- Users - libvirt List Archives

[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

[libvirt-users] User Namespace in LXC
by Saurabh Deochake 14 Nov '13

14 Nov '13

Hi all, I'm trying to restrict privileges of root user inside the container. I searched about it and got to know about "idmap" element in domain XML. I added "idmap" element in my container's XML file: <idmap> <uid start='0' target='1000' count='10'/> <gid start='0' target='1000' count='10'/> </idmap> I restarted the container with updated XML file. When I execute "id" command to know if root user inside the container has been mapped with an user from host, i still get output uid as 0 # id -u root 0 Am I doing the steps right to check the user namespacing? Please help me out with this. Thanks in advance, Saurabh Deochake. NTT DATA OSS Center, Pune, India

2 6

[libvirt-users] Doc/web site problems
by Greg Ward 13 Nov '13

13 Nov '13

Hi all -- I have one meta-question, and two real questions. 1) What's the best place to report bugs in the docs, FAQ, web site, or wiki? I have found two -- see next points. (I'm happy to make the fixes myself if I can, but I'm pretty ignorant about libvirt. Can't fix the docs without knowing the truth.) 2) In the FAQ, under "networking is unavailable in virt-manager / virsh - libvirt's default network fails to start", it says "Here is a list of the networking-related packages that need to be installed for libvirt networking to work properly:". But it does not say if those packages are supposed to be installed on the host or the guest! It might be obvious to you, but it isn't to me. 3) In http://wiki.libvirt.org/page/Networking#Guest_configuration, it says "the following XML would be used in the guest:" ... but it doesn't say WHERE to put that configuration! Is it a snippet to add to an existing file? A new file I have to create? Something else? Very frustrating! ;-( Thanks -- Greg

2 4

[libvirt-users] Migration function is not supported by the connection driver: virDomainMigrate2
by Joseph Perrin 13 Nov '13

13 Nov '13

Greetings; I'm running Fedora 19, Xen 4.2.3, libvirt 1.0.5.6. I have two identical servers x1 & x2. I've read http://libvirt.org/migration.html & I've created certificates for tls according to http://wiki.libvirt.org/page/TLSSetup I can do this on both servers, and from a third admin server: # virsh -c xen+tls://x1.localdomain hostname x1.localdomain and... # virsh -c xen+tls://x2.localdomain hostname x2.localdomain It appears I can connect without issues. No matter what I try, I cannot get migration to work :-( x 1000! I've created a test DomU named "z19", which uses a DRBD block device between the two hosts. It's running on x1, (and obviously not on x2), and I want to migrate it to x2. When I attempt to migrate I get the following. (Although, I've tried every permutation of syntax I've read about. The following appears to be most reasonable for my situation?) # virsh migrate --live z19 xen+tls://x2.localdomain error: this function is not supported by the connection driver: virDomainMigrate2 # virsh -c xen+tls://x1.localdomain migrate z19 xen+tls://x2.localdomain error: this function is not supported by the connection driver: virDomainMigrate2 # virsh migrate --live z19 xen+ssh://root@x2.localdomain error: this function is not supported by the connection driver: virDomainMigrate2 I've set libvirtd.conf: log_level=1 Which reports nothing. There is a great amount of conflicting information on this topic. It's difficult to ascertain which to listen to. Any suggestions you could give me would be greatly appreciated. -Joseph

2 1