November 2015 - Users - libvirt List Archives

[libvirt-users] "Allocate entire disk now" with an existing virtual disk
by Boylan, Ross 18 Nov '15

18 Nov '15

Have I just overwritten some of my virtual drive, and if so, is there a way to recover? On a linux host I use LVM logical volumes as the backing for my virtual disks. Using the Virtual Machine Manager 0.9.1 GUI I did this: 1. select source machine, go into details, select virtual disk and removed the hardware. 2. select destination machine, go into details, add storage. 3. click "select managed or other existing storage" 4. click the browser button under it, and select the volume detached from the source machine 5. set device type to virtio. 6. click finish. Both VM's were off, and the source VM used virtio also. KVM is the hypervisor. There were actually 2 drives, and so I repeated the steps above twice, except that the second time I clicked on "Allocate entire disk now", which deselected it (I hope) before step 6. The destination machine already had disks. So, did I just overwrite one of the disks? If so, is there a way to recover? The reason I suspect a problem is that, although the partition tables appear intact on both virtual drives, the filesystem does not. Within the virtual machine 1 partition from each drive is combined in a volume group. The main logical volume group on that fails to mount: root@jessie01:~# mount -o ro /dev/markov02/root /mnt/markov02 mount: wrong fs type, bad option, bad superblock on /dev/mapper/markov02-root, missing codepage or helper program, or other error In some cases useful info is found in syslog - try dmesg | tail or so. and e2fsck says root@jessie01:~# e2fsck /dev/markov02/root e2fsck 1.42.12 (29-Aug-2014) /dev/markov02/root: recovering journal Superblock needs_recovery flag is clear, but journal has data. It's possible this has nothing to do with the option to allocate the entire disk, since the source VM was having problems anyway (that's why I moved the disks). However, I'm pretty sure I was able to access the filesystem both in the source VM and in an earlier round when I move the disks to the destination. Thanks. Ross Boylan

1 0

[libvirt-users] virDomainCoreDumpWithFormat files created as root
by NoxDaFox 18 Nov '15

18 Nov '15

Greetings, I am dumping a guest VM memory for inspection using the command "virDomainCoreDumpWithFormat" and the created files appear to belong to root (both user and group). I have searched around but didn't find any answer. Is there a way to instruct QEMU to create those files under a different user? Thank you.

5 7

[libvirt-users] Add host
by Sakhi Hadebe 18 Nov '15

18 Nov '15

Hi, I have two host machines both installed with libvirt and virt-manager, which provides me with GUI. I have virtual machines running in one of the host. Now I want to migrate some of the virtual machines to the other machines. I am trying to migrate using the Virtual Machine Manager. When I select migrate, it pops up a window to choose the new host. The dropdown menu to choose the new host for the virtual machine is empty. How do you add the new host to the other, so they can know each other? Does it require a clustering software to achieve that? I need some assistance in that regard -- Regards, Sakhi Hadebe Engineer: South African National Research Network (SANReN)Competency Area, Meraka, CSIR Tel: +27 12 841 2308 <+27128414213> Fax: +27 12 841 4223 <+27128414223> Cell: +27 71 331 9622 <+27823034657> Email: sakhi(a)sanren.ac.za <shadebe(a)csir.co.za>

2 2

[libvirt-users] Telling virt-manager to re-read ~/.gconf in RHEL 6
by Digimer 16 Nov '15

16 Nov '15

Hi all, I've written a tool that configured Virtual Machine Manager, but I've got a condition that I am now sure how to handle... If the user isn't logged in and I configure their ~/.gconf/apps/virt-manager/*, and then they log in, everything works great. However, if they've already logged in and there was no prior virt-manager configuration, the custom config is overwritten and lost when the user logs out. Similarly, if VMM is closed, I write the config and then they open VMM (again, having logged into gnome where there was no ~/.gconf/apps/virt-manager), it acts like there is no configuration (it tries to auto-connect to localhost and doesn't show the remote machines). I am guessing this behaviour is a because VMM doesn't check it's configuration file after logging into gnome (either at all, or at least when there was no config at gnome login). Is there a way I can tell virt-manager to re-check the config? Thanks! -- Digimer Papers and Projects: https://alteeve.ca/w/ What if the cure for cancer is trapped in the mind of a person without access to education?

3 2

[libvirt-users] Qemu-ga on Windows 10
by Dominique Ramaekers 14 Nov '15

14 Nov '15

Hi, I have a windows 10 guest with qemu-ga-x64.msi installed from the fedora virtio iso. qemu-ga version is 0.12.1 host qemu version is 2.1.0 (I know its older but in this version the active block commit works perfectly...) libvirt version 1.2.12 I have these lines in my xml: <channel type='unix'> <source mode='bind' path='/var/lib/libvirt/qemu/channel/target/PCVIRTmbe.org.qemu.guest_agent.0'/> <target type='virtio' name='org.qemu.guest_agent.0' state='disconnected'/> <alias name='channel0'/> <address type='virtio-serial' controller='0' bus='0' port='1'/> </channel> /var/lib/libvirt/qemu/channel/target/PCVIRTmbe.org.qemu.guest_agent.0 exists on the host and the rights seems to be all right. On the guest, the folder C:\ProgramData\qemu-ga exists Still, I can't start the qemu-ga service. I get this error: Can't start QEMU GuestAgent-service on the local computer Error 1053: The service didn't answer correctly on the start command (or something like this, the error message in in Dutch) If I start the qemu-ga from the cmd I get this error: error opening path error opening channel failed to create guest agent channel failed to initialize guest agent channel Does anybody got an idea how to resolve this problem? Thanks in advance. Dominique.

1 1

[libvirt-users] How to fix an incorrect storage pool?
by Phill Edwards 14 Nov '15

14 Nov '15

I've created my storage pool incorrectly. I'm using LVM and I have a volume group called vms-lvm. When I look at it in virt-manager I see that the volumes it contains are home, root and swap, so when I created the storage group in virt-manager I must have specified something incorrectly. Unfortunately I can't find a way to correct this. If I try to destroy (stop) the storage group I get an error: # virsh pool-destroy vms-lvm error: Failed to destroy pool vms-lvm error: internal error: Child process (/usr/sbin/vgchange -aln centos_hyv2) unexpected exit status 5: Logical volume centos_hyv2/swap in use. Can't deactivate volume group "centos_hyv2" with 3 open logical volume(s) Same if I try to undefine it: # virsh pool-undefine vms-lvm error: Failed to undefine pool vms-lvm error: Requested operation is not valid: storage pool 'vms-lvm' is still active Can anyone please advise how I get out of this mess? There must surely be a way of undefining this?

2 3

[libvirt-users] a VM does not get dhcp
by lejeczek 12 Nov '15

12 Nov '15

hi everybody I'm looking at my setup and got stuck I have a box with two NICs on the same subnet, and another box similar, also two NICs on one subnet, all four NICs are on the same subnet. Now, that second box has also a virtual NIC (libvirtd's bridge route mode) and VMs guests are using it, traffic to that virtual net is routed via 1st real NIC. This second box dhcrelays to the first box(dhcpd). I see box-dhcrelay forwards to box-dhcpd, I see box-dhcpd receives and offers a lease but that VM guest does not get it. I have policy routing manually set in place so both boxes can ping each other all NICs. (including virtual NIC on the box-dhcrelay) Moreover, that VM guest can ping both boxes' all NICs when its IP address is set to manual. It's RHEL7 and I'm only trying IPv4. I'm hoping some can rule out (or suggest what might be broken in) libvirt/policy based routing. I see those offers box-dhcpd makes are exactly for the subnet of box-dhcrelay's virtual NIC/subnet. It's a pickle. An expert's thought would be great to hear.

2 2

[libvirt-users] xen+ssh with openauth
by Shahar Havivi 12 Nov '15

12 Nov '15

Hi, I am trying to connect to remote libvirt via script, and I get password prompt from the shell. ============================================================================ uri='xen+ssh://root@myserver/?no_verify=1' def req(credentials, user_data): print('in req...\n') for cred in credentials: if cred[0] == libvirt.VIR_CRED_AUTHNAME: cred[4] = username elif cred[0] == libvirt.VIR_CRED_PASSPHRASE: cred[4] = passwd return 0 auth = [[libvirt.VIR_CRED_AUTHNAME, libvirt.VIR_CRED_PASSPHRASE], req, None] con = libvirt.openAuth(uri, auth, 0) ============================================================================ The script never enter req() method but prompts for password, after I enter the password I am able to connect to the remote machine. I did try to put all the flags ie: auth = [[1,2,3,4,5,6,7,8], req, None] with no successes as well... The server is RHEL 5.11 with libvirt-0.8.2-29.el5_9.1 Thank you, Shahar.

2 4

[libvirt-users] ERROR_NO_SYSTEM_RESOURCES on virtio port within the Windows Guest
by Jean-Pierre Ribeauville 09 Nov '15

09 Nov '15

Hi, >From a Windows Guest , I try to read data sent by the Host , via a virtio port . ( I.e. channel device type UNIX on host side) If no data sent by Host , the blocking ReadFile() invoked within the Guest exits immediately with following error code : ERROR_NO_SYSTEM_RESOURCES 1450 (0x5AA) Insufficient system resources exist to complete the requested service. Is it the waited behavior or did I misunderstood something ? If datas are sent , I got them correctly. Thx for help. J.P. Ribeauville P: +33.(0).1.47.17.20.49 . Puteaux 3 Etage 5 Bureau 4 jpribeauville(a)axway.com<mailto:jpribeauville@axway.com> http://www.axway.com<http://www.axway.com/> P Pensez à l'environnement avant d'imprimer.

1 0

[libvirt-users] availability of target type state within a dumpxml result
by Jean-Pierre Ribeauville 09 Nov '15

09 Nov '15

Hi, I'm trying to find out what is the minimum release I need to use to have this field 'state' available ? <channel type='unix'> <source mode='bind' path='//var/lib/libvirt/qemu/dummy_agent'/> <target type='virtio' name=dummy' state='connected'/> <alias name='channel0'/> <address type='virtio-serial' controller='0' bus='0' port='1'/> </channel> By using a virsh 0.10.2 , it seems tooo old . Thanks for help. J.P. Ribeauville P: +33.(0).1.47.17.20.49 . Puteaux 3 Etage 5 Bureau 4 jpribeauville(a)axway.com<mailto:jpribeauville@axway.com> http://www.axway.com<http://www.axway.com/> P Pensez à l'environnement avant d'imprimer.

2 2

[libvirt-users] How could I do the guest mount operation from the host?
by 李笳平 09 Nov '15

09 Nov '15

Hi all, When I attach a disk like this: <disk type='file' device='disk'> <driver name='qemu' type='qcow2' cache='none'/> <source file='/kvmdata/disk/dk389_581.img'/> <backingStore/> <target dev='vda' bus='virtio'/> <alias name='virtio-disk0'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x07' function='0x0'/> </disk> And then I use “fdisk -l” on the guest, can find the disk info, if I want to mount a disk, I should log on the guest and use the command just like: mount /dev/vda1 /mnt Does Libvirt has some API which can send the guest operation(like mount/umount) from host to guest? Thus I can do the operation on host and need not to log on guest to do the operation. Thanks! Best Regards From Jiaping LI 保密：本函仅供收件人使用，如阁下并非抬头标明的收件人，请您即刻删除本函，勿以任何方式使用及传播，并请您能将此误发情形通知发件人，谢谢！

2 7

[libvirt-users] ./build-aux/bracket-spacing.pl missing from tarball
by George Trakatelis 08 Nov '15

08 Nov '15

Hi, When trying to build libvirt from recent tarballs, I noticed that ./build-aux/bracket-spacing.pl is missing. That makes 'make syntax-check' fail. Of course, one would argue that syntax-check is only for developers using git. On the other hand, the addition of the file would enable building libvirt from the tarball just by executing ./autobuild.sh I'm wondering whether that has been done on purpose or just slipped. It's always nice to have a package build out-of-the-box. George

1 0

[libvirt-users] VGA passtrough
by Dominique Ramaekers 07 Nov '15

07 Nov '15

Hi, I've been reading up on how to create guests which will have a gpu. I found a technologie of nvidia called Grid K1 or Grid K2. It's a graphical PCIe card which can make virtual gpu's available. Only I think nvidia doesn't support qemu and it looks expensive (didn't ask for a quote yet) because these cards are created for 10 tot 20 users. I only need to support about 6 tops. (and for now just one or two to start) On libvirt-qemu we can also use the VGA-passtrough support. It seems easy enough. I have the idea that when using the VGA passtrough, the screen image is send to the hdmi/dvi port on the grapfics card. It seems logical. Does this also mean I can't use the gpu power on a remote screen? I think when using spice, this is surely true but what if I use vnc? If anyone can give me some tips on this, I'll appreciate it... Greetings, Dominique.

1 0

[libvirt-users] live vnc port change
by Fırat KÜÇÜK 06 Nov '15

06 Nov '15

Hello, I want to live migrate a vm from one host to another host. Bu I got this error: error: internal error: process exited while connecting to monitor: 2015-11-06T07:53:59.006762Z qemu-system-x86_64: Failed to start VNC server on `0.0.0.0:3': Failed to bind socket: Address already in use The same VNC port is in use at the destination host. Is it possible to change port address without rebooting the server?

2 2

[libvirt-users] how to make dhcrelay service use libvirt's interfaces?
by lejeczek 05 Nov '15

05 Nov '15

hi everybody in my libvirtd I use bridge with route forward mode, Problem for me is dhcrelay.service on host which is starting before libvirtd thus not listening on libvirt interface. OS is RHEL7. I wonder if you know how to fix it, besides restarting dhcrelay that is. many thanks

2 1

[libvirt-users] Libvirt management daemon issue
by Bilal Arif 05 Nov '15

05 Nov '15

robo@robo:~$ sudo /etc/init.d/libvirt-bin start [sudo] password for robo: * Starting libvirt management daemon libvirtd/usr/sbin/libvirtd: /usr/local/lib/libvirt.so.0: version `LIBVIRT_PRIVATE_1.2.2' not found (required by /usr/sbin/libvirtd) Giving up waiting for /var/run/libvirt/libvirt-sock. After installing libvirt from source a lot of Ubuntu user face this issue, as i searched on internet about this, no solution works. this is mostly in case of ubuntu users as mentioned in many posted threads. so please help us as soon as possible. (this is only in case of installation libvirt from source)

2 1

[libvirt-users] Retrieving device state field value ...
by Jean-Pierre Ribeauville 05 Nov '15

05 Nov '15

Hi, Is there "faster" way than calling virDomainGetXMLDesc() and parsing the result to retrieve the state field ( connected or disconnected) of a channel device ? Thanks. J.P. Ribeauville P: +33.(0).1.47.17.20.49 . Puteaux 3 Etage 5 Bureau 4 jpribeauville(a)axway.com<mailto:jpribeauville@axway.com> http://www.axway.com<http://www.axway.com/> P Pensez à l'environnement avant d'imprimer.

1 0

[libvirt-users] Is it possible to retrieve Non blocking socket writeable status of a unix channel device ?
by Jean-Pierre Ribeauville 05 Nov '15

05 Nov '15

Hi, As using a blocking socket for a virtio device channel leads to be blocked if nobody has opened the corresponding /dev/virtio-ports within the Guest , I tried to use a non-blocking socket and to check whether the socket is writeable via FD_ISSET(). I'm wondering why a virtio device channel is always seen as writeable by the Host even if nobody has opened the corresponding /dev/virtio-ports within the Guest ? res= connect(sock, (struct sockaddr *) &server, sizeof(struct sockaddr_un)); if (res < 0) { perror("connecting stream socket"); close(sock); return(1); } else tv.tv_sec = 15; tv.tv_usec = 0; FD_ZERO(&myset); FD_SET(sock, &myset); if (select(sock+1, NULL, &myset, NULL, &tv) > 0) { len = sizeof(int); getsockopt(sock, SOL_SOCKET, SO_ERROR, (void*)(&result), &len); if (result) { fprintf(stderr, "Error in connection() %d - %s\n", result, strerror(result)); return(1); } else { if(FD_ISSET(sock,&myset)) fprintf(stderr, "connection()is O.K. \n"); } } else { fprintf(stderr, "Timeout or error() %d - %s\n", result, strerror(result)); return(1); } Did I misunderstood something ? Thanks for help. J.P. Ribeauville P: +33.(0).1.47.17.20.49 . Puteaux 3 Etage 5 Bureau 4 jpribeauville(a)axway.com<mailto:jpribeauville@axway.com> http://www.axway.com<http://www.axway.com/> P Pensez à l'environnement avant d'imprimer.

1 0

[libvirt-users] Retrieving datas sent by host within the Guest
by Jean-Pierre Ribeauville 04 Nov '15

04 Nov '15

Hi, I' trying to read datas sent by my KVM host within the Guest. Host send these datas trough a channel UNIX type , by using AF_UNIX socket stuff. If i try to retrive the datas by using such type of thing in the Guest : #define NAME "/dev/virtio-ports/agent" sock = socket(AF_UNIX, SOCK_STREAM, 0); 12 if (sock < 0) { 13 perror("opening stream socket"); 14 exit(1); 15 } 16 server.sun_family = AF_UNIX; 17 strcpy(server.sun_path, NAME); 18 if (bind(sock, (struct sockaddr *) &server, sizeof(struct sockaddr_un))) { 19 perror("binding stream socket"); 20 exit(1); 21 } I got ""binding stream socket Address already in use" when trying to bind the socket . But If I use such kind of thing : fd = open("/dev/virtio-ports/agent", O_RDWR | O_NONBLOCK); ssize_t size = read(fd, &rcv_buffer, 100); if(size >0) printf("Read length(%d) %s\n", size, rcv_buffer); It works fine. Is there any chance to be able to use socket stuff on Guest side or is there anything I misunderstand? Thanks for help. J.P. J.P. Ribeauville P: +33.(0).1.47.17.20.49 . Puteaux 3 Etage 5 Bureau 4 jpribeauville(a)axway.com<mailto:jpribeauville@axway.com> http://www.axway.com<http://www.axway.com/> P Pensez à l'environnement avant d'imprimer.

2 2

[libvirt-users] xen/libvirt freeze while attching network-device to vm´s
by max ustermann 04 Nov '15

04 Nov '15

Hi, first sorry for the cross-posting, but i don ́t known exactly if my problem belongs to xen or libvirt. I have an VM-Host with Xen 4.6.0 und libvirt 1.2.19. The Job of the Host is to resume in parallel 20 VM ́s (HVM WinXp) attach a block-device to each VM , wait 60 Seconds, kill the VM and start from beginning. For that, a python script create for each vm a thread and each thread get a connection to the Libvirt-Daemon via "libvirt.open(None)" and do the resume, attach and so on in a loop. This Setup works fine, it do several thousand cycle without any problems. If i complement this setup, so that each vm get , additonal to the blockdevice an network-interface after resume, via "attachDevice" call from the python-script, i observe that libvirtd/xen freeze after 20 to 100 cycle. Their are no error-messages, no crash, noting in "dmesg" or "xl dmesg". When i attached to the libvirtd via gdb and do a "info thread", i get the following: Id Target Id Frame 21 Thread 0x7f792a92c700 (LWP 1312) "libvirtd" __lll_lock_wait () at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135 20 Thread 0x7f792a12b700 (LWP 1313) "libvirtd" __lll_lock_wait () at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135 19 Thread 0x7f792992a700 (LWP 1314) "libvirtd" __lll_lock_wait () at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135 18 Thread 0x7f7929129700 (LWP 1315) "libvirtd" __lll_lock_wait () at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135 17 Thread 0x7f7928928700 (LWP 1316) "libvirtd" __lll_lock_wait () at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135 16 Thread 0x7f791bfff700 (LWP 1317) "libvirtd" __lll_lock_wait () at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135 15 Thread 0x7f791b7fe700 (LWP 1318) "libvirtd" __lll_lock_wait () at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135 14 Thread 0x7f791affd700 (LWP 1319) "libvirtd" __lll_lock_wait () at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135 13 Thread 0x7f791a7fc700 (LWP 1320) "libvirtd" __lll_lock_wait () at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135 12 Thread 0x7f7919ffb700 (LWP 1321) "libvirtd" __lll_lock_wait () at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135 11 Thread 0x7f78fd6ff700 (LWP 1334) "libvirtd" pthread_cond_wait@@GLIBC_2.3.2 () at ../nptl/sysdeps/unix/sysv/linux/x86_64/pthread_cond_wait.S:185 10 Thread 0x7f78fcefe700 (LWP 1335) "libvirtd" pthread_cond_wait@@GLIBC_2.3.2 () at ../nptl/sysdeps/unix/sysv/linux/x86_64/pthread_cond_wait.S:185 9 Thread 0x7f78f3fff700 (LWP 1336) "libvirtd" pthread_cond_wait@@GLIBC_2.3.2 () at ../nptl/sysdeps/unix/sysv/linux/x86_64/pthread_cond_wait.S:185 8 Thread 0x7f78ebfff700 (LWP 1337) "libvirtd" pthread_cond_wait@@GLIBC_2.3.2 () at ../nptl/sysdeps/unix/sysv/linux/x86_64/pthread_cond_wait.S:185 7 Thread 0x7f78f37fe700 (LWP 1338) "libvirtd" pthread_cond_wait@@GLIBC_2.3.2 () at ../nptl/sysdeps/unix/sysv/linux/x86_64/pthread_cond_wait.S:185 6 Thread 0x7f793bf94700 (LWP 1494) "libvirtd" 0x00007f79375c87fd in read () at ../sysdeps/unix/syscall-template.S:81 5 Thread 0x7f78f241d700 (LWP 17709) "libvirtd" __lll_lock_wait () at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135 4 Thread 0x7f78f1c1c700 (LWP 17717) "libvirtd" __lll_lock_wait () at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135 3 Thread 0x7f78f141b700 (LWP 17722) "libvirtd" __lll_lock_wait () at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135 2 Thread 0x7f78f0c1a700 (LWP 17751) "libvirtd" __lll_lock_wait () at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135 * 1 Thread 0x7f793c0428c0 (LWP 1269) "libvirtd" __lll_lock_wait () at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135 i start to examine at thread 1 and get the fowlling bt: #0 __lll_lock_wait () at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135 #1 0x00007f79375c411b in _L_lock_812 () from /lib64/libpthread.so.0 #2 0x00007f79375c3fe8 in __GI___pthread_mutex_lock (mutex=0x7f78d80edd30) at ../nptl/pthread_mutex_lock.c:79 #3 0x00007f793b09cd0c in virMutexLock (m=0x7f78d80edd30) at util/virthread.c:89 #4 0x00007f793b07d831 in virObjectLock (anyobj=0x7f78d80edd20) at util/virobject.c:323 #5 0x00007f793b0b71d7 in virDomainObjListFindByID (doms=0x7f78d80edd20, id=1) at conf/domain_conf.c:1237 #6 0x00007f78fe3f7eb3 in libxlDomainEventHandler (data=0x7f78d81ce940, event=0x7f793c30c2e0) at libxl/libxl_domain.c:534 #7 0x00007f793806f7c0 in egc_run_callbacks (egc=egc@entry=0x7ffdb79c7490) at libxl_event.c:1405 #8 0x00007f793806fb9d in libxl__egc_cleanup (egc=0x7ffdb79c7490) at libxl_event.c:1436 #9 libxl_osevent_occurred_fd (ctx=<optimized out>, for_libxl=0x7f78d80ec220, fd=27, events_ign=<optimized out>, revents_ign=<optimized out>) at libxl_event.c:1339 #10 0x00007f78fe3f9ac5 in libxlFDEventCallback (watch=11, fd=27, vir_events=1, fd_info=0x7f78d811c430) at libxl/libxl_driver.c:134 #11 0x00007f793b04a9c4 in virEventPollDispatchHandles (nfds=31, fds=0x7f793c30b510) at util/vireventpoll.c:509 #12 0x00007f793b04b224 in virEventPollRunOnce () at util/vireventpoll.c:658 #13 0x00007f793b0490cd in virEventRunDefaultImpl () at util/virevent.c:308 #14 0x00007f793b1f5ab6 in virNetDaemonRun (dmn=0x7f793c2e1b80) at rpc/virnetdaemon.c:701 #15 0x00007f793be0bc27 in main (argc=1, argv=0x7ffdb79c7d78) at libvirtd.c:1577 i select frame 3 and look who holds the lock: (gdb) frame 3 #3 0x00007f793b09cd0c in virMutexLock (m=0x7f78d80edd30) at util/virthread.c:89 89 pthread_mutex_lock(&m->lock); (gdb) p *m $5 = {lock = {__data = {__lock = 2, __count = 0, __owner = 1312, __nusers = 1, __kind = 512, __spins = 0, __elision = 0, __list = {__prev = 0x0, __next = 0x0}}, __size = "\002\000\000\000\000\000\000\000 \005\000\000\001\000\000\000\000\002", '\000' <repeats 21 times>, __align = 2}} i switch to frame 21 (LWP1312) i taek a look to the bt: (gdb) bt #0 __lll_lock_wait () at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135 #1 0x00007f79375c411b in _L_lock_812 () from /lib64/libpthread.so.0 #2 0x00007f79375c3fe8 in __GI___pthread_mutex_lock (mutex=0x7f79140099d0) at ../nptl/pthread_mutex_lock.c:79 #3 0x00007f793b09cd0c in virMutexLock (m=0x7f79140099d0) at util/virthread.c:89 #4 0x00007f793b07d831 in virObjectLock (anyobj=0x7f79140099c0) at util/virobject.c:323 #5 0x00007f793b0b717f in virDomainObjListSearchID (payload=0x7f79140099c0, name=0x7f791400f7c0, data=0x7f792a92b464) at conf/domain_conf.c:1225 #6 0x00007f793b055f1e in virHashSearch (ctable=0x7f78d81ccc40, iter=0x7f793b0b7148 <virDomainObjListSearchID>, data=0x7f792a92b464) at util/virhash.c:724 #7 0x00007f793b0b71f2 in virDomainObjListFindByID (doms=0x7f78d80edd20, id=2) at conf/domain_conf.c:1238 #8 0x00007f78fe3f7eb3 in libxlDomainEventHandler (data=0x7f78d81ce940, event=0x7f791c005610) at libxl/libxl_domain.c:534 #9 0x00007f793806f7c0 in egc_run_callbacks (egc=egc@entry=0x7f792a92b580) at libxl_event.c:1405 #10 0x00007f7938070f8b in libxl__egc_cleanup (egc=0x7f792a92b580) at libxl_event.c:1436 #11 libxl__ao_inprogress (ao=ao@entry=0x7f791c00c6e0, file=file@entry=0x7f793808d512 "libxl.c", line=line@entry=4226, func=func@entry=0x7f7938090150 <__func__.20288> "libxl_device_nic_add") at libxl_event.c:2014 #12 0x00007f7938048fa6 in libxl_device_nic_add (ctx=0x7f78d8112830, domid=19, nic=0x7f792a92b6b0, ao_how=<optimized out>) at libxl.c:4226 #13 0x00007f78fe402baa in libxlDomainAttachNetDevice (driver=0x7f78d81ce940, vm=0x7f79140099c0, net=0x7f791c003610) at libxl/libxl_driver.c:3199 #14 0x00007f78fe402d12 in libxlDomainAttachDeviceLive (driver=0x7f78d81ce940, vm=0x7f79140099c0, dev=0x7f791c001350) at libxl/libxl_driver.c:3231 #15 0x00007f78fe404135 in libxlDomainAttachDeviceFlags (dom=0x7f791c009150, xml=0x7f791c009230 "<interface type=\"bridge\">\n <source bridge=\"xenbr0\" />\n <model type=\"rtl8139\" />\n <mac address=\"00:27:0e:0c:01:09\" />\n </interface>", flags=1) at libxl/libxl_driver.c:3707 #16 0x00007f78fe404239 in libxlDomainAttachDevice (dom=0x7f791c009150, xml=0x7f791c009230 "<interface type=\"bridge\">\n <source bridge=\"xenbr0\" />\n <model type=\"rtl8139\" />\n <mac address=\"00:27:0e:0c:01:09\" />\n </interface>") at libxl/libxl_driver.c:3745 #17 0x00007f793b18422c in virDomainAttachDevice (domain=0x7f791c009150, xml=0x7f791c009230 "<interface type=\"bridge\">\n <source bridge=\"xenbr0\" />\n <model type=\"rtl8139\" />\n <mac address=\"00:27:0e:0c:01:09\" />\n </interface>") at libvirt-domain.c:8351 #18 0x00007f793be11a1f in remoteDispatchDomainAttachDevice (server=0x7f793c2e0a90, client=0x7f793c2f34c0, msg=0x7f793c2feca0, rerr=0x7f792a92bab0, args=0x7f791c009190) at remote_dispatch.h:2850 #19 0x00007f793be11923 in remoteDispatchDomainAttachDeviceHelper (server=0x7f793c2e0a90, client=0x7f793c2f34c0, msg=0x7f793c2feca0, rerr=0x7f792a92bab0, args=0x7f791c009190, ret=0x7f791c009130) at remote_dispatch.h:2826 #20 0x00007f793b1fccfc in virNetServerProgramDispatchCall (prog=0x7f793c2eefd0, server=0x7f793c2e0a90, client=0x7f793c2f34c0, msg=0x7f793c2feca0) at rpc/virnetserverprogram.c:437 #21 0x00007f793b1fc863 in virNetServerProgramDispatch (prog=0x7f793c2eefd0, server=0x7f793c2e0a90, client=0x7f793c2f34c0, msg=0x7f793c2feca0) at rpc/virnetserverprogram.c:307 #22 0x00007f793b1f5e6d in virNetServerProcessMsg (srv=0x7f793c2e0a90, client=0x7f793c2f34c0, prog=0x7f793c2eefd0, msg=0x7f793c2feca0) at rpc/virnetserver.c:135 #23 0x00007f793b1f5f3a in virNetServerHandleJob (jobOpaque=0x7f793c302c60, opaque=0x7f793c2e0a90) at rpc/virnetserver.c:156 #24 0x00007f793b09d9f7 in virThreadPoolWorker (opaque=0x7f793c2e0750) at util/virthreadpool.c:145 #25 0x00007f793b09d00a in virThreadHelper (data=0x7f793c2eced0) at util/virthread.c:206 #26 0x00007f79375c1ee5 in start_thread (arg=0x7f792a92c700) at pthread_create.c:309 #27 0x00007f79372f0d1d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:111 here i select frame 3 and look who holds the lock: (gdb) frame 3 #3 0x00007f793b09cd0c in virMutexLock (m=0x7f79140099d0) at util/virthread.c:89 89 pthread_mutex_lock(&m->lock); $6 = {lock = {__data = {__lock = 2, __count = 0, __owner = 1312, __nusers = 1, __kind = 512, __spins = 0, __elision = 0, __list = {__prev = 0x0, __next = 0x0}}, __size = "\002\000\000\000\000\000\000\000 \005\000\000\001\000\000\000\000\002", '\000' <repeats 21 times>, __align = 2}} so i see that thread 21 waits for a lock that holds by thread 21 from frame 4 i see, that the process want a lock for vm object: #4 0x00007f793b07d831 in virObjectLock (anyobj=0x7f79140099c0) at util/virobject.c:323 further down in the bt i see in frame 14: #14 0x00007f78fe402d12 in libxlDomainAttachDeviceLive (driver=0x7f78d81ce940, vm=0x7f79140099c0, dev=0x7f791c001350) at libxl/libxl_driver.c:3231 that vm-object 0x7f79140099c0 is the object for that also a network-device should be attached. And from that i know that there is a lock on the vm-object. so i go up in the bt an see that libxlDomainAttachDeviceLive(libvirtd) (frame14) calls libxlDomainAttachNetDevice(libvirtd) (frame13), that calls libxl_device_nic_add(libxl/xen) (frame12) from info frame for frame12 i see that he calls frame 0x7f792a92b5f0: (gdb) info f Stack level 12, frame at 0x7f792a92b670: rip = 0x7f7938048fa6 in libxl_device_nic_add (libxl.c:4226); saved rip = 0x7f78fe402baa called by frame at 0x7f792a92b760, caller of frame at 0x7f792a92b5f0 source language c. Arglist at 0x7f792a92b5e8, args: ctx=0x7f78d8112830, domid=19, nic=0x7f792a92b6b0, ao_how=<optimized out> Locals at 0x7f792a92b5e8, Previous frame's sp is 0x7f792a92b670 Saved registers: rbx at 0x7f792a92b640, rbp at 0x7f792a92b648, r12 at 0x7f792a92b650, r13 at 0x7f792a92b658, r14 at 0x7f792a92b660, rip at 0x7f792a92b668 and this is frame 11 libxl__ao_inprogress (ao=ao@entry=0x7f791c00c6e0, file=file@entry=0x7f793808d512 "libxl.c", line=line@entry=4226, func=func@entry=0x7f7938090150 <__func__.20288> "libxl_device_nic_add") at libxl_event.c:2014 so further i see that libxl__ao_inprogress calls libxl__egc_cleanup and that calls egc_run_callbacks, frame 9 info frame for frame 9 gives: (gdb) info f Stack level 9, frame at 0x7f792a92b540: rip = 0x7f793806f7c0 in egc_run_callbacks (libxl_event.c:1405); saved rip = 0x7f7938070f8b called by frame at 0x7f792a92b5f0, caller of frame at 0x7f792a92b500 source language c. Arglist at 0x7f792a92b4f8, args: egc=egc@entry=0x7f792a92b580 Locals at 0x7f792a92b4f8, Previous frame's sp is 0x7f792a92b540 Saved registers: rbx at 0x7f792a92b520, rbp at 0x7f792a92b528, r12 at 0x7f792a92b530, rip at 0x7f792a92b538 and list gives: (gdb) list 1400 1401 LIBXL_TAILQ_FOREACH_SAFE(ev, &egc->occurred_for_callback, link, ev_tmp) { 1402 LIBXL_TAILQ_REMOVE(&egc->occurred_for_callback, ev, link); 1403 LOG(DEBUG,"event %p callback type=%s", 1404 ev, libxl_event_type_to_string(ev->type)); 1405 CTX->event_hooks->event_occurs(CTX->event_hooks_user, ev); 1406 } 1407 1408 LIBXL_TAILQ_FOREACH_SAFE(aop, &egc->aops_for_callback, entry, aop_tmp) { 1409 LIBXL_TAILQ_REMOVE(&egc->aops_for_callback, aop, entry); from here i see that egc_run_callbacks is at line 1405 " CTX->event_hooks->event_occurs(CTX->event_hooks_user, ev);" , but i have no idea what he is doing there and while it "hangs" here!?! if i see it correctly, the systems freeze/hangs because it waits of the release of the lock for the vm which actuall attach a network-device, and the process of attaching the network-device didn ́t finished and hold the lock. as i say at the geginning of my mail, if i only attach a block-device to the vm`s every thing runs, and with additional attaching a network-device after 20 to 100 resume , it freeze´s. i am gratefull for any help,hints and tips for further debugging. all the best max

1 0

[libvirt-users] vm doesn't respond after live migration
by Fırat KÜÇÜK 04 Nov '15

04 Nov '15

Hi, I migrate my VM with this command: virsh migrate --live --domain my-domain --verbose --copy-storage-all --desturi qemu+ssh://root@xxx.bulutkare.com/system Migration completed successfully. But after migration VM doesn't respond. I can see the login console. I typed root but no response. What can be the problem? Regards. -- Fırat KÜÇÜK https://codv.io/

2 2

[libvirt-users] shutdown windows 7 vm although someone is logged on via RemoteDesktop
by Lentes, Bernd 03 Nov '15

03 Nov '15

Hi, how can i shutdown a windows 7 vm although someone is logged on to that vm via RemoteDesktop ? Currently libvirt waits 5 min for the vm to shutdown and then switches it off. Bernd -- Bernd Lentes Systemadministration institute of developmental genetics Gebäude 35.34 - Raum 208 HelmholtzZentrum München bernd.lentes(a)helmholtz-muenchen.de phone: +49 (0)89 3187 1241 fax: +49 (0)89 3187 2294 Wer Visionen hat soll zum Hausarzt gehen Helmut Schmidt Helmholtz Zentrum Muenchen Deutsches Forschungszentrum fuer Gesundheit und Umwelt (GmbH) Ingolstaedter Landstr. 1 85764 Neuherberg www.helmholtz-muenchen.de Aufsichtsratsvorsitzende: MinDir'in Baerbel Brumme-Bothe Geschaeftsfuehrer: Prof. Dr. Guenther Wess, Dr. Nikolaus Blum, Dr. Alfons Enhsen Registergericht: Amtsgericht Muenchen HRB 6466 USt-IdNr: DE 129521671

1 0

[libvirt-users] Creating network interfaces with specific name in LXC
by Harish Vishwanath 03 Nov '15

03 Nov '15

Hello I am trying to create interfaces within an LXC container with specific name. I am using the following XML: <interface type='network' name='blah'> <source network='dpbr_n_0'/> </interface> This doesn't seem to work; it always ends up creating interface with name as "eth0". I found some examples here: http://libvirt.org/guide/html/Application_Development_Guide-Network_Interfa… That uses interface type as "ethernet" and specifies name also, If I do that, I get back an error while starting the container as "Unsupported network type ethernet". Appreciate your help. Regards, Harish

2 4

[libvirt-users] changeBegin() on python API
by Ramon Medeiros 02 Nov '15

02 Nov '15

Hi, is changeBegin() working on python? I was trying to use the following method: if = conn.interfaceLookupByName("name") if.undefine() I saw that changeBegin() does not commit changes, but i can't use it. -- Ramon Nunes Medeiros Kimchi Developer Linux Technology Center Brazil IBM Systems & Technology Group Phone : +55 19 2132 7878 ramonn(a)br.ibm.com

3 3

[libvirt-users] virsh edit from script ??
by Jean-Pierre Ribeauville 02 Nov '15

02 Nov '15

Hi, I need to add , via script , a channel device in .xml guest config file. Is virsh edit useable from a "standard" script or is there any more suitable way to use ? Thanks. J.P. Ribeauville P: +33.(0).1.47.17.20.49 . Puteaux 3 Etage 5 Bureau 4 jpribeauville(a)axway.com<mailto:jpribeauville@axway.com> http://www.axway.com<http://www.axway.com/> P Pensez à l'environnement avant d'imprimer.

3 2