0000: 15781: debug : qemuSetupImagePathCgroup:75 : Allow path /rhev/data-center/mnt/yellow-vdsb.qa.lab.tlv.redhat.com:_Compute__NFS_GE_compute-ge-4_nfs__0/4fd23357-60 47-46c9-aa81-ba6a12a9e8bd/images/0191384a-3e0a-472f-a889-d95622cb6916/7f553f44-db08-480e-8c86-cbdeccedfafe, perms: rw Is it intended and might cause the QEMU error? Regards, Liran.

2 4

libvirt-defined network vs/and openvswitch bridge
by Po Dragonwarrior 01 Jun '20

01 Jun '20

Hi all, Can you please explain to me what is the difference (or maybe the relation?) between a libvirt defined network and an openvswitch bridge? best, Po

2 1

About Libvirt Setmem&dommemstat Function
by 🍭 20 May '20

20 May '20

I have sent a similar email about this problem. But I did not describe carefully, so I would like to explain it in more details. I am using 'virsh setmem' to ajust vm memory online. However, I don't know what is the lowerlimit that can be set to. And I try to use 'virsh dommemstat' to get 'unused' memory so that I can calculate the lowerlimit memory with this value, but it doesn't work when it comes to windows systems as I cannot set vm memory down to 'available - unused' in windows system. I suppose it's caches memory that causes this problem, so I upgrade my libvirt & qemu in order to get 'stat_disk_caches'. Although I upgrade libvirt to v4.10.0, qemu upgrade to v4.1.0, I cannot get this value. In summary, I came across two problems. How can I know the exact lowerlimit memory that can be set to in one vm? And if I need 'stat_disk_caches' value, how can I get it properly? My libvirt version: v4.10.0, qemu version:v4.1.0, hypervisor working on centos7.6 system.

1 0

macvtap direct
by Subhendu Ghosh 19 May '20

19 May '20

Hi Couple of questions around macvtap direct usage: 1) is the document here current? https://libvirt.org/formatnetwork.html#examplesDirect I have been able to get host to guest network traffic without any special configuration or switch since Fedora 28 when I first started using it. Using <forward mode=vepa> requires switch port mirroring, but just using <forward mode=bridge> doesn't. 2) do any of the language libraries make assumptions that libvirt networks must have a <bridge name=xx> attribute? Foreman's Ruby interface to libvirt errors out with attempting to build a VM on a KVM host with a network defined with <forward mode=bridge> https://projects.theforeman.org/issues/25890 thanks Subhendu

2 3