[one-users] Issues starting a VM
Ruben S. Montero
rsmontero at opennebula.org
Tue Nov 27 04:42:35 PST 2012
Hi
It seems that you have already created the VLANs, so you do not need
OpenNebula to do this for you. In your case you can just add the host and
use --net dummy. The 802.1Q drivers would create eth0.160 and a bridge for
it.
Now your problem, I think that OpenNebula is not able to detect your
hypervisor, probably because in yout system lsmod is not in "/sbin/lsmod".
Could you check that, and set it to the proper path in OpenNebulaNetwork.rb
that should be in /var/lib/one/remotes/vnm. (In multi-node installations
you would need onehost sync to propagate the changes). But as I said if you
want OpenNebula to do the bridge work for you you need to:
1.- Add to the Network templates
PHYDEV = eth0
VLAN = YES
VLAN_ID = 160 # This is optional OPenNebula would pick one for you
BRIDGE = server160 #This is optional OpenNebula would pick one for you
2.- Add a host with --net 802.1Q
I'd suggest to first keep trying with your current setup (manual VLAN),
remove the host and readd it with --net dummy; and then try the automatic
VLAN mode (as described above). For the automatic VLAN we need to solve the
issue you've reported that I believe could be the lsmod issue....
Cheers
Ruben
On Tue, Nov 27, 2012 at 8:42 AM, Stuart Longland <stuartl at vrt.com.au> wrote:
> Hi all,
>
> Firstly I'll start by pointing out that I'm very new to OpenNebula,
> having grown used to just using plain shell scripts to fire up QEMU
> instances.
>
> My workplace is looking to use OpenNebula on a multi-node cluster for
> their server infrastructure. The machines are Intel Core i3 systems
> with Intel DQ77KB mainboards. The systems run Ubuntu 12.04 AMD64.
>
> The plan is to use the mainboard's two onboard network cards in LACP to
> facilitate high-speed data transfer between VMs for live migration and
> for distributed storage using Ceph, and a third PCI express card on a
> VLAN trunk port facing the network.
>
> Right now I'm testing on a single node, not one of the ones we'll
> eventually use but rather an older, former VMWare ESX box.
>
> The status:
>
> - Ubuntu 12.04 is installed. OpenNebula 3.8.1 packages installed from
> the OpenNebula website.
> - I have the switch port connected to eth0 set up as a VLAN trunk
> - I have /etc/network/interfaces set up to create the eth0.XXX devices
> for each VLAN, and sensibly named bridge devices, `brctl show` looks
> like this:
> > root at bneprdvm0:~# brctl show
> > bridge name bridge id STP enabled interfaces
> > client130 8000.00101814cb1b no eth0.130
> > project144 8000.00101814cb1b no eth0.144
> > project159 8000.00101814cb1b no eth0.159
> > server160 8000.00101814cb1b no eth0.160
>
> This works fine with plain QEMU.
>
> Now, I've tried to create a virtual network to place a VM on the
> 'server160' bridge (i.e. I'd expect kvm to start; then it do a `brctl
> addif server160 tapX`). The following is my configuration for that
> network interface:
>
> > oneadmin at bneprdvm0:~$ onevnet show 4
> > VIRTUAL NETWORK 4 INFORMATION
> > ID : 4
> > NAME : Server 160 LAN
> > USER : oneadmin
> > GROUP : oneadmin
> > CLUSTER : -
> > TYPE : FIXED
> > BRIDGE : server160
> > VLAN : No
> > PHYSICAL DEVICE:
> > VLAN ID :
> > USED LEASES : 0
> >
> > PERMISSIONS
> > OWNER : um-
> > GROUP : ---
> > OTHER : ---
> >
> > VIRTUAL NETWORK TEMPLATE
> >
> >
> > FREE LEASES
> > LEASE=[ IP="10.87.160.13", MAC="52:54:00:12:34:58", USED="0", VID="-1" ]
>
> The VM template looks like this:
> > CPU="1.0"
> > DISK=[
> > CLONE="NO",
> > CLUSTER_ID="100",
> > DATASTORE="default",
> > DATASTORE_ID="1",
> > DEV_PREFIX="hd",
> > DISK_ID="0",
> > IMAGE="bneprdsrv5-hda",
> > IMAGE_ID="3",
> > IMAGE_UNAME="vrtadmin",
> > PERSISTENT="YES",
> > READONLY="NO",
> > SAVE="YES",
> > SOURCE="/var/lib/one/datastores/1/301aac89561c0fbeaf140cb09124f941",
> > TARGET="hda",
> > TM_MAD="shared",
> > TYPE="FILE" ]
> > MEMORY="256"
> > NAME="bneprdsrv5"
> > NIC=[
> > BRIDGE="project160",
> > IP="10.87.160.13",
> > MAC="52:54:00:12:34:58",
> > NETWORK="server160",
> > NETWORK_ID="2",
> > NETWORK_UNAME="vrtadmin",
> > PHYDEV="eth0.160",
> > VLAN="YES" ]
> > REQUIREMENTS="CLUSTER_ID = 100"
> > VMID="3"
> (Anyone know how to edit that by the way? So far all I seem to be able
> to do is blow away the VM configuration and rebuild from scratch!)
>
> I get the following error (apologies for the wall of text in advance):
> > Tue Nov 27 17:09:35 2012 [LCM][I]: New VM state is BOOT
> > Tue Nov 27 17:09:35 2012 [VMM][I]: Generating deployment file:
> /var/lib/one/3/deployment.0
> > Tue Nov 27 17:09:36 2012 [VMM][I]: Command execution fail:
> /var/tmp/one/vnm/802.1Q/pre PFZNPjxJRD4zPC9JRD48VUlEPj
> >
> A8L1VJRD48R0lEPjA8L0dJRD48VU5BTUU+b25lYWRtaW48L1VOQU1FPjxHTkFNRT5vbmVhZG1pbjwvR05BTUU+PE5BTUU+Ym5lcHJkc3J2NTwvTkF
> >
> NRT48UEVSTUlTU0lPTlM+PE9XTkVSX1U+MTwvT1dORVJfVT48T1dORVJfTT4xPC9PV05FUl9NPjxPV05FUl9BPjA8L09XTkVSX0E+PEdST1VQX1U+
> >
> MDwvR1JPVVBfVT48R1JPVVBfTT4wPC9HUk9VUF9NPjxHUk9VUF9BPjA8L0dST1VQX0E+PE9USEVSX1U+MDwvT1RIRVJfVT48T1RIRVJfTT4wPC9PV
> >
> EhFUl9NPjxPVEhFUl9BPjA8L09USEVSX0E+PC9QRVJNSVNTSU9OUz48TEFTVF9QT0xMPjA8L0xBU1RfUE9MTD48U1RBVEU+MzwvU1RBVEU+PExDTV
> >
> 9TVEFURT4yPC9MQ01fU1RBVEU+PFJFU0NIRUQ+MDwvUkVTQ0hFRD48U1RJTUU+MTM1NDAwMDE1NjwvU1RJTUU+PEVUSU1FPjA8L0VUSU1FPjxERVB
> >
> MT1lfSUQvPjxNRU1PUlk+MDwvTUVNT1JZPjxDUFU+MDwvQ1BVPjxORVRfVFg+MDwvTkVUX1RYPjxORVRfUlg+MDwvTkVUX1JYPjxURU1QTEFURT48
> >
> Q1BVPjwhW0NEQVRBWzEuMF1dPjwvQ1BVPjxESVNLPjxDTE9ORT48IVtDREFUQVtOT11dPjwvQ0xPTkU+PENMVVNURVJfSUQ+PCFbQ0RBVEFbMTAwX
> >
> V0+PC9DTFVTVEVSX0lEPjxEQVRBU1RPUkU+PCFbQ0RBVEFbZGVmYXVsdF1dPjwvREFUQVNUT1JFPjxEQVRBU1RPUkVfSUQ+PCFbQ0RBVEFbMV1dPj
> >
> wvREFUQVNUT1JFX0lEPjxERVZfUFJFRklYPjwhW0NEQVRBW2hkXV0+PC9ERVZfUFJFRklYPjxESVNLX0lEPjwhW0NEQVRBWzBdXT48L0RJU0tfSUQ
> >
> +PElNQUdFPjwhW0NEQVRBW2JuZXByZHNydjUtaGRhXV0+PC9JTUFHRT48SU1BR0VfSUQ+PCFbQ0RBVEFbM11dPjwvSU1BR0VfSUQ+PElNQUdFX1VO
> >
> QU1FPjwhW0NEQVRBW3ZydGFkbWluXV0+PC9JTUFHRV9VTkFNRT48UEVSU0lTVEVOVD48IVtDREFUQVtZRVNdXT48L1BFUlNJU1RFTlQ+PFJFQURPT
> >
> kxZPjwhW0NEQVRBW05PXV0+PC9SRUFET05MWT48U0FWRT48IVtDREFUQVtZRVNdXT48L1NBVkU+PFNPVVJDRT48IVtDREFUQVsvdmFyL2xpYi9vbm
> >
> UvZGF0YXN0b3Jlcy8xLzMwMWFhYzg5NTYxYzBmYmVhZjE0MGNiMDkxMjRmOTQxXV0+PC9TT1VSQ0U+PFRBUkdFVD48IVtDREFUQVtoZGFdXT48L1R
> >
> BUkdFVD48VE1fTUFEPjwhW0NEQVRBW3NoYXJlZF1dPjwvVE1fTUFEPjxUWVBFPjwhW0NEQVRBW0ZJTEVdXT48L1RZUEU+PC9ESVNLPjxNRU1PUlk+
> >
> PCFbQ0RBVEFbMjU2XV0+PC9NRU1PUlk+PE5BTUU+PCFbQ0RBVEFbYm5lcHJkc3J2NV1dPjwvTkFNRT48TklDPjxCUklER0U+PCFbQ0RBVEFbcHJva
> >
> mVjdDE2MF1dPjwvQlJJREdFPjxJUD48IVtDREFUQVsxMC44Ny4xNjAuMTNdXT48L0lQPjxNQUM+PCFbQ0RBVEFbNTI6NTQ6MDA6MTI6MzQ6NThdXT
> >
> 48L01BQz48TkVUV09SSz48IVtDREFUQVtzZXJ2ZXIxNjBdXT48L05FVFdPUks+PE5FVFdPUktfSUQ+PCFbQ0RBVEFbMl1dPjwvTkVUV09SS19JRD4
> >
> 8TkVUV09SS19VTkFNRT48IVtDREFUQVt2cnRhZG1pbl1dPjwvTkVUV09SS19VTkFNRT48UEhZREVWPjwhW0NEQVRBW2V0aDAuMTYwXV0+PC9QSFlE
> >
> RVY+PFZMQU4+PCFbQ0RBVEFbWUVTXV0+PC9WTEFOPjwvTklDPjxSRVFVSVJFTUVOVFM+PCFbQ0RBVEFbQ0xVU1RFUl9JRCA9IDEwMF1dPjwvUkVRV
> >
> UlSRU1FTlRTPjxWTUlEPjwhW0NEQVRBWzNdXT48L1ZNSUQ+PC9URU1QTEFURT48SElTVE9SWV9SRUNPUkRTPjxISVNUT1JZPjxPSUQ+MzwvT0lEPj
> >
> xTRVE+MDwvU0VRPjxIT1NUTkFNRT5ibmVwcmR2bTAudnJ0LmludDwvSE9TVE5BTUU+PEhJRD4yPC9ISUQ+PFNUSU1FPjEzNTQwMDAxNzU8L1NUSU1
> >
> FPjxFVElNRT4wPC9FVElNRT48Vk1NTUFEPnZtbV9rdm08L1ZNTU1BRD48Vk5NTUFEPjgwMi4xUTwvVk5NTUFEPjxUTU1BRD5zaGFyZWQ8L1RNTUFE
> >
> PjxEU19MT0NBVElPTj4vdmFyL2xpYi9vbmUvL2RhdGFzdG9yZXM8L0RTX0xPQ0FUSU9OPjxEU19JRD4wPC9EU19JRD48UFNUSU1FPjEzNTQwMDAxNzU8L1BTVElNRT48UEVUSU1FPjEzNTQwMDAxNzU8L1BFVElNRT48UlNUSU1FPjEzNTQwMDAxNzU8L1JTVElNRT48UkVUSU1FPjA8L1JFVElNRT48RVNUSU1FPjA8L0VTVElNRT48RUVUSU1FPjA8L0VFVElNRT48UkVBU09OPjA8L1JFQVNPTj48L0hJU1RPUlk+PC9ISVNUT1JZX1JFQ09SRFM+PC9WTT4=
> > Tue Nov 27 17:09:36 2012 [VMM][I]:
> /var/tmp/one/vnm/802.1Q/../OpenNebulaNic.rb:31:in `new_nic': undefined
> method `new' for nil:NilClass (NoMethodError)
> > Tue Nov 27 17:09:36 2012 [VMM][I]: from
> /var/tmp/one/vnm/802.1Q/../OpenNebulaNetwork.rb:60:in `initialize'
> > Tue Nov 27 17:09:36 2012 [VMM][I]: from
> /usr/lib/ruby/1.8/rexml/element.rb:892:in `each'
> > Tue Nov 27 17:09:36 2012 [VMM][I]: from
> /usr/lib/ruby/1.8/rexml/xpath.rb:53:in `each'
> > Tue Nov 27 17:09:36 2012 [VMM][I]: from
> /usr/lib/ruby/1.8/rexml/element.rb:892:in `each'
> > Tue Nov 27 17:09:36 2012 [VMM][I]: from
> /var/tmp/one/vnm/802.1Q/../OpenNebulaNetwork.rb:59:in `initialize'
> > Tue Nov 27 17:09:36 2012 [VMM][I]: from
> /var/tmp/one/vnm/802.1Q/../OpenNebulaNetwork.rb:110:in `new'
> > Tue Nov 27 17:09:36 2012 [VMM][I]: from
> /var/tmp/one/vnm/802.1Q/../OpenNebulaNetwork.rb:110:in `initialize'
> > Tue Nov 27 17:09:36 2012 [VMM][I]: from
> /var/tmp/one/vnm/802.1Q/HostManaged.rb:23:in `initialize'
> > Tue Nov 27 17:09:36 2012 [VMM][I]: from
> /var/tmp/one/vnm/802.1Q/../OpenNebulaNetwork.rb:100:in `new'
> > Tue Nov 27 17:09:36 2012 [VMM][I]: from
> /var/tmp/one/vnm/802.1Q/../OpenNebulaNetwork.rb:100:in `from_base64'
> > Tue Nov 27 17:09:36 2012 [VMM][I]: from /var/tmp/one/vnm/802.1Q/pre:24
> > Tue Nov 27 17:09:36 2012 [VMM][I]: ExitCode: 1
> > Tue Nov 27 17:09:36 2012 [VMM][I]: Failed to execute network driver
> operation: pre.
> > Tue Nov 27 17:09:36 2012 [VMM][E]: Error deploying virtual machine
> > Tue Nov 27 17:09:36 2012 [DiM][I]: New VM state is FAILED
>
> The error message was suspiciously like the following report…
> http://comments.gmane.org/gmane.comp.distributed.opennebula.user/8527
> … where the user miraculously fixed it by installing a missing package.
> Lord knows what that package was, as I seem to be missing it too and am
> having a hard time figuring out which one.
>
> Could someone on this list enlighten me as to what I might be doing
> wrong? Apologies if this has been asked before, but I can only find the
> above thread which gives me no further information.
>
> Regards,
> --
> ## -,-''''-. ###### Stuart Longland, Software Engineer
> ##. : ## : ## 38b Douglas Street -+*WE HAVE*+-
> ## # ## -'` .#' Milton, QLD, 4034 --+*MOVED*+--
> '#' *' '-. *' http://www.vrt.com.au
> S Y S T E M S T: 07 3535 9619 F: 07 3535 9699
> _______________________________________________
> Users mailing list
> Users at lists.opennebula.org
> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
>
--
Ruben S. Montero, PhD
Project co-Lead and Chief Architect
OpenNebula - The Open Source Solution for Data Center Virtualization
www.OpenNebula.org | rsmontero at opennebula.org | @OpenNebula
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opennebula.org/pipermail/users-opennebula.org/attachments/20121127/80e50cb3/attachment-0002.htm>
More information about the Users
mailing list