- Users - libvirt List Archives

[libvirt-users] Questions on libvirt hook scripts
by Whit Blauvelt 12 Jul '11

12 Jul '11

Hi, Reading http://libvirt.org/hooks.html I'm wondering of the following implication is accurate: Under "QEMU guest migration" it says At the beginning of the migration, the qemu hook script on the _destination_ host is executed with the "start" operation. Above that, under "Specifics" Before a QEMU guest is started, the qemu hook script is called in two locations; if either location fails, the guest is not started. The first location, since 0.9.0, is before libvirt performs any resource labeling.... This is called as: /etc/libvirt/hooks/qemu guest_name prepare begin - The second location, available Since 0.8.0, occurs after libvirt has finished labeling all resources, but has not yet started the guest, called as: /etc/libvirt/hooks/qemu guest_name start begin - Is it accurate that under 0.9.x, starting a VM will run the qemu hook script as both "prepare" and "start," but migrating a VM will only run the qemu hook script as "start"? Or is the description incomplete with libvirt also running "prepare" on migration? Another question, under "Calling libvirt functions from within a hook script" it says DO NOT DO THIS! A hook script must not call back into libvirt, as the libvirt daemon is already waiting for the script to exit. A deadlock is likely to occur. That makes obvious sense. But does it apply to calling libvirt on a second system? For example, a script might check the status of VMs on a second system as a condition of starting a VM on first, using virsh. My guess is that shouldn't deadlock. But then since during migration "the qemu hook script on the _destination_ host is executed," if the destination hook script were in turn to use virsh to check the _source_ host, then in a roundabout way a libvirt function has been called from within the hook script. I can't quite picture whether this particular instance would be dangerous. If it's accurate that the _destination_ hook script is only called with "start" and not with "prepare," the second question matters less. The checks on the second system that I want to do can be limited to the "prepare" invocation, and so never loop back. On the other hand, if the destination hook script _is_ called with "prepare" during migration, it would be useful to know how to tell if it's being called as part of a migration. TIA, Whit

1 0

[libvirt-users] NPIV + KVM
by Eric Ross 11 Jul '11

11 Jul '11

Is this possible: I would like to have a VM use a FC LUN exported to a virtual wwn (using NPIV) as its primary disk. I was thinking there might be a mechanism to say VM, boot using a LUN with this vwwn as your primary disk? Thanks in advance. -- --------------------------------------------------------------------- Eric Ross HPC Systems Administrator Argonne National Laboratory Leadership Computing Facility ---------------------------------------------------------------------

2 1

[libvirt-users] Creating a CENTOS VM
by Dale Amon 11 Jul '11

11 Jul '11

I used the following command: virt-install --connect qemu:///system -n server -r 512 --vcpus=1 -f kvm/server.cyberteams.com/root.qcow2 -s 12 -c iso/CentOS-5.6-x86_64-bin-DVD-1of2.iso --vnc --noautoconsole --os-type linux --accelerate --network=bridge:br0 --hvm which with a BlueOnyx CD some months ago worked fine. However when I connected with virt-viewer, after going through the TEST function it then claimed it could not find the CD. Is CENTOS a bit strange in some way?

1 1

[libvirt-users] Changing path of VM configuration files
by Jatin Kumar 08 Jul '11

08 Jul '11

Hello, I am using shared storage and i want to keep all the configuration files of vms of different hosts in one single location (on shared storage). Is there a way i can change the default path=/etc/libvirt/qemu/<VMName>.xml to something else. I looked into libvirtd.conf and qemu.conf but could not find any such thing. -- Jatin

3 2

[libvirt-users] KVM and <interface type='ethernet'>
by Maximilian Wilhelm 08 Jul '11

08 Jul '11

Hi! I'm trying to set up a VM with "just an tap device" with the following config (basicly inherited from 'virsh domxml-from-native qemu-argv'): <domain type='kvm'> [...] <interface type='ethernet'> <mac address='52:54:1f:ac:08:51'/> <target dev='foo_eth0'/> <model type='virtio'/> </interface> [...] </devices> </domain> This sadly failes with: | error: Failed to start domain foo | error: internal error process exited while connecting to monitor: could not configure /dev/net/tun: Device or resource busy I'm wondering why this fails as there shouldn't anything be done with the tap device besides setting it up. (The basic goal would be to set up a VM with a tap device, put an v4 and v6 IP on it and let OSPF do the rest. As the <script> parameter will not work with dropped privileges I wrote a wrapper for the kvm binary which takes care of this via sudo. This part works, but I fail to get the VM started with <interface type='ethernet'>) Any ideas? Thanks Max -- <@uschebit> Das erklärt die Sitte. Das ist so ähnlich wie mit dem Verbeugen zur Begrüßung in Japan. Das dient auch nur dazu, durch Begutachten der Schuhqualität den sozialen Rang festzustellen.

1 0

[libvirt-users] Is arping useful if a VM is brought up on a different (bridged) interface/machine?
by Whit Blauvelt 08 Jul '11

08 Jul '11

Hi, Using libvirt/qemu/kvm VMs, if a VM has stopped on one host and then the identical VM is started afresh on a different host, with the exact same (bridged) VM IP, would there be a point in having the VM run arping (or arping2) on startup to help it be quickly found on the LAN? Or is this built in at some level already? Or does it not apply here? Thanks, Whit

1 0

[libvirt-users] SSH console for qemu
by Nikita A Menkovich 08 Jul '11

08 Jul '11

Hello, I'm using qemu with libvirt and I want to create SSH emergency console to linux/unix guests, same way as it is realized in qemu VNC. Maybe someone could help me to determine a place to dive in this question. I think the best way to make it - extend qemu API. But if this could be made with libvirt - this will be also fine. -- Nikita A Menkovich http://libc6.org/ JID: menkovich(a)gmail.com

3 5

[libvirt-users] Guest-networking: Two subnets = Guests cannot connect to the other subnet
by mailinglists＠florian-kaiser.net 07 Jul '11

07 Jul '11

Hello list, We have KVM setup using libvirt (0.9.0) as configuration device on a Debian Squeeze Host. Networking is done in routed bridge mode and we have two different subnets, as well as one main ip. Subnets get routed to the main ip (this is done by our provider). Network on hosts are done like this: <network> <name>subnet_1</name> <uuid>7c60990d-ab31-c4a5-cd2b-5ce4c7c61d38</uuid> <forward dev='eth0' mode='route'/> <bridge name='virtbridge1' stp='off' delay='0' /> <mac address='52:54:00:4D:83:2A'/> <ip address='xx.xx.240.121' netmask='255.255.255.248'> </ip> </network> <network> <name>subnet_2</name> <uuid>7c60990d-ab31-c4a5-cd2b-5ce4c7c61d38</uuid> <forward dev='eth0' mode='route'/> <bridge name='virtbridge1' stp='off' delay='0' /> <mac address='52:54:00:4D:83:2A'/> <ip address='xx.xx.240.200' netmask='255.255.255.248'> </ip> </network> Routing on host Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface xx.xx.122.33 0.0.0.0 255.255.255.255 UH 0 0 0 eth0 xx.xx.240.120 0.0.0.0 255.255.255.248 U 0 0 0 virtbridge1 xx.xx.239.200 0.0.0.0 255.255.255.248 U 0 0 0 virtbridge2 0.0.0.0 xx.xx.122.33 0.0.0.0 UG 0 0 0 eth0 Guests get this: <interface type='network'> <mac address='de:ad:b6:83:85:5b'/> <source network='subnet_1'/> <model type='virtio'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> </interface> Guests put this into interfaces: iface eth0 inet static address xx.xx.239.202 netmask 255.255.255.248 gateway xx.xx.239.201 Guest routing Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface xx.xx.239.200 0.0.0.0 255.255.255.248 U 0 0 0 eth0 0.0.0.0 xx.xx.239.201 0.0.0.0 UG 0 0 0 eth0 Everything seemed to work just fine, that is: Guests could connect - to host, - to internet - to other virtual maschines in the same subnet - all vice-versa. Internet can connect to any guest and our host can connect to any guest, too. Then, we needed one guest to connect to a guest from the other subnet. And this does not work, what I get follows: PING xx.xx.240.122 (xx.xx.240.122) 56(84) bytes of data. >From xx.xx.239.201 icmp_seq=1 Destination Port Unreachable >From xx.xx.239.201 icmp_seq=2 Destination Port Unreachable Note, that I can ping the other subnets gateway just fine! But I cannot connect to any of the other subnet IPs. I believe this is just a routing problem, but I cannot figure out what route to add where. Please ask if you need any more information.

2 1

[libvirt-users] (no subject)
by Abhishek Gupta 06 Jul '11

06 Jul '11

Hello I am having a libvirt installation on kvm, ubuntu 10.04. A few months ago, I setup a cloud on the hypervisors which was functioning really cool. But from last two-three days I am getting following error on using function snapshotListNames in python. >>> snapshots=dom.snapshotListNames(0) libvir: Remote error : unknown procedure: 187 Traceback (most recent call last): File "<stdin>", line 1, in <module> File "/usr/lib/python2.7/dist-packages/libvirt.py", line 890, in snapshotListNames if ret is None: raise libvirtError ('virDomainSnapshotListNames() failed', dom=self) libvirt.libvirtError: unknown procedure: 187 Any ideas so as what could be causing the error? -- Abhishek Webpage <http://abhishekgupta92.info/> !!! Blog<http://thelazy.info/abhishek>

1 0

[libvirt-users] Associating a statefile with an image
by sethuraman subbiah 06 Jul '11

06 Jul '11

Hi , I am trying to perform offline migration (i.e) Create an incremental image using the qcow format, transfer the vm memory state to a state fie.Use the image and statefile together as a template. Now create a new vm using the template. I can successfully do it using the following commands : Save phase : stop migrate "exec:gzip -c > STATEFILE.gz" qemu-img qemu-img create -b BASE_img -f qcow2 INCRE_img Restore phase : qemu-kvm -m 1024 -hda INCRE_img -incoming "exec: gzip -c -d STATEFILE.gz" And it works fine. But I am not able to find the vm with virt-manager or other libvirt based tools. If I use : virsh save <dom_id> STATEFILE I can restore using "virsh restore STATEFILE" but I want to associate this with the incremental image that I created, but I dont know how. I think it saves the existing vm's xml file along with the STATEFILE, thus I am not able to change the disk image to the incremental image. Thus , 1. Either I should find a way to make vms created using qemu-kvm appear in libvirt-based tools. OR 2. Find a way to associate the virsh save STATEFILE with a incremental image. Any help or hint with respect to these will be very helpful. I am try to do cloning with minimal cost. This takes only 25 seconds ( to create STATEFILE) and creating incremental image is instantatenous . But I want to use the vm monitoring code based on libvirt. And it makes life difficult. Thanks for your help. - Regards, Sethuraman Subbiah Graduate Student - NC state University M.S in Computer Science

2 3

[libvirt-users] Xm driver only supports modifying persistent config.
by zhuxuegang 05 Jul '11

05 Jul '11

Dear all. I use libvirt to attach a volume to xen's guest VM,but l got the following message: Attached device: /dev/sdb [Mon Jul 4 09:18:58 2011][006640][EUCAERROR ] libvirt: Requested operation is not valid: Xm driver only supports modifying persistent config (code=55) [Mon Jul 4 09:18:58 2011][006640][EUCAERROR ] AttachVolume() failed (err=-1) XML=<disk type='block'><driver name='phy'/><source dev='/dev/sdb'/><target dev='sdae'/></d isk> [Mon Jul 4 09:18:58 2011][006640][EUCAERROR ] ERROR: doAttachVolume() failed error=1 [Mon Jul 4 09:19:02 2011][006640][EUCADEBUG ] doDescribeResource() invoked [Mon Jul 4 09:19:02 2011][006640][EUCADEBUG ] doDescribeInstances() invoked any help will be greatly appreciated. Thank you in advance... 2011-07-05 zhuxuegang

1 0

[libvirt-users] Helping on libvirt
by negar behinaeen 03 Jul '11

03 Jul '11

*Writting myVirt* ------------------------------ Dear all, I'm new in libvirt. I would be glad if you help me. I want to write a program for controlling all libvirt servers named "MYVIRT". The libvirtd works as fallow: { CLIENT } ---------------------> { SERVER A } | | ------------------------------> { SERVER B } Now, I want to place MYVIRT between client and libvirtd servers. By calling a command, MYVIRT should execute the command for all servers. { CLIENT } | | { MYVIRT } ---------------------> { SERVER A } | | ------------------------------> { SERVER B } The client softwares for libvirtd are virsh and virt-manager. Now I want them to connect to MYVIRT and look it as a libvirtd server. Would you please guide me for connecting to MYVIRT through virsh and execute a command on all servers?

1 0

[libvirt-users] Unable to boot from a vdi image
by yeswanth swami 03 Jul '11

03 Jul '11

Hi, I am not able to boot a vdi image on virtualbox using libvirt .It keeps showing me "Bootable medium not found". I have tried booting it manually with virtualbox GUI and it boots without any problem. May there is some problem with my libvirt domain XML . I also realized on some investigation that it is not detecting the vdi image file as a hard disk (when i am booting from a live cd) A part of my libvirt domain XML is here: <os> <type>hvm</type> <boot dev='hd'/> </os> ------------------------------ <devices> <disk type='file' device='disk'> <source file='disk.vdi'/> <target dev='hdd' bus='ide'/> </disk> </devices> Thanks for replying -- Cheers, Yeswanth

1 0

[libvirt-users] [PATCH] Use correct param names in Product getters
by Zeeshan Ali (Khattak) 01 Jul '11

01 Jul '11

From: "Zeeshan Ali (Khattak)" <zeeshanak(a)gnome.org> --- osinfo/osinfo_product.c | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/osinfo/osinfo_product.c b/osinfo/osinfo_product.c index 89e01bd..bbbcf67 100644 --- a/osinfo/osinfo_product.c +++ b/osinfo/osinfo_product.c @@ -158,15 +158,15 @@ void osinfo_product_add_related(OsinfoProduct *product, OsinfoProductRelationshi const gchar *osinfo_product_get_vendor(OsinfoProduct *prod) { - return osinfo_entity_get_param_value(OSINFO_ENTITY(prod), OSINFO_PRODUCT_PROP_NAME); + return osinfo_entity_get_param_value(OSINFO_ENTITY(prod), OSINFO_PRODUCT_PROP_VENDOR); } const gchar *osinfo_product_get_version(OsinfoProduct *prod) { - return osinfo_entity_get_param_value(OSINFO_ENTITY(prod), OSINFO_PRODUCT_PROP_NAME); + return osinfo_entity_get_param_value(OSINFO_ENTITY(prod), OSINFO_PRODUCT_PROP_VERSION); } const gchar *osinfo_product_get_short_id(OsinfoProduct *prod) { - return osinfo_entity_get_param_value(OSINFO_ENTITY(prod), OSINFO_PRODUCT_PROP_NAME); + return osinfo_entity_get_param_value(OSINFO_ENTITY(prod), OSINFO_PRODUCT_PROP_SHORT_ID); } const gchar *osinfo_product_get_name(OsinfoProduct *prod) { -- 1.7.5.4

2 1

[libvirt-users] ANNOUNCE: oz 0.5.0 release
by Chris Lalancette 01 Jul '11

01 Jul '11

All, I'm pleased to announce release 0.5.0 of Oz. Oz is a program for doing automated installation of guest operating systems with limited input from the user. Release 0.5.0 is a bugfix and feature release for Oz. Some of the highlights between Oz 0.4.0 and 0.5.0 are: - Replace icicle-nc binary with a shell script to try various methods. Besides being more portable, this also allows us to convert Oz to a noarch RPM (thanks to Padraig Brady) - Support for Ubuntu 6.06 - Support for md5sum/sha1sum/sha256sum checking of ISOs after download - New -x flag for oz-install to write the libvirt XML file to a user-specified location (thanks to Steve Dake) - Support for OpenSUSE customization - Support for F-15 customization (thanks to Steve Dake) - Support for running commands at the end of package installation (thanks to Steve Dake) Fedora 14 and RHEL-6 packages are available for this release. Note that to install the RHEL-6 packages, you must be running RHEL-6.1 or later. Instructions of how to get and use Oz are available at http://aeolusproject.org/oz.html If you have any questions or comments about Oz, please feel free to contact aeolus-devel(a)lists.fedorahosted.org or me (clalance(a)redhat.com) directly. -- Chris Lalancette

1 0

[libvirt-users] vnc over ssh failing
by Pete Ashdown 30 Jun '11

30 Jun '11

I'm using virt-manager 0.8.6 client side on Ubuntu 11.04. I can't get vnc over ssh to work to a kvm guest running on a Ubuntu 10.04 host. kvmhost:~> virsh vncdisplay testing :0 kvmhost:~> telnet 127.0.0.1 5900 Trying 127.0.0.1... Connected to 127.0.0.1. Escape character is '^]'. RFB 003.008 Client virt-manager debug: 2011-06-24 11:28:56,055 (console:923): Starting connect process for proto=vnc trans=tcp connhost=192.168.1.10 connuser=pashdown connport=None gaddr=127.0.0.1 gport=5900 gsocket=None 2011-06-24 11:28:56,120 (console:810): Viewer disconnected I can reach the console with: xtightvncviewer -via 192.168.1.10 127.0.0.1:0 Any suggestions are appreciated.

1 0

[libvirt-users] ANNOUNCE: libvirt Perl bindings Sys::Virt 0.2.8 release
by Daniel P. Berrange 30 Jun '11

30 Jun '11

I am happy to announce a new release[1] of the libvirt Perl bindings, Sys::Virt 0.2.8 is available for download: http://search.cpan.org/CPAN/authors/id/D/DA/DANBERR/Sys-Virt-0.2.8.tar.gz Changes in this release - Add all new APIs in libvirt 0.9.0 - Add examples for different ways of using stream APIs to download/upload storage volumes - Requires libvirt >= 0.9.0 The permanent link for the Sys::Virt package, where all historic releases can be found, is http://search.cpan.org/dist/Sys-Virt/ Thanks to everyone who contributed to this release through testing, bug reporting, patch submission, etc Regards, Daniel [1] Yes, 2 releases in 2 days. I'm playing catchup with main libvirt releases and instead of going straight to latest one (0.9.2), I'm releasing Perl bindings for the 2 intermediate releases since those are needed in various distros. Guess what to expect tomorrow.... -- |: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|

1 0

[libvirt-users] ANNOUNCE: libvirt Perl bindings Sys::Virt 0.2.7 release
by Daniel P. Berrange 30 Jun '11

30 Jun '11

I am happy to announce a new release of the libvirt Perl bindings, Sys::Virt 0.2.7 is available for download: http://search.cpan.org/CPAN/authors/id/D/DA/DANBERR/Sys-Virt-0.2.7.tar.gz Changes in this release - Honour flags parameter when dumping XML - Use thread local error APIs exclusively - Fix bogus extra URI parameter in migrate_to_uri function - Implement all virStream APIs in Sys::Virt::Stream - Impement virDomainGetSysinfo binding - Call virInitialize when loading module - Add all missing error constants - Update example RPM spec to match Fedora - Requires libvirt >= 0.8.8 The permanent link for the Sys::Virt package, where all historic releases can be found, is http://search.cpan.org/dist/Sys-Virt/ Thanks to everyone who contributed to this release through testing, bug reporting, patch submission, etc Regards, Daniel -- |: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|

1 0

[libvirt-users] Starting libvirtd cuts off host access to external network
by PLD 24 Jun '11

24 Jun '11

I've spent some hours on this without success - any help greatly appreciated. I've just done a new RHEL6 setup, with a KVM guest (first time). The basic installation works fine until I start libvirtd; at this point, the host machine loses access to the external network: step 1: ping from host to network works; external machine on network can ping both eth0 and br1 on the host step 2: /sbin/service libvirtd start step 3: ping from host to network fails; external machine on the network can't ping either eth0 or br1 on the host Stopping libvirtd (/sbin/service libvirtd stop) makes no difference - the network remains unreachable. Any idea why this would happen? I have installed a kvm guest, using the default network script at /etc/libvirtd/qemu/networks/default.xml (although I've changed the IP addresses to be on my subnet, and the DHCP server returns only one address). I have also configured a bridge; I've attached ifcfg-eth0, ifcfg-eth1, and ifcfg-br1 below. I haven't changed any iptables configuration. If I run the guest (also RHEL6) the guest can talk to the host, but it can't see the outside world. Thanks - Paul ================================================ ifcfg-eth0: ----------- NM_CONTROLLED="no" DEVICE="eth0" HWADDR=00:16:E6:8A:C7:20 ONBOOT=yes TYPE=Ethernet DEFROUTE=yes IPV4_FAILURE_FATAL=yes IPV6INIT=no NAME="System eth0" NETMASK=255.255.255.0 BOOTPROTO=dhcp ifcfg-eth1: ----------- NM_CONTROLLED="no" DEVICE="eth1" HWADDR=00:16:E6:8A:C7:21 ONBOOT=yes TYPE=Ethernet DEFROUTE=yes IPV4_FAILURE_FATAL=yes IPV6INIT=no NAME="System eth1" BRIDGE=br1 NETMASK=255.255.255.0 ifcfg-br1: ---------- NM_CONTROLLED="no" DEVICE=br1 BOOTPROTO=dhcp ONBOOT=yes TYPE=Bridge DELAY=0

2 2

Re: [libvirt-users] libvirt-users Digest, Vol 18, Issue 17
by m m 21 Jun '11

21 Jun '11

Yes, I also think that iptables module should be loaded before libvirt-bin and I have seen a post which regarded loading conntrack modules and these modules also seem to be necessary... I know a little about Upstart in Ubuntu. However, I still don't know how to properly solve this problem. ip_tables module depend on other modules being loaded and since I don't know the full list of modules which should be loaded (for IP tables to work) I don't know how to handle that. Besides, even if I had this list, it could change for some time, so it is not a good idea to rely on that. Version which I have installed is the newest release from Ubuntu repositories. Thank you for help. I will be grateful for any further suggestions. 2011/6/13 <libvirt-users-request(a)redhat.com> > Send libvirt-users mailing list submissions to > libvirt-users(a)redhat.com > > To subscribe or unsubscribe via the World Wide Web, visit > https://www.redhat.com/mailman/listinfo/libvirt-users > or, via email, send a message with subject or body 'help' to > libvirt-users-request(a)redhat.com > > You can reach the person managing the list at > libvirt-users-owner(a)redhat.com > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of libvirt-users digest..." > > > Today's Topics: > > 1. Re: libvirt-bin MOSTLY fails to create default virtual > network (Laine Stump) > 2. Boot delay (Jeff) > 3. e: How to disconnect the ISO image using virsh (Jeff) > 4. Difference between function Migrate and Migrate2 (Abhishek Gupta) > 5. Re: Boot delay (Michal Privoznik) > 6. hook scripts - not in Ubuntu packages? (Whit Blauvelt) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Sun, 12 Jun 2011 12:03:17 -0400 > From: Laine Stump <laine(a)laine.org> > To: libvirt-users(a)redhat.com > Subject: Re: [libvirt-users] libvirt-bin MOSTLY fails to create > default virtual network > Message-ID: <4DF4E345.7060104(a)laine.org> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > On 06/06/2011 03:24 PM, m m wrote: > > I have observed 3 scenarios after I reboot the host with 2 guests > > (which are supposed to autostart after boot/reboot of the host): > > 1. Default network is created but I can not ping guests. Gests don't > > have assigned IP address or they both have IP address 192.168.122.100. > > 2. Default network is not created. > > 3. Default netork is properly created and everything works fine. > > In each case guests are started and they work properly. Cases 1. and > > 2. happen a lot more often then case 3. > > > > My solutions: > > - When I restart libvirt-bin then everything is fine. > > - When I set 'sleep 40' in 'pre-start' section of > > /etc/init/libvirt-bin.conf and modify 'start on' section in the > > following way: > > > > start on (runlevel [2345] and net-device-up IFACE=br0) > > > > then everything works fine. > > > > > > However, none of the above solutions indeed solves the problem. > > Please, let me know if there is solution to this problem or what are > > your suggestions regarding this issue. > > > > **** > > HOST$ uname -a > > Linux ras 2.6.32-32-server #62-Ubuntu SMP Wed Apr 20 22:07:43 UTC 2011 > > x86_64 GNU/Linux > > > > GUEST1 and GUEST2 > > $ uname -a > > Linux server1-desktop 2.6.32-32-generic #62-Ubuntu SMP Wed Apr 20 > > 21:54:21 UTC 2011 i686 GNU/Linux > > > > HOST$ libvirtd --version > > libvirtd (libvirt) 0.7.5 > > > > My idea was that some kernel modules sometimes are loaded before > > libvirt-bin starts, and sometimes libvirt-bin starts earlier, or in > > the middle of loading of these modules, that is why 3 scenarios can be > > observed. > > That is why I have added: > > cp /proc/modules /home/wsadmin2/modules1 > > cp /proc/modules /home/wsadmin2/modules2 > > to 'pre-start' and 'post-start' parts respectively of > > /etc/init/libvirt-bin.conf. > > > > Of course, beween execution of > > - pre-start part > > - exec /usr/sbin/libvirtd -d > > - post-start part > > other kernel modules can be loaded, so we can not be sure which > > additional modules in file 'modules2' were indeed loaded before > > libvirtd was actually started. > > > > Here are my observations: > > Scenatio 1. > > 33 (fixed) modules were logged - call them BASE MODULES. > > > > Scenatio 2. > > Beside BASE MODULES + the following modules were loaded: > > > > > iptable_filter 1841 0 - Live 0xffffffffa007c000 > > > ip_tables 18201 1 iptable_filter, Live 0xffffffffa0172000 > > > x_tables 22361 1 ip_tables, Live 0xffffffffa0164000 > > In your "scenario 1" the modules necessary for iptables to function > haven't been loaded. Each virtual network requires a set of iptables > rules to function properly, and if there is no iptables, the rules can't > be added and network creation fails. Your libvirtd.log confirms that > problem: > > 10:06:06.980: error : networkAddIptablesRules:750 : failed to add > iptables rule to allow DHCP requests from 'virbr0': Invalid argument > > > 0.7.5 is quite old (> 1 year), is there perhaps a newer pre-built > version of libvirt available for your release of Ubuntu? > > It seems that Ubuntu's method of starting up libvirtd (at least on your > machine) is missing a dependency to force iptables to be loaded before > libvirtd starts. I'm unfamiliar with Ubuntu/debian startup stuff > (upstart, I guess), but that's where you need to look for the solution. > > I found the following in the ubuntu bug database, which seems similar, > but not identical (basically the system networking isn't completely > ready by the time the autostart domains are brought up): > > https://bugs.launchpad.net/ubuntu/+source/libvirt/+bug/495394 > > > > > Scenatio 3. > > Beside BASE MODULES + the following modules were loaded: > > > > $ diff modules1 modules2 > > 1c1,6 > > < iptable_filter 1841 0 - Live 0xffffffffa008b000 > > --- > > > xt_state 1490 14 - Live 0xffffffffa01ab000 > > > xt_tcpudp 2667 12 - Live 0xffffffffa01a5000 > > > nf_conntrack_ipv4 12742 14 - Live 0xffffffffa019b000 > > > nf_conntrack 73326 2 xt_state,nf_conntrack_ipv4, Live > 0xffffffffa017d000 > > conntrack is used by iptables. I haven't tried it, but wouldn't be > surprised if adding rules that required conntrack failed if it hadn't > been loaded, or something like that. > > > > nf_defrag_ipv4 1481 1 nf_conntrack_ipv4, Live 0xffffffffa0177000 > > > iptable_filter 1841 1 - Live 0xffffffffa008b000 > > 3c8 > > < x_tables 22361 1 ip_tables, Live 0xffffffffa0162000 > > --- > > > x_tables 22361 3 xt_state,xt_tcpudp,ip_tables, Live 0xffffffffa0162000 > > > > Other observations: > > Scenatio 1. > > No errors in libvirtd.log > > Scenario 2. > > > > 10:06:06.980: error : networkAddIptablesRules:750 : failed to add > > iptables rule to allow DHCP requests from 'virbr0': Invalid argument > > (...) > > 10:06:08.630: error : qemudNetworkIfaceConnect:1445 : Failed to add > > tap interface to bridge 'virbr0': No such device > > 10:06:08.630: error : qemuAutostartDomain:584 : Failed to autostart VM > > 'cluster_server1': Failed to add tap interface to bridge 'virbr0': No > > such device > > > > 10:06:08.921: info : qemudDispatchSignalEvent:390 : Received > > unexpected signal 17 > > 10:06:08.927: debug : virExecWithHook:617 : LC_ALL=C /usr/bin/kvm -help > > 10:06:08.938: info : qemudDispatchSignalEvent:390 : Received > > unexpected signal 17 > > 10:06:09.000: error : qemudNetworkIfaceConnect:1445 : Failed to add > > tap interface to bridge 'virbr0': No such device > > 10:06:09.000: error : qemuAutostartDomain:584 : Failed to autostart VM > > 'cluster_server2': Failed to add tap interface to bridge 'virbr0': No > > such device > > (...) > > Scenatio 3. > > No errors in libvirtd.log > > > > > > I didn't see any valuable logs in /var/log/boot.log, > > /var/log/messages, /var/log/syslog. Please let me know if you want to > > see them anyway or have any suggestions right away. > > > > > > _______________________________________________ > > libvirt-users mailing list > > libvirt-users(a)redhat.com > > https://www.redhat.com/mailman/listinfo/libvirt-users > > > > ------------------------------ > > Message: 2 > Date: Mon, 13 Jun 2011 08:25:00 -0500 > From: Jeff <jwk1230(a)gmail.com> > To: libvirt-users(a)redhat.com > Subject: [libvirt-users] Boot delay > Message-ID: <4609858C-FB73-4E8A-B81B-3EDCBB83B755(a)gmail.com> > Content-Type: text/plain; charset=us-ascii > > Is there a method of tweaking the boot sequence of a KVM instance so that > you can delay the process for X number of seconds? I don't have enough time > between starting the KVM and connecting via VNC. > > Thanks > > Jeff > > > > ------------------------------ > > Message: 3 > Date: Fri, 10 Jun 2011 15:12:36 -0500 > From: Jeff <jwk1230(a)gmail.com> > To: Osier Yang <jyang(a)redhat.com>, libvirt-users(a)redhat.com > Subject: [libvirt-users] e: How to disconnect the ISO image using > virsh > Message-ID: <F301FBCE-65F1-411A-980E-EF8B6A46241B(a)gmail.com> > Content-Type: text/plain; charset="utf-8" > > I am trying to detach an ISO from a KVM machine and I've tried both > attach-disk using " " in the file path as well as the update-device option > but am getting this error > > error: Failed to update device from /var/run/libvirt/qemu/kvmtest1.xml > error: XML description for unknown device type is not well formed or > invalid. > > I tried doing a virsh edit on the xml file to remove the source file path > to the ISO, as I assume is the correct method, and then running the virsh > update-device but it's not working. > > Can anyone advise the correct steps to detach a device from a running > and/or offline domain? > > Additionally - anyone have any guidance on how to attach a CD device to a > domain that is currently shutdown, for example, if I wanted to boot the > domain to a rescue CD image? > > Thanks > > Jeff > > > ? 2011?03?31? 15:17, Masanari Iida ??: > Somebody sent me a tips. > Following command do what I wanted to do. > Thanks. > > attach-disk guest01 " " hdc --driver file --type cdrom > --mode readonly > Also using "update-device" with an xml which doesn't specify > the source file path will work. > > On Thu, Mar 31, 2011 at 3:29 PM, Masanari Iida<standby24x7 gmail com> > wrote: > Hello, > > I can attach disc ISO image to cd-rom from virsh. > And I can replace the ISO image from disc1 to disc2 using the 2nd example. > > (example) > attach-disk guest01 /root/disc1.iso hdc --driver file --type cdrom > --mode readonly > attach-disk guest01 /root/disc2.iso hdc --driver file --type cdrom > --mode readonly > > My question is, how can I disconnect the image from the cdrom drive using > virsh. > Because, if I try to detach-disk hdc, I got following errors. > > virsh # detach-disk guest01 hdc > error: Failed to detach disk > error: this function is not supported by the connection driver: > This type of device cannot be hot unplugged > > What I can do is possible from virt-manager's "Disconnect" button, > (next to source path: line) from Hardware's TAB, Disk hdc screen. > I want to do this from virsh. > > OS: RHEL5.6 > libvirt-0.8.2-15.el5_6.1 is used. > > Regards, > Masanari Iida > > _______________________________________________ > libvirt-users mailing list > libvirt-users redhat com > https://www.redhat.com/mailman/listinfo/libvirt-users > > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: < > https://www.redhat.com/archives/libvirt-users/attachments/20110610/f4acfe4c… > > > > ------------------------------ > > Message: 4 > Date: Sat, 11 Jun 2011 06:39:46 +0530 > From: Abhishek Gupta <cs1090174(a)cse.iitd.ernet.in> > To: libvirt-users(a)redhat.com > Subject: [libvirt-users] Difference between function Migrate and > Migrate2 > Message-ID: <BANLkTinv6XiRMz=rS88A-U-b2Hz3SVTesg(a)mail.gmail.com> > Content-Type: text/plain; charset=ISO-8859-1 > > I was reading the API reference for the libvirt, when I came across > the function? virDomainMigrate and virDomainMigrate2. I am unable to > find the difference between the two. Can someone please tell the > difference in the application of the two? > > -- > Abhishek > > > > ------------------------------ > > Message: 5 > Date: Mon, 13 Jun 2011 17:21:02 +0200 > From: Michal Privoznik <mprivozn(a)redhat.com> > To: Jeff <jwk1230(a)gmail.com> > Cc: libvirt-users(a)redhat.com > Subject: Re: [libvirt-users] Boot delay > Message-ID: <4DF62ADE.60209(a)redhat.com> > Content-Type: text/plain; charset=ISO-8859-1 > > On 13.06.2011 15:25, Jeff wrote: > > Is there a method of tweaking the boot sequence of a KVM instance so that > you can delay the process for X number of seconds? I don't have enough time > between starting the KVM and connecting via VNC. > > > You might consider using virt-viewer. It simply waits for domain to > boot, if it is down. > > Your other option is to create hook script: > http://libvirt.org/hooks.html > and put conditional sleep there, because libvirt just before running a > domain tries to execute qemu with some arguments (e.g. -help) to qeury > its capabilities. And you don't want your hook script to delay that. > > So you need something like this: > > if [ $2 == "start" ]; then > sleep 5 > fi > > > Thanks > > > > Jeff > > > > _______________________________________________ > > libvirt-users mailing list > > libvirt-users(a)redhat.com > > https://www.redhat.com/mailman/listinfo/libvirt-users > > Michal > > > > ------------------------------ > > Message: 6 > Date: Mon, 13 Jun 2011 11:42:44 -0400 > From: Whit Blauvelt <whit.virt(a)transpect.com> > To: libvirt-users(a)redhat.com > Subject: [libvirt-users] hook scripts - not in Ubuntu packages? > Message-ID: <20110613154244.GA28777(a)black.transpect.com> > Content-Type: text/plain; charset=us-ascii > > On Mon, Jun 13, 2011 at 05:21:02PM +0200, Michal Privoznik wrote: > > > Your other option is to create hook script: > > http://libvirt.org/hooks.html > > That looks useful. > > On several Ubuntu systems, with Ubuntu libvirt 0.8.8 and 0.8.3 packages > installed, there's no libvirt/hooks/ directory at all. Is this a directory > that can simply be created and used under libvirt, or does using this > feature on Ubuntu require custom installation? > > Thanks, > Whit > > > > ------------------------------ > > _______________________________________________ > libvirt-users mailing list > libvirt-users(a)redhat.com > https://www.redhat.com/mailman/listinfo/libvirt-users > > End of libvirt-users Digest, Vol 18, Issue 17 > ********************************************* >

2 1

[libvirt-users] Bypass USB to Xen VM
by Lorenzo Milesi 20 Jun '11

20 Jun '11

Hi. I'm trying to bypass an USB device from the host to a paravirtualized Xen VM. I found this page: http://libvirt.org/formatdomain.html#elementsUSB which says how to bypass usb and pci devices to guests. It says that works on KVM only, but is that notice about PCI or about ALL? Anyway I tried both from virt-manager and from virsh edit domain by adding the following: <hostdev mode='subsystem' type='usb'> <source> <vendor id='0x03f0'/> <product id='0x0114'/> </source> </hostdev> but the edit is ignored, right after I save, if I do a dumpxml the "hostdev" stanza is not present. Why? Any other way to bypass? usbip? thanks

1 0

[libvirt-users] Display port allocation
by Tom Hughes 20 Jun '11

20 Jun '11

The automatic port allocation algorithm for displays in libvirt seems to assume that it is the only thing that will be allocating ports in the normal VNC range starting at port 5900. My host is running vino to export it's display, which has bound itself to port 5900, yet libvirt is still trying to use port 5900 for both VNC and Spice displays if I leave it in auto allocation mode. I thought that this used to work in F14 but having just tried it again it looks like I may have been mistaken. Certainly it doesn't seem to work properly in F15 either with the standard libvirt or with the virt-preview one. Tom -- Tom Hughes (tom(a)compton.nu) http://compton.nu/

1 0

Re: [libvirt-users] virsh slow on execution
by Lorenzo Milesi 18 Jun '11

18 Jun '11

> Also, when I start a VM via OCF, while guest is booting the command > virsh domstate mydom > returns "no state". This makes PaceMaker believe the resource is not > running and continuously stops and restart the machine. I've let a shell open with virsh domstate mydom active in loop, and it seems that it returns the correct state one time once very much requests! Say over 20 requests, only one or two says "idle" or "running", all the others return "no state". I'd like to know why virsh fails so many times returning correct state... thanks

1 0

[libvirt-users] virsh slow on execution
by Lorenzo Milesi 18 Jun '11

18 Jun '11

Hi. I've set up a Debian6+Xen4+libvirt0.9 server. VMs are going to be managed by PaceMaker with VirtualDomain primitive. As found on some mailing list I've set export VIRSH_DEFAULT_CONNECT_URI="xen:///" in root's .bashrc, but when I run virsh commands, it takes up to 10s to get the result! Is this normal? At some point today I had issues connecting to libvirt, and in this case it was responding fast, less than couple of seconds. Also, when I start a VM via OCF, while guest is booting the command virsh domstate mydom returns "no state". This makes PaceMaker believe the resource is not running and continuously stops and restart the machine. How can I speed up virsh output? Thanks

1 0

[libvirt-users] Problems migrating xen domains
by Lorenzo Milesi 18 Jun '11

18 Jun '11

Hi. I've set up a cluster with two Debian 6 nodes, running Xen 4 and libvirt 0.9 (from backports). My VMs run over a DRBD resource, to allow live migration between the nodes. I've successfully tried live migration using xm migrate, now I'd like to make it work using virsh, but cannot make it. Here's the error I'm getting: # virsh migrate efw xen+ssh://10.22.22.152/ error: POST operation failed: xend_post: error from xen daemon: (xend.err "can't connect: (-2, 'Name or service not known')") As said, xm migrate worked at first, so xend should be set up correctly. I've done several searches but with no luck. What could be the problem? thanks Relevant configs: # cat /etc/xen/xend-config.sxp |grep -v ^#|grep -v ^$ (xend-http-server yes) (xend-unix-server yes) (xend-relocation-server yes) (xend-port 8000) (xend-relocation-port 8002) (vif-script vif-bridge) (dom0-min-mem 196) (enable-dom0-ballooning yes) (total_available_memory 0) (dom0-cpus 0) (vncpasswd '') # env|grep VIRSH_DEFAULT_CONNECT_URI VIRSH_DEFAULT_CONNECT_URI=xen:/// root@xensrv01:~# virsh version Compiled against library: libvir 0.9.0 Using library: libvir 0.9.0 Using API: Xen 3.0.1 Running hypervisor: Xen 4.0.0

1 0