- Users - libvirt List Archives

[libvirt-users] Progressively worsening unresponsiveness
by Michael Warnecke 27 Aug '14

27 Aug '14

Hello all, I'm new to the list, and hoping someone can point me in the right direction. I've got an Ubuntu 14 dom0 with, what I think, are excellent specs. 12 cores, hyperthreaded, VT, 64GB RAM, gobs of disk, etc... all to run 8 virtual machines (at the moment). My problem is each of the guests - Windows, Ubuntu, FreeBSD, starts out working fine. After several minutes, the guest becomes unresponsive for a moment, and comes back. As time goes on, this happens more, and more frequently, and longer and longer until after about a day the guest is completely useless, and needs a shutdown, and start. The dom0 is NEVER unresponsive. I've tried to track this down through every means I have available, and have come up empty handed. Here is some of what I've done so far, please let me know what other information would help to debug this: 1. I've set guest OS processor to be default, and copied from host. Neither made any noticable difference. 2. I've switched all disks and network to be VirtIO. Big improvement over IDE, but the unresponsive problem persists. 3. Linux guests have the following in dmesg: hrtimer: interrupt took 10109276 ns [sched_delayed] sched: RT throttling activated 4. I found nothing suspicious in the dom0's dmesg. 5. Unresponsiveness does not correlate with disk usage. 6. Host uses 4 disks in software RAID-5 (yes I know I'm bad, but there are reasons) using BTRFS. 7. Guest disks are all raw. Please let me know if there is some other useful information, or if you have an idea where I should look next. Thanks!

2 2

[libvirt-users] Compile error on centos 7 : undefined reference to `virConnectNetworkEventRegisterAnyCheckACL'
by mxs kolo 27 Aug '14

27 Aug '14

Hi all Centos 7.0.1406, 3.10.0-123.el7.x86_64 Configure libvirt-1.2.7 as: ./configure --prefix=/usr --with-xen=no --with-qemu=no --with-openvz=no --with-vmware=no --with-esx=no --with-parallels=no --with-bhyve=no --with-uml=no --with-vbox=no --with-selinux=no make ... make[3]: Entering directory `/usr/local/src/libvirt-1.2.7/daemon' CC libvirtd-remote.o CCLD libvirtd libvirtd-remote.o: In function `remoteRelayDomainEventCheckACL': /usr/local/src/libvirt-1.2.7/daemon/remote.c:157: undefined reference to `virConnectDomainEventRegisterAnyCheckACL' libvirtd-remote.o: In function `remoteRelayNetworkEventCheckACL': /usr/local/src/libvirt-1.2.7/daemon/remote.c:184: undefined reference to `virConnectNetworkEventRegisterAnyCheckACL' collect2: error: ld returned 1 exit status make[3]: *** [libvirtd] Error 1 make[3]: Leaving directory `/usr/local/src/libvirt-1.2.7/daemon' make[2]: *** [all] Error 2 make[2]: Leaving directory `/usr/local/src/libvirt-1.2.7/daemon' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/usr/local/src/libvirt-1.2.7' make: *** [all] Error 2 I trying build libvirt-remote.0 manualy, by added file ../src/.libs/libvirt_driver_access.a where function virConnectDomainEventRegisterAnyCheckACL defined. But as result libvirtd cant' start with multiple errrors: Aug 25 19:00:54 arvis systemd: Starting LSB: daemon for libvirt virtualization API... Aug 25 19:00:54 journal: libvirt version: 1.2.7 Aug 25 19:00:54 journal: failed to load module /usr/lib/libvirt/connection-driver/libvirt_driver_network.so /usr/lib/libvirt/connection-driver/libvirt _driver_network.so: undefined symbol: virConnectListAllNetworksCheckACL Aug 25 19:00:54 journal: failed to load module /usr/lib/libvirt/connection-driver/libvirt_driver_storage.so /usr/lib/libvirt/connection-driver/libvirt _driver_storage.so: undefined symbol: virConnectListAllStoragePoolsCheckACL Aug 25 19:00:54 journal: failed to load module /usr/lib/libvirt/connection-driver/libvirt_driver_nodedev.so /usr/lib/libvirt/connection-driver/libvirt _driver_nodedev.so: undefined symbol: virConnectListAllNodeDevicesCheckACL Aug 25 19:00:54 journal: failed to load module /usr/lib/libvirt/connection-driver/libvirt_driver_secret.so /usr/lib/libvirt/connection-driver/libvirt_ driver_secret.so: undefined symbol: virConnectListAllSecretsCheckACL Aug 25 19:00:54 journal: failed to load module /usr/lib/libvirt/connection-driver/libvirt_driver_nwfilter.so /usr/lib/libvirt/connection-driver/libvir t_driver_nwfilter.so: undefined symbol: virConnectNumOfNWFiltersCheckACL Aug 25 19:00:54 journal: failed to load module /usr/lib/libvirt/connection-driver/libvirt_driver_interface.so /usr/lib/libvirt/connection-driver/libvi rt_driver_interface.so: undefined symbol: virConnectNumOfDefinedInterfacesCheckACL Aug 25 19:00:54 journal: failed to load module /usr/lib/libvirt/connection-driver/libvirt_driver_qemu.so /usr/lib/libvirt/connection-driver/libvirt_dr iver_qemu.so: undefined symbol: virConnectListDomainsCheckACL Aug 25 19:00:54 journal: failed to load module /usr/lib/libvirt/connection-driver/libvirt_driver_lxc.so /usr/lib/libvirt/connection-driver/libvirt_dri ver_lxc.so: undefined symbol: virConnectListDomainsCheckACL Aug 25 19:00:54 journal: failed to load module /usr/lib/libvirt/connection-driver/libvirt_driver_uml.so /usr/lib/libvirt/connection-driver/libvirt_dri ver_uml.so: undefined symbol: virConnectListDomainsCheckACL b.r. Maxim Kozin

2 6

[libvirt-users] Building virt-top to SLES 11 SP3
by cjlove＠san.rr.com 25 Aug '14

25 Aug '14

Hi all, I was curious whether anyone has successfully built virt-top (version 1.0.8) on SUSE SLES11 SP3. SUSE's ocaml support is rather limited, so I've built ocaml 4.00.1 from source and installed the necessary dependencies using Opam. I'm still running into a compile error on virt_top.ml, line 697: Error: This expression has type int -> D.typed_param list array but an expression was expected of type (string * D.typed_param_value) list array As a frame of reference I was able to build virt-top 1.0.8 from source on a Fedora Core 20 box with the necessary ocaml -devel RPMs installed (also ocaml 4.00.1), so I don't know if I'm running into other dependency issues or something else. The ocaml-libvirt package version 0.6.1.2 compiles successfully, which suggests that SLES 11 SP3 has a sufficient version of libvirt. Thanks in advance, Chris

3 3

[libvirt-users] Remote connection. Get URI
by Pavel Khrulev 25 Aug '14

25 Aug '14

Hello. I have an app (virt-manager) for remote connection to server where libvirtd is running. When I make a new connection with uri (qemu+tcp://paul@172.16.111.20/system) I want to know username (paul here) on the side of server. But an object virConnectPtr has URI (qemu:///system ) on the remote server. Why and how I can know username on the server's side?

2 1

[libvirt-users] NetworkManager and bridged networking
by Gene Czarcinski 23 Aug '14

23 Aug '14

Adam Williamson (adamw) has done a fine job at determining how to get bridged networking working with NetworkManager: https://www.happyassassin.net/2014/07/23/bridged-networking-for-libvirt-wit… According to Adam, he did the work on Fedora 21 whereas I did it on Fedora 20. the undesirable features (problems) I found may be corrected on F21 but I believe that they have not been fixed. While Adam's procedure for creating a bridge definition for NetworkManager using the Gnome or nm-connection-editor are basically correct. However, The virt-manager & libvirt do not "like" the default names. [*not a show stopper but it should be fixed*] Instead, the the bridge definition, use "br0" for the connection name and the device name. For the "slave" definition (the real physical NIC), use the <physical_device_name> such as p33p1 or p4p1 or em0 or whatever. I then rebooted the host to make sure everything was clean and the netowrk came up fine on br0. Then, using virt-manager, I took an existing VM, deleted the virtual NIC, added a new virtual NIC using br0. After booting up the VM, you need to edit its network configuration to delete the existing definition and add a new one pointing to the new virtual NIC. This works!!! And on Fedora 20 ... oops, not quite Fedora 20. This was done on a Fedora 20 host which as the updates from fedora-virt-preview applied. I still have a system without the preview updates so I will go back and see if it works or it needs the preview updates. Gene

1 0

[libvirt-users] vlan tags filtered in linux bridge interfaces?
by Dennis Jacobfeuerborn 22 Aug '14

22 Aug '14

Hi, I've set up a virtual machine and plugged it into a bridge on the host. When i configure an ip for eth0 in the guest i can ping other machines fine but when I add an interface eth0.10 in the guest and add an ip pings using this interface do not work. When using tcpdump in vnet0 on the host I can see the icmp packets when I ping using eth0 in the guest but not when I use the eth0.10 interface. Do tagged packets get filtered by libvirt and can this be disabled so that packets just get forwarded without such filtering? Regards, Dennis

2 1

[libvirt-users] How to enable SELinux driver for libvirt?
by Qiang Guan 22 Aug '14

22 Aug '14

Hi guys, It seems libvirt disable SELinux driver by default. How to enable SELinux driver for libvirt? -- ------------ Jackie Best Regards

2 3

[libvirt-users] virt-sandbox-service
by Gene Czarcinski 21 Aug '14

21 Aug '14

Is there any documentation/tutorial that explains how to use virt-sandbox-service? After looking at some writeups about virt-sandbox-service, this looks like a good tool for something I need to do. But, following the "examples" I cannot get anything to work correctly. With a simple xfce install with httpd, lighttpd, and libvirt-sandbox installed, I tried: 1. virt-sandbox-service create -C -u lighttpd.service webserver1 which produced a python traceback?? 2. virt-sandbox-service create -C -u httpd.service webserver2 Which appeared to work correctly but when I tried: virt-sandbox-service start webserver2 I got an error message saying that "start" was not valid??? 3. On another system with lighttpd and httpd already installed and able to run, I updated libvirt and friends to the preview version. My virtual systems seem to run OK. I then installed libvirt-sandbox and tried: virt-sandbox-service create -C -u httpd.service tstserver which got some error messages about sockets. I assume that this stuff has to work for someone. HELP! Gene

1 1

[libvirt-users] virsh snapshot
by Adam King 21 Aug '14

21 Aug '14

Hi, I had a 'domain' called APP03. I was performing auto snapshots each night using /usr/bin/virsh snapshot-create APP03 . I then wanted the domain to be called APP01 for clarity's sake, so I did virsh dumpxml APP03 > APP03.xml, edited the name to APP01 and changed the ID. I then did virsh define APP03.xml to get APP01. I have now realised that the qcow2 images are huge (1.1TB image but 80G inside the virtual machine), I am assuming this is the snapshots tagged onto the virtual hard disk? How do I reduce the size of those back to something realistic? I've removed old snapshots from APP03 using virsh snapshot-delete APP03 '<snapshotidnumber> Hope this makes sense

2 5

[libvirt-users] Reboot, halt, poweroff called inside an lxc domain
by Michal Witanowski 21 Aug '14

21 Aug '14

Hi all, Is there any possibility (using libvirt C API or whatever else) to "propagate" shutdown events (reboot, halt, power off) to a host? The API seems to lack this functionality - there exists function virConnectDomainEventRegisterAny, but the virDomainEventStoppedDetailType enum does not distinguish between these 3 different situations. Moreover, while rebooting a machine, an additional argument can be passed to reboot() syscall, which will be used by bootloader during the startup. Ideally, when calling e.g. "reboot recovery" inside an lxc container I'd like to be informed (inside host) that "reboot" was called with "recovery" as additional argument, so I can take appropriate actions. Thanks in advance. Best regards, Michal

2 1