Re: [ovirt-users] nic order

2017-06-07 Thread Yevgeny Zaspitsky
Sorry, please ignore my suggestion. Now I realize that you actually want to
avoid re-ordering.

On Wed, Jun 7, 2017 at 2:05 PM, Yevgeny Zaspitsky <yzasp...@redhat.com>
wrote:

> You can activate reordering by using oVirt REST API. Sending POST request
> to http://${engine_address}/vms/${vm_id}/reordermacaddresses URL should
> do the job.
> Please note that it would reorder all vnics of the VM that PCI address
> wasn't assigned to them, in other words the VM wasn't run since the vnic
> was added to it.
>
> On Thu, Jun 1, 2017 at 2:57 PM, Nathanaël Blanchet <blanc...@abes.fr>
> wrote:
>
>>
>>
>> Le 01/06/2017 à 12:39, Dan Kenigsberg a écrit :
>>
>> On Wed, May 31, 2017 at 8:23 PM, Nathanaël Blanchet <blanc...@abes.fr> 
>> <blanc...@abes.fr> wrote:
>>
>> Hi all,
>>
>> I added a new nic on all my vms, but the MAC address pool was too small. So
>> I extended it from 00:1a:4a:22:1f:00 to 00:1a:4a:22:1f:ff =>
>> 00:1a:4a:22:1e:00 to 00:1a:4a:22:1f:ff.
>>
>> It was okay but when rebooting some vms, nics are interverted and I couldn't
>> join my vm on ip!
>>
>> I noticed nics were ordered now from the smallest MAC address to the
>> greatest one, so the last plugged nic from the new pool was now the first in
>> the vm, and all nics were moved at the bottom.
>>
>> Is it an expected behaviour?
>>
>> No. As far as I recall, and Yevgeny can add more details, reordering
>> of vNICs happens only when allocating multiple nics in the initial
>> definition of a VM; but not on plugging of a nic to an existing VM.
>>
>> Which version of ovirt-engine do you use?
>>
>> oVirt Engine Version: 4.1.2.2-1.el7.centos
>>
>>
>> --
>> Nathanaël Blanchet
>>
>> Supervision réseau
>> Pôle Infrastrutures Informatiques
>> 227 avenue Professeur-Jean-Louis-Viala
>> 34193 MONTPELLIER CEDEX 5
>> Tél. 33 (0)4 67 54 84 55
>> Fax  33 (0)4 67 54 84 14blanc...@abes.fr
>>
>>
>
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users


Re: [ovirt-users] nic order

2017-06-07 Thread Yevgeny Zaspitsky
You can activate reordering by using oVirt REST API. Sending POST request
to http://${engine_address}/vms/${vm_id}/reordermacaddresses URL should do
the job.
Please note that it would reorder all vnics of the VM that PCI address
wasn't assigned to them, in other words the VM wasn't run since the vnic
was added to it.

On Thu, Jun 1, 2017 at 2:57 PM, Nathanaël Blanchet  wrote:

>
>
> Le 01/06/2017 à 12:39, Dan Kenigsberg a écrit :
>
> On Wed, May 31, 2017 at 8:23 PM, Nathanaël Blanchet  
>  wrote:
>
> Hi all,
>
> I added a new nic on all my vms, but the MAC address pool was too small. So
> I extended it from 00:1a:4a:22:1f:00 to 00:1a:4a:22:1f:ff =>
> 00:1a:4a:22:1e:00 to 00:1a:4a:22:1f:ff.
>
> It was okay but when rebooting some vms, nics are interverted and I couldn't
> join my vm on ip!
>
> I noticed nics were ordered now from the smallest MAC address to the
> greatest one, so the last plugged nic from the new pool was now the first in
> the vm, and all nics were moved at the bottom.
>
> Is it an expected behaviour?
>
> No. As far as I recall, and Yevgeny can add more details, reordering
> of vNICs happens only when allocating multiple nics in the initial
> definition of a VM; but not on plugging of a nic to an existing VM.
>
> Which version of ovirt-engine do you use?
>
> oVirt Engine Version: 4.1.2.2-1.el7.centos
>
>
> --
> Nathanaël Blanchet
>
> Supervision réseau
> Pôle Infrastrutures Informatiques
> 227 avenue Professeur-Jean-Louis-Viala
> 34193 MONTPELLIER CEDEX 5 
> Tél. 33 (0)4 67 54 84 55
> Fax  33 (0)4 67 54 84 14blanc...@abes.fr
>
>
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users


Re: [ovirt-users] rhevm listening on ipv6 ?

2017-03-05 Thread Yevgeny Zaspitsky
Šimon,

According to [1] oVirt should be accessible through IPv6 address.

>From my experience engine web can be accessed by using an URL with an
IPv6 address in a browser.
Please notice that an IPv6 literal address (not a hostname that resolves to
an IPv6 address) in a valid URL should be wrapped by brackets, e.g. http://
[1::2]:80/path.

You can verify that the engine listens to an IPv6 by running the following
command from the Linux console of the host that runs ovirt-engine (while
ovirt-engine service runs):

netstat -lnp6 | grep 

where  is the port ovirt-engine should be listening to and be
accessed through (default is 80).

It seems to me like a problem on ManageIQ side.

Please let me know if you need further assistance.

Regards,
Yevgeny

[1] https://bugzilla.redhat.com/show_bug.cgi?id=1310837

On Fri, Mar 3, 2017 at 4:55 PM, Šimon  wrote:

> Hello Yevgeny,
>
> I am working on ipv6-only support for ManageIQ. And I have noticed that
> providers QA has around listen on ipv4 only.
>
> Is it possible for rhevm web/api to listen on ipv6? What are the steps I
> need to do?
>
> Thanks a lot!
> ~š.
>
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users


Re: [ovirt-users] Import guest at rhv4.0.6 failed after convert a regular ova file by virt-v2v

2017-01-05 Thread Yevgeny Zaspitsky
Kun,

I found that the OVF file includes a VM with a single vNic, but "
rasd:MACAddress" entry is missing there.
Could you please tell what was the MAC address in the source system and
what it is in oVirt once you've succeed importing the VM?
Also could you please tell what kind of system is the source system and
what is the tool you use for exporting VM out of there?

Are you able to find the MAC address info in the export file prior using
virt-v2v?

Regards,
Yevgeny

On Thu, Jan 5, 2017 at 1:42 PM, Kun Wei <ku...@redhat.com> wrote:

>
> Hi,I have been sent ovf file to Yevgeny in private
> Thanks
>
>
> - Original Message -
> From: "Yevgeny Zaspitsky" <yzasp...@redhat.com>
> To: "Kun Wei" <ku...@redhat.com>
> Cc: "Shahar Havivi" <shav...@redhat.com>, "users" <users@ovirt.org>,
> "Tingting Zheng" <tzh...@redhat.com>, "Ming Xie" <m...@redhat.com>
> Sent: Thursday, January 5, 2017 6:37:58 PM
> Subject: Re: Import guest at rhv4.0.6 failed after convert a regular ova
> file by virt-v2v
>
> Seems like engines tries to validate a vNic with an empty MAC address and
> fails on that. The fact that in the direct flow it works sounds weird to
> me: either we loose the MAC in one flow or we do not validate it in the
> other one.
>
> Kun, could you please send (in private if you like) the OVF file from the
> input tarball you've used, so we'd be able to debug the case more
> thoroughly?
>
> Regards,
> Yevgeny
>
> On Thu, Jan 5, 2017 at 10:03 AM, Kun Wei <ku...@redhat.com> wrote:
>
> >
> > Hi,shavivi.
> >
> > Thank you .Yes,as you know i can import ova directly and it can be import
> > successful.
> >
> > But  i am testing v2v now  and  i want to confirm if it is a new bug.
> >
> >
> >
> > - Original Message -
> > From: "Shahar Havivi" <shav...@redhat.com>
> > To: "Kun Wei" <ku...@redhat.com>, "Yevgeny Zaspitsky" <
> yzasp...@redhat.com
> > >
> > Cc: "users" <users@ovirt.org>, "Tingting Zheng" <tzh...@redhat.com>,
> > "Ming Xie" <m...@redhat.com>
> > Sent: Thursday, January 5, 2017 3:05:45 PM
> > Subject: Re: Import guest at rhv4.0.6 failed after convert a regular ova
> > file by virt-v2v
> >
> > Yevgeny, please look why it fails on "isMacInRange"
> >
> > Kun, you know that you can import ova directly instead of using virt-v2v
> to
> > export domain and then import to data domain,
> > you can do that from the VMs tab->Import-> and change the source to OVA.
> >
> > On Thu, Jan 5, 2017 at 4:35 AM, Kun Wei <ku...@redhat.com> wrote:
> >
> > >
> > > Hi,  I want to confirm a questions if it is a bug,or if something wrong
> > > with me.Thanks
> > >
> > > When i convert a ova guest file to rhv4.0.6 by virt-v2v,but import is
> > > failed.The attachment are engine.log and vdsm.log.
> > >
> > >
> > >
> > > Version:
> > > libguestfs-1.32.7-3.el7_3.2.x86_64
> > > virt-v2v-1.32.7-3.el7_3.2.x86_64
> > > qemu-img-rhev-2.6.0-28.el7_3.2.x86_64
> > > libvirt-2.0.0-10.el7_3.3.x86_64
> > > rhv4.0: 4.0.6.3
> > >
> > > Steps :
> > > 1:Prepare a regular ova file guest .
> > > 2:Use virt-v2v to convert the ova file to rhv4.0.
> > > # virt-v2v -i ova esx-rhel6.8-vmdkgz.tar -o rhev -os 10.73.131.93:
> > /home/nfs_export
> > > -of qcow2
> > > [   0.0] Opening the source -i ova esx-rhel6.8-vmdkgz.tar
> > > [  44.1] Creating an overlay to protect the source from being modified
> > > [  45.0] Initializing the target -o rhev -os 10.73.131.93:
> > /home/nfs_export
> > > [  45.3] Opening the overlay
> > > [  53.1] Inspecting the overlay
> > > [  70.2] Checking for sufficient free disk space in the guest
> > > [  70.2] Estimating space required on target for each disk
> > > [  70.2] Converting Red Hat Enterprise Linux Server release 6.8
> > (Santiago)
> > > to run on KVM
> > > virt-v2v: This guest has virtio drivers installed.
> > > [ 215.6] Mapping filesystem data to avoid copying unused and blank
> areas
> > > [ 216.0] Closing the overlay
> > > [ 217.2] Checking if the guest needs BIOS or UEFI to boot
> > > [ 217.2] Assigning disks to buses
> > > [ 217.2] Copying disk 1/1 to /tmp/v2v.8kDoXb/908b80fa-09a6-
> > > 4827-8e0a-a6a54520f685/images/47a1919e-1914-44b4-920f-
> > > b8e208fa4bc1/da74f1c3-8f43-4a15-934a-bb9d380615c3 (qcow2)
> > > (100.00/100%)
> > > [ 254.6] Creating output metadata
> > > [ 254.7] Finishing off
> > > 3.Then login rhv4.0 and try to import the guest from export domain to
> > data
> > > domain after finishing the conversion.But it is failed with the
> > > error:"Failed to import Vm rhel6u8 to Data Center Default, Cluster
> > Default"
> > > as appendix picture.
> > >
> > > -Additional info :
> > > I can import a guest from kvm or vmware.Only can not import this way.
> > >
> > >
> > > -
> > > Best Regards,
> > > Kun Wei
> > >
> > >
> > >
> >
>
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users


Re: [ovirt-users] Import guest at rhv4.0.6 failed after convert a regular ova file by virt-v2v

2017-01-05 Thread Yevgeny Zaspitsky
Seems like engines tries to validate a vNic with an empty MAC address and
fails on that. The fact that in the direct flow it works sounds weird to
me: either we loose the MAC in one flow or we do not validate it in the
other one.

Kun, could you please send (in private if you like) the OVF file from the
input tarball you've used, so we'd be able to debug the case more
thoroughly?

Regards,
Yevgeny

On Thu, Jan 5, 2017 at 10:03 AM, Kun Wei <ku...@redhat.com> wrote:

>
> Hi,shavivi.
>
> Thank you .Yes,as you know i can import ova directly and it can be import
> successful.
>
> But  i am testing v2v now  and  i want to confirm if it is a new bug.
>
>
>
> - Original Message -
> From: "Shahar Havivi" <shav...@redhat.com>
> To: "Kun Wei" <ku...@redhat.com>, "Yevgeny Zaspitsky" <yzasp...@redhat.com
> >
> Cc: "users" <users@ovirt.org>, "Tingting Zheng" <tzh...@redhat.com>,
> "Ming Xie" <m...@redhat.com>
> Sent: Thursday, January 5, 2017 3:05:45 PM
> Subject: Re: Import guest at rhv4.0.6 failed after convert a regular ova
> file by virt-v2v
>
> Yevgeny, please look why it fails on "isMacInRange"
>
> Kun, you know that you can import ova directly instead of using virt-v2v to
> export domain and then import to data domain,
> you can do that from the VMs tab->Import-> and change the source to OVA.
>
> On Thu, Jan 5, 2017 at 4:35 AM, Kun Wei <ku...@redhat.com> wrote:
>
> >
> > Hi,  I want to confirm a questions if it is a bug,or if something wrong
> > with me.Thanks
> >
> > When i convert a ova guest file to rhv4.0.6 by virt-v2v,but import is
> > failed.The attachment are engine.log and vdsm.log.
> >
> >
> >
> > Version:
> > libguestfs-1.32.7-3.el7_3.2.x86_64
> > virt-v2v-1.32.7-3.el7_3.2.x86_64
> > qemu-img-rhev-2.6.0-28.el7_3.2.x86_64
> > libvirt-2.0.0-10.el7_3.3.x86_64
> > rhv4.0: 4.0.6.3
> >
> > Steps :
> > 1:Prepare a regular ova file guest .
> > 2:Use virt-v2v to convert the ova file to rhv4.0.
> > # virt-v2v -i ova esx-rhel6.8-vmdkgz.tar -o rhev -os 10.73.131.93:
> /home/nfs_export
> > -of qcow2
> > [   0.0] Opening the source -i ova esx-rhel6.8-vmdkgz.tar
> > [  44.1] Creating an overlay to protect the source from being modified
> > [  45.0] Initializing the target -o rhev -os 10.73.131.93:
> /home/nfs_export
> > [  45.3] Opening the overlay
> > [  53.1] Inspecting the overlay
> > [  70.2] Checking for sufficient free disk space in the guest
> > [  70.2] Estimating space required on target for each disk
> > [  70.2] Converting Red Hat Enterprise Linux Server release 6.8
> (Santiago)
> > to run on KVM
> > virt-v2v: This guest has virtio drivers installed.
> > [ 215.6] Mapping filesystem data to avoid copying unused and blank areas
> > [ 216.0] Closing the overlay
> > [ 217.2] Checking if the guest needs BIOS or UEFI to boot
> > [ 217.2] Assigning disks to buses
> > [ 217.2] Copying disk 1/1 to /tmp/v2v.8kDoXb/908b80fa-09a6-
> > 4827-8e0a-a6a54520f685/images/47a1919e-1914-44b4-920f-
> > b8e208fa4bc1/da74f1c3-8f43-4a15-934a-bb9d380615c3 (qcow2)
> > (100.00/100%)
> > [ 254.6] Creating output metadata
> > [ 254.7] Finishing off
> > 3.Then login rhv4.0 and try to import the guest from export domain to
> data
> > domain after finishing the conversion.But it is failed with the
> > error:"Failed to import Vm rhel6u8 to Data Center Default, Cluster
> Default"
> > as appendix picture.
> >
> > -Additional info :
> > I can import a guest from kvm or vmware.Only can not import this way.
> >
> >
> > -
> > Best Regards,
> > Kun Wei
> >
> >
> >
>
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users


Re: [ovirt-users] ovirtmgmt network change

2016-12-21 Thread Yevgeny Zaspitsky
AFAIK if you use a DNS name of the host when you add it to an engine, then
its IP could be changed as long as that name continue be resolved to the
host new address.

On Wed, Dec 21, 2016 at 2:41 PM, Dan Kenigsberg  wrote:

> On Wed, Dec 21, 2016 at 12:41 AM, Bill Bill  wrote:
> > Hello,
> >
> >
> >
> > How can the IP of the hosts be changed? It is greyed out and I need to
> > change those. It’s the only thing I’m not able to change..
>
> I am afraid that you cannot change the address of the management
> network while the host is attached to Engine. The reason is that
> Engine always uses this address to communicate with the host.
>
> You could change it only if you remove the host from Engine, modify
> the address manually, and re-add the host using the new address.
> ___
> Users mailing list
> Users@ovirt.org
> http://lists.ovirt.org/mailman/listinfo/users
>
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users


Re: [ovirt-users] host does not comply with the cluster Default networks

2016-12-11 Thread Yevgeny Zaspitsky
I guess ovirtmgmt is your management network (which is required), then
engine tries to configure that on the host upon adding that. I assume that
the engine failed in that for the reason that you should be able to find in
the engine log.
Was ovirtmgmt network configured on the host prior adding it to the engine?
Could you please attach the log file?
Could you please tell what exact version are you using?

On Thu, Dec 8, 2016 at 2:37 AM, Mohmed Kasem  wrote:

> Dear all
> I have a problem in ovirt4.0 when I add host to engine i got erorr ( host
> does not comply with the cluster Default networks, the following networks
> are missing on host: 'ovirtmgmt)
> I facing this problem only in ovirt4.0 any help please
>
> ___
> Users mailing list
> Users@ovirt.org
> http://lists.phx.ovirt.org/mailman/listinfo/users
>
>
___
Users mailing list
Users@ovirt.org
http://lists.phx.ovirt.org/mailman/listinfo/users


Re: [ovirt-users] Assigned ipv6 vm addreses

2016-09-25 Thread Yevgeny Zaspitsky
Adding users list back (my fault ;-) )

On Sun, Sep 25, 2016 at 4:20 PM, Allen Warren <dallenwar...@gmail.com>
wrote:

> Thanks for the reply.  I figured it out yesterday.  I just edited the VM
> Custom Mac Address and it works like a champ now...
> I will setup that range and it should be good to go..  I am running oVirt
> 4.0..  oVirt is a nice product, beats the heck out of Xen Server..
>
> On Sun, Sep 25, 2016 at 6:57 AM, Yevgeny Zaspitsky <yzasp...@redhat.com>
> wrote:
>
>> oVirt does not produce IP addresses, but it allocates MAC addresses to
>> vNics on VMs it creates.
>> A MAC address serves as the base for an IPv6 address allocation (either
>> by IPv6 router advertisement for global addresses or by an operation system
>> for local addresses).
>>
>> BTW, 2 vNics with the same MAC address in the same LAN, would cause layer
>> 2 (Ethernet) malfunctioning too.
>>
>> In order to make the 2 oVirt's produce different MAC addresses you need
>> to adjust the definition of the MAC-pool ranges to be non-overlapping each
>> to another.
>> You could find MAC-pool definitions under a DC definitions in v4.0 and
>> under a cluster in v4.1.
>>
>> Regards,
>> Yevgeny
>>
>> On Sat, Sep 24, 2016 at 5:44 PM, Allen Warren <dallenwar...@gmail.com>
>> wrote:
>>
>>> I have two oVirt servers running as local servers on the same network.
>>> How do I prevent the servers from creating the same ipv6 address for two
>>> different vm's. If I start a vm on the first server and then start one on
>>> the second server the ipv6 addresses are the same..
>>>
>>> ___
>>> Users mailing list
>>> Users@ovirt.org
>>> http://lists.ovirt.org/mailman/listinfo/users
>>>
>>>
>>
>
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users


Re: [ovirt-users] Ovirt Node NG 4.0.0 wants to bridge my infiniband link

2016-07-04 Thread Yevgeny Zaspitsky
Infiniband should not be bridged.
Ovirt tries to build a bridge over a NIC if the network is a VM one. Please
verify that it isn't.

On Mon, Jul 4, 2016 at 5:10 PM, Giorgio Biacchi <gior...@di.unimi.it> wrote:

> As I said, Infiniband is not ethernet so it cannot be L2 bridged. This is
> normal
> and I understand why it can't be done. I'd like to know if there's a way to
> configure host networking for infiniband link without bridging because is
> not
> necessary to have a bridge there. In addition to this, 3.6 behavior was
> different. Here's all the relevant configuration files generated by 3.6
> and 4.0:
>
> - 3.6 ---
>
> [root@node3.6]# cat /etc/sysconfig/network-scripts/ifcfg-ib0
> # Generated by VDSM version 4.17.11-7.gitc0752ac.el7
> DEVICE=ib0
> HWADDR=80:00:00:48:fe:80:00:00:00:00:00:00:00:02:c9:03:00:0a:06:9f
> ONBOOT=yes
> BOOTPROTO=dhcp
> MTU=65520
> DEFROUTE=no
> NM_CONTROLLED=no
> IPV6INIT=no
>
> [root@node3.6]# cat /var/lib/vdsm/persistence/netconf/nets/ibstorage
> {
> "nic": "ib0",
> "bootproto": "dhcp",
> "blockingdhcp": true,
> "bridged": "false",
> "mtu": "65520"
> }
>
> -
>
> - 4.0 ---
>
> [root@node4.0]# cat /etc/sysconfig/network-scripts/ifcfg-ib0
> # Generated by VDSM version 4.18.5.1-1.el7.centos
> DEVICE=ib0
> BRIDGE=ibstorage
> ONBOOT=yes
> MTU=65520
> NM_CONTROLLED=no
> IPV6INIT=no
>
> [root@node4.0]# cat /var/lib/vdsm/persistence/netconf/nets/ibstorage
> {
> "ipv6autoconf": false,
> "nic": "ib0",
> "mtu": 65520,
> "switch": "legacy",
> "dhcpv6": false,
> "stp": false,
> "bridged": true,
> "defaultRoute": false
> }
>
> [root@node4.0]# cat /var/lib/vdsm/persistence/netconf/nets/ib0
> {
> "ipv6autoconf": false,
> "nic": "ib0",
> "mtu": 65520,
> "switch": "legacy",
> "dhcpv6": false,
> "stp": false,
> "bridged": true,
> "defaultRoute": false
> }
>
> [root@node4.0]# cat /etc/sysconfig/network-scripts/ifcfg-ibstorage
> # Generated by VDSM version 4.18.5.1-1.el7.centos
> DEVICE=ibstorage
> TYPE=Bridge
> DELAY=0
> STP=off
> ONBOOT=yes
> DEFROUTE=no
> NM_CONTROLLED=no
> IPV6INIT=no
> MTU=65520
>
> -
>
> So it seems that 4.0 generates more files and wants to bridge an ethernet
> unbridgeable link. Steps to attempt to configure the connection for 4.0
> was the
> same done successfully for 3.6.
>
> Thanks again
>
> On 07/04/2016 03:25 PM, Arman Khalatyan wrote:
> > As I remember the IBoIP is not fully implemented Ethernet protocol, it
> does not
> > support bridging.
> >
> >
> > ***
> >
> >  Dr. Arman Khalatyan  eScience -SuperComputing
> >  Leibniz-Institut für Astrophysik Potsdam (AIP)
> >  An der Sternwarte 16, 14482 Potsdam, Germany
> >
> > ***
> >
> > On Mon, Jul 4, 2016 at 2:58 PM, Yevgeny Zaspitsky <yzasp...@redhat.com
> > <mailto:yzasp...@redhat.com>> wrote:
> >
> > Hello,
> >
> > What's "couple of vdsm hooks"? Could you provide the source code of
> those?
> >
> > Regards,
> > Yevgeny
> >
> > On Mon, Jul 4, 2016 at 1:43 PM, Giorgio Biacchi <gior...@di.unimi.it
> > <mailto:gior...@di.unimi.it>> wrote:
> >
> > Hi list,
> > I'm trying a test installation of the new Ovirt Node Image and
> engine before
> > upgrading my 3.6 installation.
> >
> > In my current environment every hypervisor access the NFS shared
> data
> > domain via
> > infiniband. With a couple of vdsm hooks I was able to set the
> infiniband
> > link in
> > connected mode and set a MTU of 65520. In my case the infiniband
> link is
> > used as
> > a common NIC, I don't use RDMA, only IP over infiniband to
> access the
> > data domain.
> >
> > Now, when I attempt to setup host networking, the system try to
> bridge the
> > connection "ibstorage" (defined in networks tab) with ib0
> device. But a non
> > 

Re: [ovirt-users] Ovirt Node NG 4.0.0 wants to bridge my infiniband link

2016-07-04 Thread Yevgeny Zaspitsky
Hello,

What's "couple of vdsm hooks"? Could you provide the source code of those?

Regards,
Yevgeny

On Mon, Jul 4, 2016 at 1:43 PM, Giorgio Biacchi  wrote:

> Hi list,
> I'm trying a test installation of the new Ovirt Node Image and engine
> before
> upgrading my 3.6 installation.
>
> In my current environment every hypervisor access the NFS shared data
> domain via
> infiniband. With a couple of vdsm hooks I was able to set the infiniband
> link in
> connected mode and set a MTU of 65520. In my case the infiniband link is
> used as
> a common NIC, I don't use RDMA, only IP over infiniband to access the data
> domain.
>
> Now, when I attempt to setup host networking, the system try to bridge the
> connection "ibstorage" (defined in networks tab) with ib0 device. But a non
> ethernet device cannot be bridged that way and host networking setup fails.
>
> Any idea or suggestion? With 3.6 (node and engine) all was fine.
>
> Thanks
> --
> gb
>
> PGP Key: http://pgp.mit.edu/
> Primary key fingerprint: C510 0765 943E EBED A4F2 69D3 16CC DC90 B9CB 0F34
> ___
> Users mailing list
> Users@ovirt.org
> http://lists.ovirt.org/mailman/listinfo/users
>
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users


Re: [ovirt-users] Networking issues with oVirt Guest VM.

2016-07-04 Thread Yevgeny Zaspitsky
Adding mailing list back...

On Mon, Jul 4, 2016 at 3:38 PM, Yevgeny Zaspitsky <yzasp...@redhat.com>
wrote:

> Clint,
>
> Sorry, I missed that you already tried that.
>
> Here are my thoughts (some more shooting in the dark) after reading your
> description again:
>
>- You have quite complicate setup. IIUC, ovirt-engine and its host are
>vSphere VMs. Then, a kind of no-macspoof should be applied from the vSphere
>side. BTW, are both of them on the same vShepre host? Is DHCP server
>another VM on that host?
>- Where/how did you "turn on Port Mirroring"?
>- I'd start the troubleshooting by using tcpdump utility in order to
>pinpoint the component that blocks the traffic.
>- Did you try assigning a static IP instead of DHCP and then check
>connectivity? If that works, then the problem is on the DHCP sever side
>probably.
>- If you do not see any requests in the DHCP server log, then I guess,
>"dhclient -B" wouldn't help.
>- Please turn iptables/firewalld off.
>
>
> Regards,
> Yevgeny
>
> On Sun, Jul 3, 2016 at 9:06 PM, Yevgeny Zaspitsky <yzasp...@redhat.com>
> wrote:
>
>> Hello,
>>
>> IIUC using vdsm macspoof hook would help - reading [1] should help you
>> configuring that.
>>
>> [1] https://github.com/oVirt/vdsm/blob/master/vdsm_hooks/macspoof/README
>>
>> Hope that helps,
>> Yevgeny
>>
>> On Thu, Jun 30, 2016 at 6:11 AM, Clint Smith <clint.smith.m...@gmail.com>
>> wrote:
>>
>>> Hello,
>>>
>>> I have been experimenting with oVirt for the last couple of weeks and I
>>> must say it has a lot of nice features.  I really like it, however, I am
>>> having a heck of a time getting the guest networking all set up correctly.
>>> I am hoping that someone can give me a little guidance in figuring this
>>> out.  I apologize in advance if some of my terminology is off,  I am new.
>>>
>>> Here is a brief intro to my setup:
>>> I created a Centos 7 VM within a vSphere/ESXI environment and then
>>> installed ovirt-engine on it.  I also created another Centos 7 VM and set
>>> it up as a host.  I have configured the Cluster and Host via the oVirt
>>> Administration Portal.  For simplicity, I am using the default ovirtmgmt
>>> network as my only logical network, however I have tried several different
>>> schemes with no luck.  I have a DHCP server and a DNS server that are
>>> siblings to the oVirt host and the engine.  Both the engine and the host
>>> have been upgraded to version 4.0.
>>>
>>> The problem:
>>> My thought was that I would have the guest VMs on the oVirt host use my
>>> existing DHCP server to get their IP addresses, at least at first.  The
>>> problem I am having is that the DHCPACK is not making it back across the
>>> ovirtmgmt bridge and on to the guest.  If I tell dhclient(from the guest)
>>> to force a Broadcast (by using the –B option) on the DHCP server, it will
>>> work.  This is not a solution, just a clue.  Another clue is that ARP
>>> replies from the gateway don’t make it back to the machine, preventing
>>> pings even when I force the IP.  Lastly, If I turn on Port Mirroring,
>>> everything works fine, but it’s my understanding that this is only for
>>> debugging purposes.
>>>
>>> What I have tried (in no particular order):
>>>
>>>- Reading the docs
>>>- Turning on VLAN tagging.
>>>- Installing the mac-spoofing hook, making the configuration changes
>>>to the engine, and then turning it on in the VM config.  I also verified
>>>that the ‘filterref’ tag was removed using virsh.
>>>- Setting up a second logical network on a different subnet, and
>>>connecting it to an additional network interface that I added to the 
>>> host.
>>>On the host, I setup dnsmasq as a DNS and DHCP server.  I got this 
>>> working
>>>up to the point of having the same issues that I was having using the
>>>existing DHCP and DNS servers on the ovirtmgmt network.
>>>- I have tried various changes to iptables as well as the original
>>>settings as well as verified that ebtables is not blocking any traffic.  
>>> I
>>>did configure iptables for logging and noticed it was dropping some 
>>> traffic
>>>related to DHCP, however it seemed like it was DISCOVER or REQUEST 
>>> traffic
>>>due to the IN, OUT, SRC, and DST variables in the log.  I have viewed the
>&g

Re: [ovirt-users] Network redundancy with Manual balancing per VLAN

2016-06-26 Thread Yevgeny Zaspitsky
Dan, Edy,

Could you guys answer this?

IIUC, the requirements are:

   - stream the traffic of few VLANs(network roles) through a single bond
   - be able to bind a VLAN to a bond slave with an option of fallback
   - have redundancy
   - assign different QoS to every VLAN (my addition)

I guess this is a new RFC that we do not support currently, but would we be
able to provide in any future?

-- Forwarded message --
From: Fernando Frediani 
Date: Sat, Jun 25, 2016 at 11:17 PM
Subject: [ovirt-users] Network redundancy with Manual balancing per VLAN
To: users@ovirt.org


Hello,

In VMware it is possible to bond two network interfaces and for each
Portgroup (equivalent to a VLAN) is possible to tell which of the physical
interfaces underneath it you wish the traffic to flow primarily and which
stays as secondary(bond mode=1 equivalent). So for certain VLANs
(Management, Live Migration, etc) is possible to force traffic flow via one
physical NIC of the bond and for other VLANs (Virtual Machine's traffic)
outs via the other NIC with failover to each other should a cable or switch
fails.

This is specially good for better utilize the fewer NICs available and
still have redundancy.

In oVirt it is also possible to have bonds, but would it still be possible
to do that same and favor the traffic per VLAN basis ? I guess it is
something related to Linux Bond module but perhaps someone has done this
already.

Thanks

Fernando

___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users


Re: [ovirt-users] Network Interface order changed after reboot

2016-06-26 Thread Yevgeny Zaspitsky
BTW, what oVirt (vdsm) version do you use?

On Sun, Jun 26, 2016 at 3:15 PM, Yevgeny Zaspitsky <yzasp...@redhat.com>
wrote:

> Christoph,
>
> Are those nodes bare-metal or they're VMs?
>
> Regards,
> Yevgeny
>
> On Thu, Jun 23, 2016 at 3:49 PM, <ov...@timmi.org> wrote:
>
>> Hi List,
>>
>> I have two nodes (running CentOS 7) and the network interface order
>> changed for some interfaces after every reboot.
>>
>> The configurations are done through the oVirt GUI. So the ifcfg-ethX
>> scripts are configured automatically by VDSM.
>>
>> Is there any option to get this configured to be stable?
>>
>> Best regards and thank you
>>
>> Christoph
>>
>> ___
>> Users mailing list
>> Users@ovirt.org
>> http://lists.ovirt.org/mailman/listinfo/users
>>
>
>
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users


Re: [ovirt-users] Network Interface order changed after reboot

2016-06-26 Thread Yevgeny Zaspitsky
Christoph,

Are those nodes bare-metal or they're VMs?

Regards,
Yevgeny

On Thu, Jun 23, 2016 at 3:49 PM,  wrote:

> Hi List,
>
> I have two nodes (running CentOS 7) and the network interface order
> changed for some interfaces after every reboot.
>
> The configurations are done through the oVirt GUI. So the ifcfg-ethX
> scripts are configured automatically by VDSM.
>
> Is there any option to get this configured to be stable?
>
> Best regards and thank you
>
> Christoph
>
> ___
> Users mailing list
> Users@ovirt.org
> http://lists.ovirt.org/mailman/listinfo/users
>
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users


Re: [ovirt-users] adding network via ovirt-shell

2016-02-24 Thread Yevgeny Zaspitsky
Bill,

Here are the CLI equivalents for the GUI operations you have asked for:

   1. Assign network to a cluster:
   add network --name ${network_name} --parent-cluster-name ${cluster_name}
   2. After a network is attached to a cluster, make it required (or not
   required):
   update network --name ${network_name} --parent-cluster-name
   ${cluster_name} --expect [True|False]


Regards,
Yevgeny

On Wed, Feb 24, 2016 at 1:29 AM, Bill James  wrote:

> I kind of found the answer.
> Under Cluster -> Logical Networks -> Manager Networks
> there is a check box for "Assign" & "Required".
> Checking the Assign box makes it show up in "Setup Host Networks" dialog
> box.
>
> No idea how to check that box using the cli though.
>
>
>
> On 02/23/2016 10:54 AM, Bill James wrote:
>
>> I'm trying to add a network using ovirt-shell.
>> It adds fine to the GUI but I can't add the network to a host because it
>> isn't listed in the "Setup Host Networks" dialog.
>> Adding a network interface using the GUI is added fine to the dialog
>> window.
>> What am I missing?
>>
>> [oVirt shell (connected)]# add network --data_center-name Default --name
>> Vlan7 --description '10.176.7' --vlan-id 7
>>
>>
>> Also not sure how to add "Network Label" via cli. (adding it via GUI
>> doesn't make interface usable in dialog window)
>>
>> ovirt-engine-3.6.2.6-1.el7.centos.noarch
>>
>>
>> Problem networks are Vlan5 & Vlan 7.
>>
>>
>> [oVirt shell (connected)]# list networks
>>
>> id : 80b5bffb-afb7-4c14-b228-e505b6a93152
>> name   : Gluster-KS
>> description: kickstart
>>
>> id : ----0009
>> name   : ovirtmgmt
>> description: Management Network
>>
>> id : 7d9d55d9-1158-4a64-a2b9-07b763ae2b6d
>> name   : vlan1
>> description: 10.176.1
>>
>> id : e81540f9-99f2-4826-b483-b9c08031cdaa
>> name   : Vlan5
>> description: 10.176.5
>>
>> id : 6d1bf691-d438-48e2-a230-21360c609889
>> name   : Vlan6
>> description: 10.176.6
>>
>> id : 2c7db3f8-151f-4c5a-a88d-c840f7f2ce57
>> name   : Vlan7
>> description: 10.176.7
>>
>>
>> Cloud Services for Business www.j2.com
>> j2 | eFax | eVoice | FuseMail | Campaigner | KeepItSafe | Onebox
>>
>>
>> This email, its contents and attachments contain information from j2
>> Global, Inc. and/or its affiliates which may be privileged, confidential or
>> otherwise protected from disclosure. The information is intended to be for
>> the addressee(s) only. If you are not an addressee, any disclosure, copy,
>> distribution, or use of the contents of this message is prohibited. If you
>> have received this email in error please notify the sender by reply e-mail
>> and delete the original message and any copies. (c) 2015 j2 Global, Inc.
>> All rights reserved. eFax, eVoice, Campaigner, FuseMail, KeepItSafe, and
>> Onebox are registered trademarks of j2 Global, Inc. and its affiliates.
>> ___
>> Users mailing list
>> Users@ovirt.org
>> http://lists.ovirt.org/mailman/listinfo/users
>>
>
> ___
> Users mailing list
> Users@ovirt.org
> http://lists.ovirt.org/mailman/listinfo/users
>
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users


Re: [ovirt-users] [ovirt-devel] Welcome: New ovirt.org website!

2016-02-21 Thread Yevgeny Zaspitsky
Hi,

I updated my feature page [1] on the legacy site last week and added the
new picture file [2]. That's missing in the new site and github.
Looks like source/images/wiki folder was updated on github on Jun 19th last
time, which is quite long time ago...
IMHO it re-synced again...

[1] http://old.ovirt.org/Features/IPv6_support
[2] http://old.ovirt.org/File:Edit_network_ipv6.jpg

Regards,
Yevgeny

On Fri, Feb 19, 2016 at 5:57 PM, Mikey Ariel  wrote:

> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA256
>
> Following yesterday's pre-launch email[1], I'm happy to announce that
> the new ovirt.org[2] website is now live in **public beta** and
> available for community review.
>
> All the information in this email is also available in the first blog
> post of our shiny new oVirt blog[3]. Read on for information about new
> features, known issues, and a call for reviews.
>
> The old MediaWiki site is still available in read-only[4], and will be
> taken offline on **March 1, 2016**. This is to ensure that you can
> compare pages and review migrated content.
>
> What's New
> ==
>
> The new Website is full of improvements and enhancements, check out
> these highlights:
>
> * Source content is now formatted in Markdown instead of MediaWiki.
> This means that you can create and edit documentation, blog posts, and
> feature pages with the same Markdown syntax you know.
>
> * The Website is deployed with Middleman and stored on GitHub. This
> means that you can make changes to content with the same GitHub
> contribution workflow that you know (fork, clone, edit, commit, submit
> pull request). We even have an "Edit this page on GitHub" link at the
> bottom of every page!
>
> * New layout and design, from breadcrumbs to sidebards and an upgraded
> landing page. (This is still WIP, please see known issues.)
>
> * Automatic redirects from the old MediaWiki site. This means that if
> the wiki page exists in the new website, previously-released URLs will
> redirect to that page. If the page was removed, the Search page will
> open with the page title auto-filled in the search box.
>
> * Hierarchical content structure. This means that instead of flat
> Wiki-style files, the deployed Website reflects an organized source
> repo with content sorted into directories and sub-directories.
>
> * Official oVirt blog! Our new blog welcomes contributions. This means
> that if you solved a problem with oVirt, want to share your oVirt
> story, or describe a cool integration, you can submit a pull request
> with a blog post and we will provide editorial reviews and help
> publish your posts.
>
> * Standardized contribution process. The GitHub repo now includes a
> README.md[5] file that you can use to learn about how to add and edit
> content on the website. We welcome pull requests!
>
> Known Issues
> 
>
> Despite our best efforts, there are still a few kinks with the new
> website that you should be aware of:
>
> * Attempting to navigate to ovirt.org (without www.) leads to a
> redirect loop. We have a ticket open with OpenShift, our hosting
> service to fix this.
>
> * Only http is available. We also have a ticket with OpenShift to add
> SSL and enable https.
>
> * Home page and Download page are still being upgraded by our UX
> experts, expect some cool new changes soon!
>
> * Feature pages look-and-feel is still under construction. You can
> still edit and push feature pages as usual.
>
> What's Next
> ===
>
> Even though the Website is live, the work is hardly over. We'd like to
> ask for your help in:
>
> * Reviewing content for anything obsolete or outdated (each page in
> the new website includes a header toolbar with metadata from the
> original wiki page for your convenience)
>
> * Submitting blog posts or any other content that you wish to share
> with the oVirt community
>
> * Reporting bugs and proposing enhancements, for example broken links
> or missing pages
>
> We hope you will enjoy the new oVirt Website, looking forward to your
> feedback and contributions!
>
> [1] http://lists.ovirt.org/pipermail/devel/2016-February/012372.html
> [2] http://www.ovirt.org/
> [3] http://www.ovirt.org/blog/2016/02/welcome-to-new-ovirt-site/
> [4] http://old.ovirt.org/Home
> [5] https://github.com/oVirt/ovirt-site/blob/master/README.md
>
> - --
> Mikey Ariel
> Community Lead, oVirt
> www.ovirt.org
>
> "To be is to do" (Socrates)
> "To do is to be" (Jean-Paul Sartre)
> "Do be do be do" (Frank Sinatra)
>
> Mobile: +420-702-131-141
> IRC: mariel / thatdocslady
> Twitter: @ThatDocsLady
>
>
> -BEGIN PGP SIGNATURE-
> Version: GnuPG v2
>
> iQEcBAEBCAAGBQJWxztZAAoJEHYPPTOszxHoOF4H/2UeK9JTspw4dYU+pUu8fKqf
> pSrE9ha3k9KpPnGIAx6VTnbXolNr9tHFiT8gcds1C9CVoMEPrKUxns9CTz3kPYBH
> /mGYCwQ+ncUNBCn+54uiKmXssaAtpVLnuxh8f1z9BDUoFjoY7ASLTzwXu9XTuH9V
> C9gaR0iGcRgSmBPjfPuAVWuHgimM068x5Ig2BoMSzAeCM7pO6LwiRzt3xX15xlR4
> TMkUWLHdnJ1uL1GH267tMofrMnk25veDPIzuzIaPFjwIuTFgtwnUQI82ULbxcet2
> 

Re: [ovirt-users] Logical Networks with different MTUs

2015-10-14 Thread Yevgeny Zaspitsky
Hello Cam,

The problem you described seems to be similar to the described in
https://bugzilla.redhat.com/show_bug.cgi?id=1242227.
The mentioned bug was fixed in one of the recent oVirt builds.
The updated Ovirt code allows to mix networks (regardless of them being VM
or non-VM) on a single NIC with the following conditions:

   - not more than one non-VLAN network in conjunction with any number of
   VLANs
   - MTU of the VLAN networks have to be equal to that of a non-VLAN one if
   such exists

So I'd advise you to update you oVirt to the latest build.

Regards,
Yevgeny

On Tue, Oct 6, 2015 at 2:47 AM, Cam Wright 
wrote:

> Hi,
>
> I've just started building an oVirt farm as a replacement for our
> existing infrastructure and I'm having issues configuring multiple
> VLAN bridges with different MTUs.
>
> When I try to add a logical network with a VLAN tag + VM Network + MTU
> 9000 to a bonded interface on my host with only the ovirtmgmt network
> (MTU 1500, default) it throws up an error message and won't let me
> continue, error message below:
>
> Cannot setup Networks. The following Logical Networks do not have the
> same MTU value: ovirtmgmt(default),
> net_bne_vlan10(9000).
>
> Eventually, I will want to add more logical networks with the default
> MTU, so setting the ovirtmgmt MTU to 9000 is unfortunately not going
> to solve my issue.
>
> This is something I can do in vanilla Linux networking where I set the
> 802.3ad bond to be MTU 9000 and configure two different bridges off
> that interface to 1500 and 9000, however this does not seem to be
> possible in the oVirt web interface.
>
> I've dug through the Administration guide and feature requests and
> there doesn't seem to be anything there around using different MTU
> values for different logical networks.
>
> Has anyone been able to successfully do this? The
> /etc/sysconfig/network-scripts and /var/lib/vdsm/persistence/netconf
> files seem to get rewritten whenever the vdsm service does anything,
> so I'm reluctant to edit them manually.
>
> The hosts were built using the hosted-engine installer on Fedora 22.
> The engine is running Fedora 22 also, see below:
>
> [root@bne-ovirtengine01 ~]# cat /etc/redhat-release
> Fedora release 22 (Twenty Two)
> [root@bne-ovirtengine01 ~]# uname -r
> 4.0.4-301.fc22.x86_64
>
> oVirt Engine Version: 3.6.0-1.fc22
>
> Any assistance would be appreciated.
>
> Cam Wright | Systems Administrator
> CUTTINGEDGE
> 90 Victoria St, West End, Brisbane, QLD, 4101
> T +61 7 3013 6222
> E cwri...@cuttingedge.com.au | W www.cuttingedge.com.au
>
> /SYD /BNE /MEL /TYO
> ___
> Users mailing list
> Users@ovirt.org
> http://lists.ovirt.org/mailman/listinfo/users
>
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users


Re: [ovirt-users] [ovirt-devel] Feature Page: Mac Pool per DC

2014-04-27 Thread Yevgeny Zaspitsky
Now for users@ovirt.org indeed.

- Original Message -
From: Yevgeny Zaspitsky yzasp...@redhat.com
To: Martin Mucha mmu...@redhat.com
Cc: us...@ovrit.org, de...@ovirt.org
Sent: Sunday, April 27, 2014 2:29:46 PM
Subject: Re: [ovirt-devel] Feature Page: Mac Pool per DC

Martin,

I'd like to propose a different approach on how the ranges to be defined and 
stored.

Discussing this feature with Moti raised the alternative UX design: 
Defining ranges could be added as a left-tab on create DC dialog and a sub-tab 
on an existing DC. It would be a table of start and end address fields and we 
can add a calculated # of MACs in the range and/or summary for the DC. 
Also that will make string parsing unneeded, prevent possible user mistakes in 
the string format and make possible validating every field of the range on the 
UI side easier.
As you can see on the screenshot you've attached even a single range doesn't 
fit to the text box. In case of multiple ranges managing them in a single line 
textbox would be very uncomfortable.

A range is an object with at least 2 members (start and end). And we have few 
of these for each data center.
Storing a collection of the objects in a single field in a relational DB seems 
a bit awkward to me. 
That has few disadvantages:
1. is not normalized
2. make data validation nearly impossible
3. make querying the data very difficult
4. is restraining our ability to extend the object (e.g. a user might like to 
give a description to a range)
So IMHO a satellite table with the FK to storage_pool would be a more robust 
design.

Best regards, 
 
Yevgeny Zaspitsky 
Senior Software Engineer 
Red Hat Israel 


- Original Message -
From: Martin Mucha mmu...@redhat.com
To: users@ovirt.org, de...@ovirt.org
Sent: Thursday, April 10, 2014 9:59:44 AM
Subject: [ovirt-devel] new feature

Hi,

I'd like to notify you about new feature, which allows to specify distinct MAC 
pools, currently one per data center.
http://www.ovirt.org/Scoped_MacPoolManager

any comments/proposals for improvement are very welcomed.
Martin.
___
Devel mailing list
de...@ovirt.org
http://lists.ovirt.org/mailman/listinfo/devel
___
Devel mailing list
de...@ovirt.org
http://lists.ovirt.org/mailman/listinfo/devel
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users