[ovirt-users] self hosted engine vm won't start

[Martin Sivak]

The address is assigned by the engine and hosted engine tools only
consumes it via the OVF mechanism so an engine database dump might be
needed to investigate this, I have no idea how the address is
generated.

--
Martin Sivak
SLA / oVirt

On Tue, Oct 31, 2017 at 11:25 AM, Martin Sivak <msivak at redhat.com> wrote:
> Hi,
>
> adding a second NIC is supported, but we should definitely investigate
> why it has the same PCI slot address.
>
> Martin
>
> On Tue, Oct 31, 2017 at 10:56 AM, Hristo Pavlov <ichkata_66 at mail.ru> wrote:
>> Simone, Thanks a lot for the support!
>>
>> Yep, after remove the line of the additional network and starting with
>> custom vm file , the hosted engine virtual machine runs successfully.
>>
>> So far, the cluster will be in Global Maintainance, because the other nodеs
>> don't know about it.
>>
>> At the network settings of HostedEngine in Administration Portal still has
>> the second network adapter, I want to ask if I remove it, will remove it
>> from the original file /var/run/ovirt-hosted-engine-ha/vm.conf,  and to turn
>> off Global Maintainance to have high availability?
>>
>> Thank you very much!
>>
>> Понедельник, 30 октября 2017, 19:09 +02:00 от Simone Tiraboschi
>> <stirabos at redhat.com>:
>>
>>
>>
>>
>> On Mon, Oct 30, 2017 at 5:20 PM, Hristo Pavlov <ichkata_66 at mail.ru> wrote:
>>
>> I had added the second ethernet card through the Administration Panel, Edit
>> HostedEngine virtual machine and add new ethernet adapter. It didn't give me
>> any error.
>>
>>
>> This smells like a bug, we have to investigate it.
>>
>> In the mean time, try this:
>>
>> cp /var/run/ovirt-hosted-engine-ha/vm.conf
>> /var/run/ovirt-hosted-engine-ha/vm_fixed.conf
>> # edit /var/run/ovirt-hosted-engine-ha/vm_fixed.conf removing the line of
>> the additional network
>> hosted-engine --vm-start
>> --vm-conf=/var/run/ovirt-hosted-engine-ha/vm_fixed.conf
>>
>>
>>
>>
>> Понедельник, 30 октября 2017, 18:13 +02:00 от Simone Tiraboschi
>> <stirabos at redhat.com>:
>>
>>
>>
>>
>> On Mon, Oct 30, 2017 at 5:03 PM, Hristo Pavlov <ichkata_66 at mail.ru> wrote:
>>
>> Francesco,
>>
>> I noticed something in file /var/run/ovirt-hosted-engine-ha/vm.conf
>>
>> cpuType=Broadwell
>> emulatedMachine=pc-i440fx-rhel7.3.0
>> vmId=da98112d-b9fb-4098-93fa-1f1374b41e46
>> smp=2
>> memSize=6144
>> maxVCpus=16
>> spiceSecureChannels=smain,sdisplay,sinputs,scursor,splayback,srecord,ssmartcard,susbredir
>> vmName=HostedEngine
>> display=vnc
>> devices={index:0,iface:virtio,format:raw,bootOrder:1,address:{slot:0x06,bus:0x00,domain:0x0000,type:pci,function:0x0},volumeID:91738265-d3d1-4e8b-99f7-32e8838803b0,imageID:9a144135-8bea-4dfb-904c-3dd034e97e59,readonly:false,domainID:f0c9b7eb-2bd5-4af7-9b72-b5c279bb74cf,deviceId:9a144135-8bea-4dfb-904c-3dd034e97e59,poolID:00000000-0000-0000-0000-000000000000,device:disk,shared:exclusive,propagateErrors:off,type:disk}
>> devices={nicModel:pv,macAddr:00:16:3e:4f:93:b6,linkActive:true,network:ovirtmgmt,deviceId:030ba378-a536-4e24-af77-993bd76e2bd5,address:{slot:0x03,bus:0x00,domain:0x0000,type:pci,function:0x0},device:bridge,type:interface}
>> devices={nicModel:pv,macAddr:00:1a:4a:16:01:b8,linkActive:true,network:br202,deviceId:92833efc-352a-40e0-aa31-568e6dd66ea5,address:{slot:0x06,bus:0x00,domain:0x0000,type:pci,function:0x0},device:bridge,type:interface}
>>
>>
>> AFAIK hosted-engine VM has only one nic interface and you cannot add another
>> one from the engine.
>> How did you managed to get two nics?
>>
>>
>> devices={device:vnc,type:graphics,deviceId:fa67234b-013a-432c-9e8d-43dc35827a54,address:None}
>> devices={index:2,iface:ide,shared:false,readonly:true,deviceId:8c3179ac-b322-4f5c-9449-c52e3665e0ae,address:{controller:0,target:0,unit:0,bus:1,type:drive},device:cdrom,path:,type:disk}
>> devices={device:virtio,specParams:{source:urandom},model:virtio,type:rng}
>> devices={device:console,type:console}
>>
>> is it normal to duplicate these two pci addresses?
>>
>> Some time ago I had added the second ethernet card live at engine host.
>>
>> Can it be edited in any way?
>>
>> Thanks!
>>
>>
>>
>>
>>
>>
>> Понедельник, 30 октября 2017, 16:59 +02:00 от Francesco Romani
>> <fromani at redhat.com>:
>>
>>
>> On 10/30/2017 03:56 PM, Hristo Pavlov wrote:
>>
>> Thank you all!
>>
>> Francesco,
>>
>> [root at alpha ~]# journalctl -u libvirtd
>> -- Logs begin at Sun 2017-10-29 09:39:58 EET, end at Mon 2017-10-30 16:35:36
>> EET. --
>> Oct 29 09:41:24 alpha.datamax.bg systemd[1]: Starting Virtualization
>> daemon...
>> Oct 29 09:41:26 alpha.datamax.bg systemd[1]: Started Virtualization daemon.
>> Oct 29 22:30:26 alpha.datamax.bg libvirtd[4207]: libvirt version: 2.0.0,
>> package: 10.el7_3.9 (CentOS BuildSystem <http://bugs.centos.org>,
>> 2017-05-25-20:52:28, c1bm.rdu2.centos.org)
>> Oct 29 22:30:26 alpha.datamax.bg libvirtd[4207]: hostname: alpha.datamax.bg
>> Oct 29 22:30:26 alpha.datamax.bg libvirtd[4207]: XML error: Attempted double
>> use of PCI slot 0000:00:06.0 (may need "multifunction='on'" for device on
>> function 0)
>>
>>
>> This error will prevent the VM to start. Now, we need to see why it happens
>> in the first place. Will look at the vdsm logs as soon as possible.
>>
>> Bests,
>>
>> --
>> Francesco Romani
>> Senior SW Eng., Virtualization R&D
>> Red Hat
>> IRC: fromani github: @fromanirh
>>
>>
>>
>>
>>
>> _______________________________________________
>> Users mailing list
>> Users at ovirt.org
>> http://lists.ovirt.org/mailman/listinfo/users
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> _______________________________________________
>> Users mailing list
>> Users at ovirt.org
>> http://lists.ovirt.org/mailman/listinfo/users
>>