- Users - libvirt List Archives

[libvirt-users] Using hostdev to plug a PCI-E host device into Q35 pcie-root port
by Thomas Kuther 19 Nov '13

19 Nov '13

Hello, I'm trying to migrate a working qemu command line configuration to libvirt. The part I'm currently failing on is: $ qemu-system-x86_64 -M Q35 ... -device vfio-pci,host=05:00.0,bus=pcie.0 The right way to translate this into libvirt XML seems to be using <hostdev>, but I seem to be unable to plug it into the pcie-root port This is how the interesting part looks like when I let "virsh edit" generate an <address> <controller type='pci' index='0' model='pcie-root'/> <controller type='pci' index='1' model='dmi-to-pci-bridge'> <address type='pci' domain='0x0000' bus='0x00' slot='0x02' function='0x0'/> </controller> <controller type='pci' index='2' model='pci-bridge'> <address type='pci' domain='0x0000' bus='0x01' slot='0x01' function='0x0'/> </controller> [...] <hostdev mode='subsystem' type='pci' managed='yes'> <driver name='vfio'/> <source> <address domain='0x0000' bus='0x03' slot='0x00' function='0x0'/> </source> <address type='pci' domain='0x0000' bus='0x02' slot='0x06' function='0x0'/> </hostdev> [...] To my understanding, this will plug the host device into the pci-bridge controller. The guest OS doesn't boot with this and resets right after bios. Manually setting <address type='pci' domain='0x0000' bus='0x00' slot='0x1E' function='0x0'/> cause XML validation failure. Is there any way in libvirt XML to plug a host's PCI-E device directly into the pcie-root port, like it works on qemu command line? I'm aware I could use something like <qemu:commandline> <qemu:arg value='-device'/> <qemu:arg value='vfio-pci,host=05:00.0,bus=pcie.0'/> </qemu:commandline> but I insist on running the VM as non-root, and if I got that right I need to configure at least one vfio device (or memory locking) in order for libvirt to set a proper RLIMIT_MEMLOCK value. Any help would be be appreciated. Regards, Thomas

2 3

Re: [libvirt-users] Unable to use more than 4 serial devices at once.
by Cole Robinson 19 Nov '13

19 Nov '13

On 11/17/2013 09:18 PM, Matthew Harrold wrote: > On Sun, Nov 17, 2013 at 7:44 PM, Cole Robinson <crobinso(a)redhat.com > <mailto:crobinso@redhat.com>> wrote: > On 11/17/2013 01:26 PM, Matthew Harrold wrote: > > Hi All. > > > > I'm trying to setup a guest within my hypervisor that acts as a Console > > Server, using 10 USB to Serial adapters. The Host is all setup and can > access > > each Serial Port without problems. > > > > I have attempted to add the Serial Connections to a Guest, using both > Virtual > > Machine Manager and by editing the .XML file, and I can add up to 4 Serial > > devices (Including one for console access to the guest) without any > issues at > > all. However when I try to add a fifth , the virtual machine refuses to > start > > > > > > Distro: Ubuntu 13.04 > > Kernel: 3.8.0-33-generic > > libvirtd: 1.0.2 > > QEMU:1.4.0 (Debian 1.4.0+dfsg-1expubuntu4) > > > > The log- > > > > _4 Serial Devices_ > > > > 2013-11-17 17:37:45.901+0000: starting up > > LC_ALL=C PATH=/usr/local/sbin:/usr/local/bin:/usr/bin:/usr/sbin:/sbin:/bin > > QEMU_AUDIO_DRV=none /usr/bin/kvm -name Ubuntu12.04-Cisco -S -M pc-1.0 -m > 1024 > > -smp 2,sockets=2,cores=1,threads=1 -uuid > 1837c56f-6f84-2abe-8bdd-d7c974803b84 > > -no-user-config -nodefaults -chardev > > > socket,id=charmonitor,path=/var/lib/libvirt/qemu/Ubuntu12.04-Cisco.monitor,server,nowait > > -mon chardev=charmonitor,id=monitor,mode=control -rtc base=utc -no-shutdown > > -device piix3-usb-uhci,id=usb,bus=pci.0,addr=0x1.0x2 -drive > > > file=/mnt/DiskRepo/Ubuntu12.04-TFTP.qcow2,if=none,id=drive-virtio-disk0,format=qcow2 > > -device > > > virtio-blk-pci,scsi=off,bus=pci.0,addr=0x4,drive=drive-virtio-disk0,id=virtio-disk0,bootindex=1 > > -drive if=none,id=drive-ide0-1-0,readonly=on,format=raw -device > > ide-cd,bus=ide.1,unit=0,drive=drive-ide0-1-0,id=ide0-1-0 -netdev > > tap,fd=31,id=hostnet0,vhost=on,vhostfd=32 -device > > > virtio-net-pci,netdev=hostnet0,id=net0,mac=52:54:00:e6:9c:c0,bus=pci.0,addr=0x3 > -chardev > > pty,id=charserial0 -device isa-serial,chardev=charserial0,id=serial0 > -chardev > > tty,id=charserial1,path=/dev/c1841-A -device > > isa-serial,chardev=charserial1,id=serial1 -chardev > > tty,id=charserial2,path=/dev/c2620-A -device > > isa-serial,chardev=charserial2,id=serial2 -chardev > > tty,id=charserial3,path=/dev/c2620-B -device > > isa-serial,chardev=charserial3,id=serial3 -vnc 0.0.0.0:0 > <http://0.0.0.0:0> <http://0.0.0.0:0> > > -vga cirrus -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x5 > > W: kvm binary is deprecated, please use qemu-system-x86_64 instead > > char device redirected to /dev/pts/0 (label charserial0) > > > > > > _5 Serial Devices_ > > > > 2013-11-17 18:05:59.925+0000: starting up > > LC_ALL=C PATH=/usr/local/sbin:/usr/local/bin:/usr/bin:/usr/sbin:/sbin:/bin > > QEMU_AUDIO_DRV=none /usr/bin/kvm -name Ubuntu12.04-Cisco -S -M pc-1.0 -m > 1024 > > -smp 2,sockets=2,cores=1,threads=1 -uuid > 1837c56f-6f84-2abe-8bdd-d7c974803b84 > > -no-user-config -nodefaults -chardev > > > socket,id=charmonitor,path=/var/lib/libvirt/qemu/Ubuntu12.04-Cisco.monitor,server,nowait > > -mon chardev=charmonitor,id=monitor,mode=control -rtc base=utc -no-shutdown > > -device piix3-usb-uhci,id=usb,bus=pci.0,addr=0x1.0x2 -drive > > > file=/mnt/DiskRepo/Ubuntu12.04-TFTP.qcow2,if=none,id=drive-virtio-disk0,format=qcow2 > > -device > > > virtio-blk-pci,scsi=off,bus=pci.0,addr=0x4,drive=drive-virtio-disk0,id=virtio-disk0,bootindex=1 > > -drive if=none,id=drive-ide0-1-0,readonly=on,format=raw -device > > ide-cd,bus=ide.1,unit=0,drive=drive-ide0-1-0,id=ide0-1-0 -netdev > > tap,fd=31,id=hostnet0,vhost=on,vhostfd=32 -device > > > virtio-net-pci,netdev=hostnet0,id=net0,mac=52:54:00:e6:9c:c0,bus=pci.0,addr=0x3 > -chardev > > pty,id=charserial0 -device isa-serial,chardev=charserial0,id=serial0 > -chardev > > tty,id=charserial1,path=/dev/c1841-A -device > > isa-serial,chardev=charserial1,id=serial1 -chardev > > tty,id=charserial2,path=/dev/c2620-A -device > > isa-serial,chardev=charserial2,id=serial2 -chardev > > tty,id=charserial3,path=/dev/c2620-B -device > > isa-serial,chardev=charserial3,id=serial3 -chardev > > tty,id=charserial4,path=/dev/c2950-A -device > > isa-serial,chardev=charserial4,id=serial4 -vnc 0.0.0.0:0 > <http://0.0.0.0:0> <http://0.0.0.0:0> > > -vga cirrus -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x5 > > W: kvm binary is deprecated, please use qemu-system-x86_64 instead > > char device redirected to /dev/pts/0 (label charserial0) > > qemu-system-x86_64: -device isa-serial,chardev=charserial4,id=serial4: > Device > > 'isa-serial' could not be initialized > > 2013-11-17 18:06:00.314+0000: shutting down > > > > Does anyone have any idea what's causing this? I have attempted using > > different Serial to USB adapters to rule out a hardware / configuration > issue > > on one of the adapters, but it makes no difference. If I try to start more > > than 4 Serial Devices, it refuses to start up. > > > Pretty sure that's just a qemu or PC limit for isa-serial. The alternative is > to use virtio serial which can handle many more ports. > - Cole > > > That sounds likely. Is there any documentation covering 'binding' a host > serial port to a virtio serial device? I've search but not found anything > relevant. > Hmm, that's a good question. Actually it doesn't look like there is any way to have <serial> use virtio. You can use channel to get something close, like: <channel type='pty'> <target type='virtio' name='my.port.2'/> </channel> However it won't appear as ttyS0 in the guest so it might not fit your usecase. Recent qemu has a PCI serial device which should allow for more than 4 serial ports, but support isn't wired up in libvirt yet. - Cole

1 0

[libvirt-users] Asterisk, pci passthrough, and pri card
by Mauricio Tavares 19 Nov '13

19 Nov '13

Has anyone been successfully running asterisk in libvirt with a pri card? I thought we had it working but we seem to be experiencing long (1h+) calls being dropped.

2 1

[libvirt-users] macvtap direct and ip spoofing
by vlad halilov 19 Nov '13

19 Nov '13

Hi there. I have configured kvm domain (rhel6.4) with ethernet bridged over macvtap, and found no filtration applied except mac. 'virsh' just silently ignoring attributes 'filterref' and 'ip address' in different formats. No error on validate stage. Config examples: ... <interface type='direct'> <mac address='52:54:00:31:ae:1a'/> <source dev='em1' mode='private'/> <filterref filter='clean-traffic'> <parameter name='IP' value='10.1.101.44'/> </filterref> <model type='virtio'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> </interface> ... or like these: ... <interface type='direct'> <mac address='52:54:00:31:ae:1a'/> <source dev='em1' mode='private'/> <ip address='10.1.101.44'/> <filterref filter='clean-traffic'/> <model type='virtio'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> </interface> ... With 'virsh create domain.xml', vm created, but dumpxml show that filterref is disappeared. I have not found any success stories with filtering rules and 'direct' interface types. Is it supported with this type? Or may be other tricks to protect network from vm spoofing and direct type? - vlad f halilov

2 2

[libvirt-users] Hot resize virtio-scsi device with multipath device backend
by Baptiste AGASSE 18 Nov '13

18 Nov '13

Hi all, I'm currently testing virtio-scsi on EL6.4. I've seen that we can resize scsi disk without restarting the guest for qcow files backend for example (this point is well documented). I want to know if it's possible to "hot resize" a multipath device attached to a guest via virtio-scsi. I use multipath device hosted to a SAN array to store data used on virtual guests. I've attached this device with virtio-scsi device to a guest. Now i want to resize this device without restarting my guest. I've resized the disk on my SAN array, refreshed iscsi connections and multipathed device to reflect the new size, but how to say to my guest that device have been resized ? I've tried "echo 1 > /sys/class/scsi_disk/2\:0\:0\:2/device/rescan" in my guest but it didn't works. I think i'm missing a step (libvirt notify the guest that the device have changed ?) and i don't find any documentation about resizing direct attached physical device via virtio-scsi. Is it possible to do this ? kernel.x86_64 2.6.32-358.14.1.el6 libvirt.x86_64 0.10.2-18.el6_4.14 qemu-kvm.x86_64 2:0.12.1.2-2.355.0.1.el6_4.9 Have a nice day. Regards. Bapiste.

1 0

[libvirt-users] Unable to use more than 4 serial devices at once.
by Matthew Harrold 17 Nov '13

17 Nov '13

Hi All. I'm trying to setup a guest within my hypervisor that acts as a Console Server, using 10 USB to Serial adapters. The Host is all setup and can access each Serial Port without problems. I have attempted to add the Serial Connections to a Guest, using both Virtual Machine Manager and by editing the .XML file, and I can add up to 4 Serial devices (Including one for console access to the guest) without any issues at all. However when I try to add a fifth , the virtual machine refuses to start *The Host-* Distro: Ubuntu 13.04 Kernel: 3.8.0-33-generic libvirtd: 1.0.2 QEMU:1.4.0 (Debian 1.4.0+dfsg-1expubuntu4) The log- *4 Serial Devices* 2013-11-17 17:37:45.901+0000: starting up LC_ALL=C PATH=/usr/local/sbin:/usr/local/bin:/usr/bin:/usr/sbin:/sbin:/bin QEMU_AUDIO_DRV=none /usr/bin/kvm -name Ubuntu12.04-Cisco -S -M pc-1.0 -m 1024 -smp 2,sockets=2,cores=1,threads=1 -uuid 1837c56f-6f84-2abe-8bdd-d7c974803b84 -no-user-config -nodefaults -chardev socket,id=charmonitor,path=/var/lib/libvirt/qemu/Ubuntu12.04-Cisco.monitor,server,nowait -mon chardev=charmonitor,id=monitor,mode=control -rtc base=utc -no-shutdown -device piix3-usb-uhci,id=usb,bus=pci.0,addr=0x1.0x2 -drive file=/mnt/DiskRepo/Ubuntu12.04-TFTP.qcow2,if=none,id=drive-virtio-disk0,format=qcow2 -device virtio-blk-pci,scsi=off,bus=pci.0,addr=0x4,drive=drive-virtio-disk0,id=virtio-disk0,bootindex=1 -drive if=none,id=drive-ide0-1-0,readonly=on,format=raw -device ide-cd,bus=ide.1,unit=0,drive=drive-ide0-1-0,id=ide0-1-0 -netdev tap,fd=31,id=hostnet0,vhost=on,vhostfd=32 -device virtio-net-pci,netdev=hostnet0,id=net0,mac=52:54:00:e6:9c:c0,bus=pci.0,addr=0x3 -chardev pty,id=charserial0 -device isa-serial,chardev=charserial0,id=serial0 -chardev tty,id=charserial1,path=/dev/c1841-A -device isa-serial,chardev=charserial1,id=serial1 -chardev tty,id=charserial2,path=/dev/c2620-A -device isa-serial,chardev=charserial2,id=serial2 -chardev tty,id=charserial3,path=/dev/c2620-B -device isa-serial,chardev=charserial3,id=serial3 -vnc 0.0.0.0:0 -vga cirrus -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x5 W: kvm binary is deprecated, please use qemu-system-x86_64 instead char device redirected to /dev/pts/0 (label charserial0) *5 Serial Devices* 2013-11-17 18:05:59.925+0000: starting up LC_ALL=C PATH=/usr/local/sbin:/usr/local/bin:/usr/bin:/usr/sbin:/sbin:/bin QEMU_AUDIO_DRV=none /usr/bin/kvm -name Ubuntu12.04-Cisco -S -M pc-1.0 -m 1024 -smp 2,sockets=2,cores=1,threads=1 -uuid 1837c56f-6f84-2abe-8bdd-d7c974803b84 -no-user-config -nodefaults -chardev socket,id=charmonitor,path=/var/lib/libvirt/qemu/Ubuntu12.04-Cisco.monitor,server,nowait -mon chardev=charmonitor,id=monitor,mode=control -rtc base=utc -no-shutdown -device piix3-usb-uhci,id=usb,bus=pci.0,addr=0x1.0x2 -drive file=/mnt/DiskRepo/Ubuntu12.04-TFTP.qcow2,if=none,id=drive-virtio-disk0,format=qcow2 -device virtio-blk-pci,scsi=off,bus=pci.0,addr=0x4,drive=drive-virtio-disk0,id=virtio-disk0,bootindex=1 -drive if=none,id=drive-ide0-1-0,readonly=on,format=raw -device ide-cd,bus=ide.1,unit=0,drive=drive-ide0-1-0,id=ide0-1-0 -netdev tap,fd=31,id=hostnet0,vhost=on,vhostfd=32 -device virtio-net-pci,netdev=hostnet0,id=net0,mac=52:54:00:e6:9c:c0,bus=pci.0,addr=0x3 -chardev pty,id=charserial0 -device isa-serial,chardev=charserial0,id=serial0 -chardev tty,id=charserial1,path=/dev/c1841-A -device isa-serial,chardev=charserial1,id=serial1 -chardev tty,id=charserial2,path=/dev/c2620-A -device isa-serial,chardev=charserial2,id=serial2 -chardev tty,id=charserial3,path=/dev/c2620-B -device isa-serial,chardev=charserial3,id=serial3 -chardev tty,id=charserial4,path=/dev/c2950-A -device isa-serial,chardev=charserial4,id=serial4 -vnc 0.0.0.0:0 -vga cirrus -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x5 W: kvm binary is deprecated, please use qemu-system-x86_64 instead char device redirected to /dev/pts/0 (label charserial0) qemu-system-x86_64: -device isa-serial,chardev=charserial4,id=serial4: Device 'isa-serial' could not be initialized 2013-11-17 18:06:00.314+0000: shutting down Does anyone have any idea what's causing this? I have attempted using different Serial to USB adapters to rule out a hardware / configuration issue on one of the adapters, but it makes no difference. If I try to start more than 4 Serial Devices, it refuses to start up. Thanks for your time, Matthew

2 1

[libvirt-users] /var/lib/libvirt/images on BTRFS (it works!)
by Gene Czarcinski 17 Nov '13

17 Nov '13

Some of you may be interested in running your qemu-kvm (or other) disk image files on BTRFS. After some fumbling around, I have found out how to do it so it works and works very well (especially if it is on an SSD). Anyone investigating running on BTRFS immediately finds out that there is a fragmentation problem (just putting image files on btrfs can result in them having between 50,000 and 100,000 extents). Some may have seen that adding nodatacow to the mount option is a solution ... it is not. So what is the solution? 1. execute: chattr +C /var/lib/libvirt/images or wherever your image files are. This needs to be done on the directory whether you have a separate subvol or not. 2. This ONLY works for new files and ONLY works for "raw" storage format files. Pretty much by definition, qcow2 files are sparse format files and that is the primary problem on btrfs. So that the vm creation wizard will create a new disk using the raw storage format, you need to select: Edit->Preferences->VM Details->Default Storage Format->Raw After then, an created virtual will have the correct disk format. You can use "filefrag" to verify that things are working. Gene

1 0

[libvirt-users] How to tell libvirt not to use qemu "accel=kvm" for a FreeBSD guest
by Christian Gagneraud 14 Nov '13

14 Nov '13

Hi there, I have used qemu to install FreeBSD 9.2 release, and i can boot it fine with: $ qemu-system-x86_64 -hda images/FreeBSD-9.2-RELEASE-amd64.img I then made virsh aware of my brand new VM (see attached XML), but when i run it via virsh, It hangs after the bootloader whith this message: Booting... CPU doesn't support long mode From the log, here is the generated qemu command: LC_ALL=C PATH=/usr/local/sbin:/usr/local/bin:/usr/bin:/usr/sbin:/sbin:/bin /usr/bin/qemu-system-x86_64 -name fbsd92-amd64 -S -machine pc-i440fx-1.5,accel=kvm,usb=off -m 512 -realtime mlock=off -smp 1,sockets=1,cores=1,threads=1 -uuid 11ae165e-b8a2-5083-fb77-37c0042725bf -nographic -no-user-config -nodefaults -chardev socket,id=charmonitor,path=/var/lib/libvirt/qemu/fbsd92-amd64.monitor,server,nowait -mon chardev=charmonitor,id=monitor,mode=control -rtc base=utc -no-shutdown -device piix3-usb-uhci,id=usb,bus=pci.0,addr=0x1.0x2 -drive file=/home/christiaga/projects/vms/images/FreeBSD-9.2-RELEASE-amd64.img,if=none,id=drive-ide0-0-0,format=raw -device ide-hd,bus=ide.0,unit=0,drive=drive-ide0-0-0,id=ide0-0-0,bootindex=1 -netdev tap,fd=24,id=hostnet0 -device rtl8139,netdev=hostnet0,id=net0,mac=52:54:00:9c:94:3b,bus=pci.0,addr=0x3 -chardev pty,id=charserial0 -device isa-serial,chardev=charserial0,id=serial0 -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x4 After many attempts, I pinned down the problem. To be able to boot the machine with the above command, I had to remove: * the accel=kvm in -machine * the -S (to "autostart" the VM) And change the -net tap,fd=24 to -net user,id=... So basically, the only thing i would like to tell libvirt is not to use accel=kvm, i guess the FreeBSD kernel is missing some feature (is it this PAE thing?) Does anyone how can i tell libvirt to do that? regards, Chris

3 5

[libvirt-users] Problem about lxc-enter-namespace interface
by hzguanqiang＠corp.netease.com 14 Nov '13

14 Nov '13

Hi experts, When I test lxc-enter-namespace interface, it turned out such an error: hzguanqiang@debian:~$ vir version Compiled against library: libvirt 1.1.4 Using library: libvirt 1.1.4 Using API: LXC 1.1.4 Running hypervisor: LXC 3.9.6 hzguanqiang@debian:~$ vir list Id Name State ---------------------------------------------------- 1544 instance-00000015 running hzguanqiang@debian:~$ vir lxc-enter-namespace 1544 --noseclabel /bin/df -hl error: Cannot get namespaces for 1545: Function not implemented hzguanqiang@debian:~$ ls /proc/self/ns ipc mnt net pid user uts What's the problem? ------------------ Best regards! GuanQiang 2013-11-14

2 10

[libvirt-users] UTS namespace
by Aarti Sawant 14 Nov '13

14 Nov '13

Hello, I am trying to set up UTS name space on centos 6.4 LXC container. one method is that we set the host name in /etc/sysconfig/network. hostname=test1 Is there any way so that i can set the hostname in libvirt xml while creating a container? I know using lxc-tools we can set hostname while creating by 2 ways - first while creating container using lxc-create -u test1 - Second by specifying lxc.utsname=test1 in config file Thanks Aarti Sawant NTTDATA OSS Center Pune

2 1