Re: [ovirt-users] upgrade 3.5 to 3.6 auto import storage domain failed

[Резников А . А .]

14.03.2017 11:40, Резников А.А. пишет:

Hi list!

During the upgrade ovirt from 3.5 to 3.6, I get errors:
"VDSM command failed: Can not acquire host id: 
(u'8905c9ac-d892-478d-8346-63b8fa1c5763 ', SanlockException (22,' 
Sanlock lockspace add failure ',' Invalid argument '))",
"Failed to attach Storage Domain hosted_storage to Data Center 
Default. (User: SYSTEM)",

"Failed to import the Hosted Engine Storage Domain".

Thus in my data storage nfs OVF_STORE disks do not appear.

The following messages appear in vdsm.log:

Thread-2806 :: DEBUG :: 2017-03-14 10: 40: 29,569 :: task :: 827 :: 
Storage.TaskManager.Task: :( resourceAcquired) Task = 
`cd294427-d6fa-454b-9289-6660ad228c8f`: : _resourcesAcquired: 
Storage.8905c9ac-d8

92-478d-8346-63b8fa1c5763 (shared)
Thread-2806 :: DEBUG :: 2017-03-14 10: 40: 29,569 :: task :: 993 :: 
Storage.TaskManager.Task :: (_ decref) Task = 
`cd294427-d6fa-454b-9289-6660ad228c8f`: : Ref 1 aborting False
Thread-2806 :: ERROR :: 2017-03-14 10: 40: 29,569 :: task :: 866 :: 
Storage.TaskManager.Task :: (_ setError) Task = 
`cd294427-d6fa-454b-9289-6660ad228c8f`: : Unexpected error

Traceback (most recent last call last):
  File "/usr/share/vdsm/storage/task.py", line 873, in _run
Return fn (* args, ** kargs)
  File "/usr/share/vdsm/logUtils.py", line 49, in wrapper
Res = f (* args, ** kwargs)
  File "/usr/share/vdsm/storage/hsm.py", line 3315, in getImagesList
Images = dom.getAllImages ()
  File "/usr/share/vdsm/storage/fileSD.py", line 373, in getAllImages
Self.getPools () [0],
IndexError: list index out of range



3 ', Clearing records.


Does anyone have any idea how this can be fixed?

Thanks, Alex.


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



Any body help me?

Thanks, Alex.

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

Re: [ovirt-users] Error on Node upgrade 2

[Yuval Turgeman]

Adding Ryan, Fernando, can you please attach /tmp/imgbased.log ?

On Mar 13, 2017 2:43 PM, "FERNANDO FREDIANI" 
wrote:

> Hi Yedidyah.
> Running oVirt-Node *4.1.0--0.20170201.0+1* on the nodes and tried a
> normal upgrade. It detected it has to upgrade to
> *4.1.1-0.3.rc1.20170303133657.git20d3594.el7.centos* from the
> ovirt-4.1-pre repository.
>
> The upgrade finished with the following problems:
>
> ...
> Running transaction
>   Updating   : ovirt-node-ng-image-4.1.1-0.3.
> rc1.20170303133657.git20d3594.el7.centos.noarch
> 1/4
>   Updating   : ovirt-node-ng-image-update-4.1.1-0.3.rc1.20170303133657.
> git20d3594.el7.centos.noarch   2/4
> mount: special device /dev/onn_kvm01/ovirt-node-ng-4.1.1-0.20170303.0+1
> does not exist
> cp: target ‘/tmp/tmp.N6JgSdFcu6/usr/share/imgbased/’ is not a directory
> rm: cannot remove ‘/tmp/tmp.N6JgSdFcu6/usr/share/imgbased/*image-update*.rpm’:
> No such file or directory
> umount: /tmp/tmp.N6JgSdFcu6: not mounted
> ...
>
> So it seems it was not able to do it correctly.
> This is the second time it happens and I had to remove manually these
> packages containing 4.1.1 version.
>
> Tried also:
>
> # lvdisplay | grep ovirt-node-ng-4.1.1-0.20170303.0+1
>   LV Path/dev/onn_kvm01/ovirt-node-ng-4.1.1-0.20170303.0+1
>   LV Nameovirt-node-ng-4.1.1-0.20170303.0+1
> # fdisk -l /dev/onn_kvm01/ovirt-node-ng-4.1.1-0.20170303.0+1
> fdisk: cannot open /dev/onn_kvm01/ovirt-node-ng-4.1.1-0.20170303.0+1: No
> such file or directory
> # vgchange -ay
> # fdisk -l /dev/onn_kvm01/ovirt-node-ng-4.1.1-0.20170303.0+1 (then worked)
> But the upon reboot it came back to 4.1.0 as if 4.1.1 never existed
>
> Fernando
>
> On 12/03/2017 03:30, Yedidyah Bar David wrote:
>
> On Fri, Mar 10, 2017 at 2:37 PM, FERNANDO FREDIANI 
>  wrote:
>
> I am not sure if another email I sent went through but has anyone got
> problems when upgrading a running oVirt-node-ng from 4.1.0 to 4.1.1.
>
>
> What kind of problems?
>
>
>
> Is the only solution a complete reinstall of the node ?
>
>
> No, this should work.
>
> Best,
>
>
>
> Thanks
>
> Fernando
>
> ___
> Users mailing listUsers@ovirt.orghttp://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] Suggestion for configuration with only 2 nics and equallogic ISCSI

[Yaniv Kaul]

On Tue, Mar 14, 2017, 10:13 PM Gianluca Cecchi 
wrote:

> Il 14 Mar 2017 20:54, "Yaniv Kaul"  ha scritto:
>
> On Tue, Mar 14, 2017, 9:46 PM Gianluca Cecchi 
> wrote:
>
> On Tue, Mar 14, 2017 at 6:30 PM, Pavel Gashev  wrote:
>
> Gianluca,
>
>
>
> Are you sure your 10Gbit nics doesn’t support iSCSI offload?
>
> - I remeber that at 5.5 / 6beta times even when configuring offload iSCSI
> you still needed to configure an ip on the interfaces in OS, because I seem
> to rememebr the discovery phase (or login perhaps?) happened on tcp and not
> capable in offload. Is this still true or can I only configure an ip for
> the offload macs?
>
>
> oVirt sees it as FC. No need for IP.
> Y.
>
>
> Hi Yanuv, can you elaborate on this? I have not understood your comment...
> Can I configure chap on storage side?
>

Depends on your card (I'd be surprised if it can't though)
oVirt sees the LUNs already connected, just like FC.
Y.


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

Re: [ovirt-users] Suggestion for configuration with only 2 nics and equallogic ISCSI

[Gianluca Cecchi]

Il 14 Mar 2017 20:54, "Yaniv Kaul"  ha scritto:



On Tue, Mar 14, 2017, 9:46 PM Gianluca Cecchi 
wrote:

> On Tue, Mar 14, 2017 at 6:30 PM, Pavel Gashev  wrote:
>
> Gianluca,
>
>
>
> Are you sure your 10Gbit nics doesn’t support iSCSI offload?
>
> - I remeber that at 5.5 / 6beta times even when configuring offload iSCSI
> you still needed to configure an ip on the interfaces in OS, because I seem
> to rememebr the discovery phase (or login perhaps?) happened on tcp and not
> capable in offload. Is this still true or can I only configure an ip for
> the offload macs?
>

oVirt sees it as FC. No need for IP.
Y.


Hi Yanuv, can you elaborate on this? I have not understood your comment...
Can I configure chap on storage side?

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

Re: [ovirt-users] Suggestion for configuration with only 2 nics and equallogic ISCSI

[Yaniv Kaul]

On Tue, Mar 14, 2017, 9:46 PM Gianluca Cecchi 
wrote:

> On Tue, Mar 14, 2017 at 6:30 PM, Pavel Gashev  wrote:
>
> Gianluca,
>
>
>
> Are you sure your 10Gbit nics doesn’t support iSCSI offload?
>
>
>
>
> I Pavel, thanks for the comment.
> It could be a nice way to proceed ;-) Are you using it with oVirt?
>
> Some considerations/questions:
>
> - in the past I tested iSCSI offload with RH EL 5.5 (it was late 2010)
> with some problems; then I also tested with RH EL 6 when still in beta with
> better results. Not tested more anddidn't go to production as it was not
> considered crucial. Can we consider it stable in RH EL / CentOS 7.3?
>
> - I remeber that at 5.5 / 6beta times even when configuring offload iSCSI
> you still needed to configure an ip on the interfaces in OS, because I seem
> to rememebr the discovery phase (or login perhaps?) happened on tcp and not
> capable in offload. Is this still true or can I only configure an ip for
> the offload macs?
>

oVirt sees it as FC. No need for IP.
Y.


> - is the offload configuration transparent to oVirt in the sense that
> oVirt only needs that the hosts see the multipath iSCSI devices, or do I
> need to make any particular config in oVirt too, then?
>
> Thanks in advance,
>
> Gianluca
>
> ___
> 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] Suggestion for configuration with only 2 nics and equallogic ISCSI

[Gianluca Cecchi]

On Tue, Mar 14, 2017 at 6:30 PM, Pavel Gashev  wrote:

> Gianluca,
>
>
>
> Are you sure your 10Gbit nics doesn’t support iSCSI offload?
>
>
>

I Pavel, thanks for the comment.
It could be a nice way to proceed ;-) Are you using it with oVirt?

Some considerations/questions:

- in the past I tested iSCSI offload with RH EL 5.5 (it was late 2010) with
some problems; then I also tested with RH EL 6 when still in beta with
better results. Not tested more anddidn't go to production as it was not
considered crucial. Can we consider it stable in RH EL / CentOS 7.3?

- I remeber that at 5.5 / 6beta times even when configuring offload iSCSI
you still needed to configure an ip on the interfaces in OS, because I seem
to rememebr the discovery phase (or login perhaps?) happened on tcp and not
capable in offload. Is this still true or can I only configure an ip for
the offload macs?

- is the offload configuration transparent to oVirt in the sense that oVirt
only needs that the hosts see the multipath iSCSI devices, or do I need to
make any particular config in oVirt too, then?

Thanks in advance,

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

Re: [ovirt-users] assigned Host was not available

[Bill James]

We found if we move the rebuilt host to a new cluster setup for 4.1 
compatibility it could startup VMs just fine.

So we are guessing its just a 3.6 compat mode issue.
I'll try upgrading the rest of the nodes to 4.1 and change compat mode.
Hopefully all will just work then.
We'll see this weekend

Thanks.


On 03/14/2017 02:41 AM, Bill James wrote:
I'm not sure if its related or not but I noticed 3 other hosts are 
doing similar things.
They all have in common that they are earlier hardware nodes from the 
other 3 nodes in the cluster.
All are HP DL360's, but that later 3 have 256 GB ram, older ones have 
128GB ram.

The other 3 nodes are still running ovirt 3.6.4.


On 3/12/17 11:57 PM, Bill James wrote:
only thing different on the VM config is that I says Deploy to host 
ovirt4.test, which ovirt then ignores because "host is not available".


I attached logs from ovirt engine and ovirt4.test.
VM I tried to start is called "billj4-2".
I also tried to migrate another VM that is suppose to run on ovirt4 
called "billj4".

This time it successfully migrated and is running on ovrit4.test.

Thank you for your help!


[root@ovirt4 test vdsm]# virsh -r list --all
 IdName   State

 2 billj4 running


On 3/12/17 6:16 AM, Yaniv Kaul wrote:



On Tue, Mar 7, 2017 at 9:50 PM Bill James > wrote:


I have a hardware node that I recently upgraded from 3.6.4 to
4.1.0.4-1.
It is having issues that certain VMs I cannot deploy to it
because ovirt
says host is not available.
However that host is status up and has 14 running VMs on it.

How do I tell why ovirt thinks the host is not available?


Anything different on the VM's configuration?
If some VMs do run on it, look like the issue is with the VM 
configuration. Can you provide full logs?

Y.


Attached is output from "vdsClient -s ovirt4.test.j2noc.com
 getVdsStats"


vdsm-4.19.4-1.el7.centos.x86_64
ovirt-engine-4.1.0.4-1.el7.centos.noarch


engine.log:
2017-03-07 11:06:49,269-08 WARN
[org.ovirt.engine.core.dal.dbbroker.auditloghandling.AuditLogDirector]
(ForkJo
inPool-1-worker-0) [] EVENT_ID: USER_RUN_VM_ON_NON_DEFAULT_VDS(152),
Correlation ID: 81490f5b-3029-44d4-b223-df
6bb506efb3, Job ID: 76900dc9-ed9a-4fda-a17a-31825aa515c9, Call
Stack:
null, Custom Event ID: -1, Message: Guest
  billj4 started on Host ovirt7.test.j2noc.com
. (Default Host parameter
was ignored - assigned Host was not available).


___
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 3.5 / Virtual Machine / 10GB possible?

[Yaniv Kaul]

On Tue, Mar 14, 2017, 7:50 PM Devin Acosta  wrote:

>
> Yaniv,
>
> So are you telling me that the virtual machine will get 10Gbit speeds even
> though it only reports 1Gbit?
>

Yes, theoretically. The historical reason is that when it was developed,
there was a need to display some link speed (ethtool, Windows device
manager) and back then 1g seemed like high speed.

Getting really 10g is a challenge, but I have seen 7-8g on a bright day.
Y.


> On Tue, Mar 14, 2017 at 10:43 AM, Yaniv Kaul  wrote:
>
>
>
> On Tue, Mar 14, 2017, 7:04 PM Devin Acosta  wrote:
>
>
> We are running oVIRT 3.5, until one of vendors supports us moving to 4.0.
> Our RHV hosts have 20GB aggregated NICs on the box, however our VM's I
> notice only have 1GB NICs available. I do see that the Virtual machine is
> using the "virtio-net-pci" device, is there a way to get this Virtual
> Machine to have a 10GB NIC? When i run 'atop' on the RHV host I notice that
> it claims the virtual NIC is hitting 100% utilization at times.
>
>
> Disregard that number. It's a virtual NIC, and so is the 1g number.
> Y.
>
>
> What methods do I have available to increase the 1GB NIC to something
> higher hopefully to 10GB ?
>
> --
>
> Devin Acosta
> Red Hat Certified Architect, LinuxStack
> 602-354-1220 <(602)%20354-1220> || de...@linuxguru.co
> ___
> Users mailing list
> Users@ovirt.org
> http://lists.ovirt.org/mailman/listinfo/users
>
>
>
>
> --
>
> Devin Acosta
> Red Hat Certified Architect, LinuxStack
> 602-354-1220 || de...@linuxguru.co
>
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] oVIRT 3.5 / Virtual Machine / 10GB possible?

[Devin Acosta]

Yaniv,

So are you telling me that the virtual machine will get 10Gbit speeds even
though it only reports 1Gbit?

On Tue, Mar 14, 2017 at 10:43 AM, Yaniv Kaul  wrote:

>
>
> On Tue, Mar 14, 2017, 7:04 PM Devin Acosta  wrote:
>
>>
>> We are running oVIRT 3.5, until one of vendors supports us moving to 4.0.
>> Our RHV hosts have 20GB aggregated NICs on the box, however our VM's I
>> notice only have 1GB NICs available. I do see that the Virtual machine is
>> using the "virtio-net-pci" device, is there a way to get this Virtual
>> Machine to have a 10GB NIC? When i run 'atop' on the RHV host I notice that
>> it claims the virtual NIC is hitting 100% utilization at times.
>>
>
> Disregard that number. It's a virtual NIC, and so is the 1g number.
> Y.
>
>
>> What methods do I have available to increase the 1GB NIC to something
>> higher hopefully to 10GB ?
>>
>> --
>>
>> Devin Acosta
>> Red Hat Certified Architect, LinuxStack
>> 602-354-1220 <(602)%20354-1220> || de...@linuxguru.co
>> ___
>> Users mailing list
>> Users@ovirt.org
>> http://lists.ovirt.org/mailman/listinfo/users
>>
>


-- 

Devin Acosta
Red Hat Certified Architect, LinuxStack
602-354-1220 || de...@linuxguru.co
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

[ovirt-users] hosted engine without the appliance?

[Devin A. Bougie]

Hi, All.  Is it still possible or supported to run a hosted engine without 
using the oVirt Engine Appliance?  In other words, to install our own OS on a 
VM and have it act as a hosted engine?  "hosted-engine --deploy" now seems to 
insist on using the oVirt Engine Appliance, but if it's possible and not a 
mistake we'd prefer to run and manage our own OS.

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

Re: [ovirt-users] oVIRT 3.5 / Virtual Machine / 10GB possible?

[Yaniv Kaul]

On Tue, Mar 14, 2017, 7:04 PM Devin Acosta  wrote:

>
> We are running oVIRT 3.5, until one of vendors supports us moving to 4.0.
> Our RHV hosts have 20GB aggregated NICs on the box, however our VM's I
> notice only have 1GB NICs available. I do see that the Virtual machine is
> using the "virtio-net-pci" device, is there a way to get this Virtual
> Machine to have a 10GB NIC? When i run 'atop' on the RHV host I notice that
> it claims the virtual NIC is hitting 100% utilization at times.
>

Disregard that number. It's a virtual NIC, and so is the 1g number.
Y.


> What methods do I have available to increase the 1GB NIC to something
> higher hopefully to 10GB ?
>
> --
>
> Devin Acosta
> Red Hat Certified Architect, LinuxStack
> 602-354-1220 || de...@linuxguru.co
> ___
> 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] Suggestion for configuration with only 2 nics and equallogic ISCSI

[Pavel Gashev]

Gianluca,

Are you sure your 10Gbit nics doesn’t support iSCSI offload?

From:  on behalf of Gianluca Cecchi 

Date: Tuesday, 14 March 2017 at 14:16
To: users 
Subject: [ovirt-users] Suggestion for configuration with only 2 nics and 
equallogic ISCSI

Hello,
I'm going to configure an environment where I would have:
- 2 x 1Gbit/s nics that I have to dedicate to ovirt mgmt network only, due to 
design constraints

- 2 x 10Gbit/s nics where I have to configure all other VLANS

In particular I have to define:
1) a couple of VLANs for VMs
2) VLAN for live migration (can I defined more that 1 of this kind?)
3) VLAN for iSCSI against an EQL system

EQL guys keep saying that the only supported way to connect is through 2 
different adapters and using multipath then. The iSCSI lan is on VLAN100 and I 
have to use it.

This brings in a problem in the way I can configure logical networks in oVirt 
and a scenario where I create bond on top of vlan and not the reverse as I 
normally do

The desiderata would be something like this, having the two 10Gbit nics the 
names eth2 and eth3 and vlan100 the one for iSCSI and vlan65 the one for VMs, 
for example:



eth2  --- eth2.100 -- iSCSI IP 1

\ eth2.65 ---

  \ __ bond65 -- IP

eth3  --- eth3.100 -- iSCSI IP 2  /

\ eth3.65 ---
Questions:
- does this work at all in Linux in general? I saw something with bonding 
option use_carrier=1
- in that case, can I then configure oVirt to use this configuration for 
networks?
- is LACP supporto on switch an enabler or a stopper...?

Other suggestions?

Thanks in advance,
Gianluca
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] Scheduled major outage for several centos services and impact on oVirt project

[Simone Tiraboschi]

On Tue, Mar 14, 2017 at 5:34 PM, Simone Tiraboschi 
wrote:

> Hi,
> as announced and scheduled, several centos services are going to be out of
> service for about 48 hours since March 14th, 13:00 UTC:
> https://seven.centos.org/2017/03/infra-scheduled-major-
> outage-for-several-services/
>
> The list of involved services includes https://cbs.centos.org (Koji).
> Our stable release tracker rpm ( http://resources.ovirt.org/
> pub/yum-repo/ovirt-release41.rpm ) requires packages from there and so
> oVirt from stable release is currently not installable as well on Centos
> based systems.
>
> [centos-ovirt41-candidate]
> name=CentOS-7 - oVirt 4.1
> baseurl=http://cbs.centos.org/repos/virt7-ovirt-41-candidate/$basearch/os/
> gpgcheck=0
> enabled=1
>
> [centos-ovirt-common-candidate]
> name=CentOS-7 - oVirt common
> baseurl=http://cbs.centos.org/repos/virt7-ovirt-common-
> candidate/$basearch/os/
> gpgcheck=0
> enabled=1
>
> We are trying to find a recovery solution.
>

We are working to build a new release tracker rpm for 4.1.

In the mean time the quick fix is to manually move from cbs server to
mirror repos editing
/etc/yum.repos.d/ovirt-4.1-dependencies.repo
changing it from:
[centos-ovirt41-candidate]
name=CentOS-7 - oVirt 4.1
baseurl=http://cbs.centos.org/repos/virt7-ovirt-41-candidate/$basearch/os/
gpgcheck=0
enabled=1

[centos-ovirt-common-candidate]
name=CentOS-7 - oVirt common
baseurl=
http://cbs.centos.org/repos/virt7-ovirt-common-candidate/$basearch/os/
gpgcheck=0
enabled=1

to:
[ovirt-centos-ovirt41]
name=CentOS-$releasever - oVirt 4.1
baseurl=
http://mirror.centos.org/centos/$releasever/virt/$basearch/ovirt-4.1/
gpgcheck=1
enabled=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-SIG-Virtualization

[centos-opstools-release]
name=CentOS-7 - OpsTools - release
baseurl=http://mirror.centos.org/centos/$releasever/opstools/$basearch/
gpgcheck=1
enabled=1
skip_if_unavailable=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-SIG-OpsTools
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

[ovirt-users] oVIRT 3.5 / Virtual Machine / 10GB possible?

[Devin Acosta]

We are running oVIRT 3.5, until one of vendors supports us moving to 4.0.
Our RHV hosts have 20GB aggregated NICs on the box, however our VM's I
notice only have 1GB NICs available. I do see that the Virtual machine is
using the "virtio-net-pci" device, is there a way to get this Virtual
Machine to have a 10GB NIC? When i run 'atop' on the RHV host I notice that
it claims the virtual NIC is hitting 100% utilization at times.

What methods do I have available to increase the 1GB NIC to something
higher hopefully to 10GB ?

-- 

Devin Acosta
Red Hat Certified Architect, LinuxStack
602-354-1220 || de...@linuxguru.co
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

[ovirt-users] Scheduled major outage for several centos services and impact on oVirt project

[Simone Tiraboschi]

Hi,
as announced and scheduled, several centos services are going to be out of
service for about 48 hours since March 14th, 13:00 UTC:
https://seven.centos.org/2017/03/infra-scheduled-major-outage-for-several-services/

The list of involved services includes https://cbs.centos.org (Koji).
Our stable release tracker rpm (
http://resources.ovirt.org/pub/yum-repo/ovirt-release41.rpm ) requires
packages from there and so oVirt from stable release is currently
not installable as well on Centos based systems.

[centos-ovirt41-candidate]
name=CentOS-7 - oVirt 4.1
baseurl=http://cbs.centos.org/repos/virt7-ovirt-41-candidate/$basearch/os/
gpgcheck=0
enabled=1

[centos-ovirt-common-candidate]
name=CentOS-7 - oVirt common
baseurl=
http://cbs.centos.org/repos/virt7-ovirt-common-candidate/$basearch/os/
gpgcheck=0
enabled=1

We are trying to find a recovery solution.
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] oVirt 4 and 10GbE NICs

[Darrell Budic]

> On Mar 14, 2017, at 7:54 AM, FERNANDO FREDIANI  
> wrote:
> 
> Isn't the traffic shown on the dashboard based in 1Gbps always, even if the 
> hosts have 10Gb interfaces ?
> 

Yep, all dirt interfaces show as 1Gb.

> Is there anywhere in oVirt config files or Database that you can tell to the 
> dashboard to consider 10Gb instead of 1Gb for those cases ?
> 
> 

Not that I know of, but it doesn’t affect that available performance, it’s just 
visible.

I’ve gotten ~3.5Gbps out of iperf with no appreciable tuning to a VM, so it’s 
definitely possible to get more speed out of them.

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

Re: [ovirt-users] ovirt 4.1 selfhosted engine deployment

[Yedidyah Bar David]

On Tue, Mar 14, 2017 at 3:56 PM, Lauri-Alo Adamson
 wrote:
> Hello!
>
> Thanks  - try to remember put cc
>
> More information:
> Our server has separated management Network interfaces (nic) - in HP server i 
> use ILO web managem interface to access server remote console. (HP iLO 
> Integrated Remote Console)
> I used this renote console to access this freezed node and i was no abel to  
> ping  or access ohter host in the same network. Then  open this remote 
> console i saw many  "vcpu unhandled rdmsr" messages.
> My node had GbE nic interface with ip address that was not used by ovirtmagmt 
> interface. I was no tabel ping or access this interfiace then node freezed.

OK. Please try first using virsh/virt-manager/virt-install directly,
to find out if it's a bug in kvm etc. Best,

>
> Lauri-Alo Adamson
>
>
>
>
> -Original Message-
> From: Yedidyah Bar David [mailto:d...@redhat.com]
> Sent: Tuesday, March 14, 2017 3:36 PM
> To: Lauri-Alo Adamson 
> Subject: Re: [ovirt-users] ovirt 4.1 selfhosted engine deployment
>
> Hi,
>
> Please keep the list in Cc, so that others can try to help, and/or learn. 
> Thanks.
>
> On Tue, Mar 14, 2017 at 3:14 PM, Lauri-Alo Adamson 
>  wrote:
>> Hello !
>>
>> Additional clarifications:
>>
>> Host/node freezed after the hosted engine virtuaal machine tried first
>> time to start. After that i was no tabel to Access node over ssh Whole node 
>> freezed. Only after reboot i was abel to access thise host again  over ssh.
>
> You said you saw something on the console. I thought you meant you managed to 
> login.
> If you didn't, then indeed the problem wasn't (probably) only in networking, 
> but something deeper, in the kernel probably.
>
> Best,
>
>>
>> Lauri-Alo Adamson
>>
>> -Original Message-
>> From: Yedidyah Bar David [mailto:d...@redhat.com]
>> Sent: Tuesday, March 14, 2017 3:04 PM
>> To: Lauri-Alo Adamson 
>> Cc: users@ovirt.org
>> Subject: Re: [ovirt-users] ovirt 4.1 selfhosted engine deployment
>>
>> On Tue, Mar 14, 2017 at 2:48 PM, Lauri-Alo Adamson 
>>  wrote:
>>> Hello!
>>>
>>>
>>>
>>> I tried to install/deploy ovirt 4.1 self hosted engine  with
>>> glusterfs (thee node cluster two HP 380D Gen9 server with Intel(R)
>>> Xeon(R) CPU
>>> E5-2650 v3 @ 2.30GHz and one IBM 3650 M4 ) on Centos 7.3.1611 x86-64 .
>>> But then the hosted engine
>>>
>>> virtuaal machine is starting  then whole  host (node)  freezes –
>>> server is not accessible over network with ssh. Engine virtuaal
>>> machines fails to start and deploiment fails. Same time the on server
>>> console and server log apper lines „vcpu  unhandled rdmsr“ . I
>>> searched Google this problem and dis not found the solution.
>>>
>>>
>>>
>>> There was talk about  adding  /etc/modprobe.d/kvm.conf  „options kvm
>>> ignore_msrs=1“  Which is the same as „echo 1 >
>>> /sys/module/kvm/parameters/ignore_msrs“
>>>
>>> http://wiki.stoney-cloud.org/wiki/Workaround_unhandled_rdmsr/wrmsr
>>>
>>>
>>>
>>> I tried but these . Sadly they  do not work. Has any one found
>>> solution to the described probleem.
>>
>> Are you sure your problem isn't a simple networking problem? At what stage 
>> does it die? Do you run it under 'screen'? If so, you can reattach to it 
>> ('screen -r -d') and see where you are. Also please check/post the setup 
>> logs (/var/log/ovirt-hosted-engine-setup/*).
>>
>> If indeed your problem is kernel/kvm related, then I have no idea.
>> Does starting a kvm vm using virsh/virt-install work?
>>
>> Good luck,
>>
>>>
>>>
>>>
>>> Lauri-Alo Adamson
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> ___
>>> Users mailing list
>>> Users@ovirt.org
>>> http://lists.ovirt.org/mailman/listinfo/users
>>>
>>
>>
>>
>> --
>> Didi
>
>
>
> --
> Didi



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

Re: [ovirt-users] ovirt 4.1 selfhosted engine deployment

[Lauri-Alo Adamson]

Hello!

Thanks  - try to remember put cc 

More information:
Our server has separated management Network interfaces (nic) - in HP server i 
use ILO web managem interface to access server remote console. (HP iLO 
Integrated Remote Console)
I used this renote console to access this freezed node and i was no abel to  
ping  or access ohter host in the same network. Then  open this remote console 
i saw many  "vcpu unhandled rdmsr" messages.
My node had GbE nic interface with ip address that was not used by ovirtmagmt 
interface. I was no tabel ping or access this interfiace then node freezed.

Lauri-Alo Adamson




-Original Message-
From: Yedidyah Bar David [mailto:d...@redhat.com] 
Sent: Tuesday, March 14, 2017 3:36 PM
To: Lauri-Alo Adamson 
Subject: Re: [ovirt-users] ovirt 4.1 selfhosted engine deployment

Hi,

Please keep the list in Cc, so that others can try to help, and/or learn. 
Thanks.

On Tue, Mar 14, 2017 at 3:14 PM, Lauri-Alo Adamson  
wrote:
> Hello !
>
> Additional clarifications:
>
> Host/node freezed after the hosted engine virtuaal machine tried first 
> time to start. After that i was no tabel to Access node over ssh Whole node 
> freezed. Only after reboot i was abel to access thise host again  over ssh.

You said you saw something on the console. I thought you meant you managed to 
login.
If you didn't, then indeed the problem wasn't (probably) only in networking, 
but something deeper, in the kernel probably.

Best,

>
> Lauri-Alo Adamson
>
> -Original Message-
> From: Yedidyah Bar David [mailto:d...@redhat.com]
> Sent: Tuesday, March 14, 2017 3:04 PM
> To: Lauri-Alo Adamson 
> Cc: users@ovirt.org
> Subject: Re: [ovirt-users] ovirt 4.1 selfhosted engine deployment
>
> On Tue, Mar 14, 2017 at 2:48 PM, Lauri-Alo Adamson 
>  wrote:
>> Hello!
>>
>>
>>
>> I tried to install/deploy ovirt 4.1 self hosted engine  with 
>> glusterfs (thee node cluster two HP 380D Gen9 server with Intel(R) 
>> Xeon(R) CPU
>> E5-2650 v3 @ 2.30GHz and one IBM 3650 M4 ) on Centos 7.3.1611 x86-64 .
>> But then the hosted engine
>>
>> virtuaal machine is starting  then whole  host (node)  freezes – 
>> server is not accessible over network with ssh. Engine virtuaal 
>> machines fails to start and deploiment fails. Same time the on server 
>> console and server log apper lines „vcpu  unhandled rdmsr“ . I 
>> searched Google this problem and dis not found the solution.
>>
>>
>>
>> There was talk about  adding  /etc/modprobe.d/kvm.conf  „options kvm 
>> ignore_msrs=1“  Which is the same as „echo 1 > 
>> /sys/module/kvm/parameters/ignore_msrs“
>>
>> http://wiki.stoney-cloud.org/wiki/Workaround_unhandled_rdmsr/wrmsr
>>
>>
>>
>> I tried but these . Sadly they  do not work. Has any one found 
>> solution to the described probleem.
>
> Are you sure your problem isn't a simple networking problem? At what stage 
> does it die? Do you run it under 'screen'? If so, you can reattach to it 
> ('screen -r -d') and see where you are. Also please check/post the setup logs 
> (/var/log/ovirt-hosted-engine-setup/*).
>
> If indeed your problem is kernel/kvm related, then I have no idea.
> Does starting a kvm vm using virsh/virt-install work?
>
> Good luck,
>
>>
>>
>>
>> Lauri-Alo Adamson
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> ___
>> Users mailing list
>> Users@ovirt.org
>> http://lists.ovirt.org/mailman/listinfo/users
>>
>
>
>
> --
> Didi



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

Re: [ovirt-users] ovirt 4.1 selfhosted engine deployment

[Yedidyah Bar David]

On Tue, Mar 14, 2017 at 2:48 PM, Lauri-Alo Adamson
 wrote:
> Hello!
>
>
>
> I tried to install/deploy ovirt 4.1 self hosted engine  with glusterfs (thee
> node cluster two HP 380D Gen9 server with Intel(R) Xeon(R) CPU E5-2650 v3 @
> 2.30GHz and one IBM 3650 M4 ) on Centos 7.3.1611 x86-64 . But then the
> hosted engine
>
> virtuaal machine is starting  then whole  host (node)  freezes – server is
> not accessible over network with ssh. Engine virtuaal machines fails to
> start and deploiment fails. Same time the on server console and server log
> apper lines „vcpu  unhandled rdmsr“ . I searched Google this problem and dis
> not found the solution.
>
>
>
> There was talk about  adding  /etc/modprobe.d/kvm.conf  „options kvm
> ignore_msrs=1“  Which is the same as „echo 1 >
> /sys/module/kvm/parameters/ignore_msrs“
>
> http://wiki.stoney-cloud.org/wiki/Workaround_unhandled_rdmsr/wrmsr
>
>
>
> I tried but these . Sadly they  do not work. Has any one found solution to
> the described probleem.

Are you sure your problem isn't a simple networking problem? At what
stage does it die? Do you run it under 'screen'? If so, you can
reattach to it ('screen -r -d') and see where you are. Also please
check/post the setup logs (/var/log/ovirt-hosted-engine-setup/*).

If indeed your problem is kernel/kvm related, then I have no idea.
Does starting a kvm vm using virsh/virt-install work?

Good luck,

>
>
>
> Lauri-Alo Adamson
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> ___
> Users mailing list
> Users@ovirt.org
> http://lists.ovirt.org/mailman/listinfo/users
>



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

Re: [ovirt-users] oVirt 4 and 10GbE NICs

[FERNANDO FREDIANI]

Isn't the traffic shown on the dashboard based in 1Gbps always, even if 
the hosts have 10Gb interfaces ?


Is there anywhere in oVirt config files or Database that you can tell to 
the dashboard to consider 10Gb instead of 1Gb for those cases ?


Thanks

Fernando


On 14/03/2017 07:57, Davide Ferrari wrote:

Hello

the oVirt hosts in my cluster have all 10GbE interfaces (bonded 2 by 
2), and I was wondering if a single Linux VM with a virtio interface 
can reach for example 8Gbps.


I'm asking because I'm seeing a 100% network usage in some VMs in 
ovirt VM dashboard but the real traffic in that VM is between 1-2Gbps.


Thanks

--
Davide Ferrari
Senior Systems Engineer


___
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

[ovirt-users] ovirt 4.1 selfhosted engine deployment

[Lauri-Alo Adamson]

Hello!

I tried to install/deploy ovirt 4.1 self hosted engine  with glusterfs (thee 
node cluster two HP 380D Gen9 server with Intel(R) Xeon(R) CPU E5-2650 v3 @ 
2.30GHz and one IBM 3650 M4 ) on Centos 7.3.1611 x86-64 . But then the hosted 
engine
virtuaal machine is starting  then whole  host (node)  freezes - server is not 
accessible over network with ssh. Engine virtuaal machines fails to start and 
deploiment fails. Same time the on server console and server log apper lines 
"vcpu  unhandled rdmsr" . I searched Google this problem and dis not found the 
solution.

There was talk about  adding  /etc/modprobe.d/kvm.conf  "options kvm 
ignore_msrs=1"  Which is the same as "echo 1 > 
/sys/module/kvm/parameters/ignore_msrs"
http://wiki.stoney-cloud.org/wiki/Workaround_unhandled_rdmsr/wrmsr



I tried but these . Sadly they  do not work. Has any one found solution to the 
described probleem.



Lauri-Alo Adamson











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

Re: [ovirt-users] Maintenance GUI vs CLI

[Yedidyah Bar David]

On Tue, Mar 14, 2017 at 2:31 PM, Marcin Kruk  wrote:
> So if the maintenance mode from the GUI is necessary to upgrade host, how to
> launch it from the CLI on that host?

Generally speaking, it should be doable using the api/sdk.
I don't know the details. If you have many hosts, you might
want to try the upgrade manager:

http://www.ovirt.org/develop/release-management/features/engine/upgrademanager/

>
> 2017-03-14 11:30 GMT+01:00 Yedidyah Bar David :
>>
>> On Tue, Mar 14, 2017 at 12:12 PM, Marcin Kruk 
>> wrote:
>> > What is the difference between:
>>
>> There are two different notions of host maintenance:
>>
>> 1. in the engine, meaning the engine will migrate away VMs from
>> this host, not start new ones on it, etc. This applies to all
>> hosts, not just hosted-engine ones
>>
>> 2. in ovirt-hosted-engine-ha, the high availability daemons.
>> Here it means similar things, but applies only to the hosted
>> engine vm, and is maintained in the HE shared storage (not in
>> the engine db).
>>
>> > host CLI command:  hosted-engine --vm-maintenance --mode=local
>>
>> This one does (2.).
>>
>> > and
>> > RHVEM gui click action:  Hosts ->  -> Maintenance?
>>
>> In the past, this did only (1.), but now it does both.
>>
>> See also:
>>
>>
>> https://www.ovirt.org/develop/release-management/features/engine/self-hosted-engine-maintenance-flows/
>> https://bugzilla.redhat.com/show_bug.cgi?id=1047649
>> https://bugzilla.redhat.com/show_bug.cgi?id=1277646
>>
>> Best,
>> --
>> Didi
>
>



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

Re: [ovirt-users] Maintenance GUI vs CLI

[Marcin Kruk]

So if the maintenance mode from the GUI is necessary to upgrade host, how
to launch it from the CLI on that host?

2017-03-14 11:30 GMT+01:00 Yedidyah Bar David :

> On Tue, Mar 14, 2017 at 12:12 PM, Marcin Kruk 
> wrote:
> > What is the difference between:
>
> There are two different notions of host maintenance:
>
> 1. in the engine, meaning the engine will migrate away VMs from
> this host, not start new ones on it, etc. This applies to all
> hosts, not just hosted-engine ones
>
> 2. in ovirt-hosted-engine-ha, the high availability daemons.
> Here it means similar things, but applies only to the hosted
> engine vm, and is maintained in the HE shared storage (not in
> the engine db).
>
> > host CLI command:  hosted-engine --vm-maintenance --mode=local
>
> This one does (2.).
>
> > and
> > RHVEM gui click action:  Hosts ->  -> Maintenance?
>
> In the past, this did only (1.), but now it does both.
>
> See also:
>
> https://www.ovirt.org/develop/release-management/features/
> engine/self-hosted-engine-maintenance-flows/
> https://bugzilla.redhat.com/show_bug.cgi?id=1047649
> https://bugzilla.redhat.com/show_bug.cgi?id=1277646
>
> Best,
> --
> Didi
>
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] Serious Trouble - Lost Domains

[JC Clark]

Thank you for your response.

My storage is NFS storage.

I will try with "qemu-img info" ..
 
I made an attempt to import the domain and I got an error.  I will attempt to replicate tonight.


Do I have to alter any of the metafiles when importing?  I seemed to remember 
it didn't like to import another data center's domain.


Sincerely JC


Date: Tue, 14 Mar 2017 10:16:17 +0100
From: Arman Khalatyan<arm2...@gmail.com>
To: JC Clark<j...@saipantech.com>
Cc: users<users@ovirt.org>
Subject: Re: [ovirt-users] Serious Trouble - Lost Domains
Message-ID:
<caaqdm6ap1ydv4thra+z4h-2k3o9vq8qp4pawq7vshymeegx...@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"

What kind of storage are you using?
If you check images with "qemu-img info" are you able to see the
filesystems?
Can you simply import the domain unto the new ovirt?

Am 14.03.2017 9:59 vorm. schrieb "JC Clark"<j...@saipantech.com>:


Dear Fellows and Fellettes,

I am having serious disaster problems.  After a power transformer outside
the building literally exploded,  surged and fried the motherboard to the
main SPM domain computer in my Ovirt 4.0 system. After getting the computer
working again with a new mother board.  I managed to basically loose
integrity of the old engine.  I have rebuilt the engine.

I have backups of the old engine and 4 data domains an ISO domain and an
export domain which appear to not be damaged. They are all accessible from
the CL.
  I had to create a new host and SPM.  How do I get the floating domains
into the new Engine?

I have 1 storage container (85GB) I must get back.  MUST GET BACK!!

I really appreciate you reading my sob story.  Hope you can help..

Thank You much


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


-- next part --
An HTML attachment was scrubbed...
URL:<http://lists.ovirt.org/pipermail/users/attachments/20170314/4e4ee976/attachment.html>


--
Warm Regards,
 
JC Clark

IT Director
Mount Carmel School
Saipan, CNMI
(670) 235-6272 office
(670) 483-8684 cell

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

[ovirt-users] Suggestion for configuration with only 2 nics and equallogic ISCSI

[Gianluca Cecchi]

Hello,
I'm going to configure an environment where I would have:
- 2 x 1Gbit/s nics that I have to dedicate to ovirt mgmt network only, due
to design constraints

- 2 x 10Gbit/s nics where I have to configure all other VLANS

In particular I have to define:
1) a couple of VLANs for VMs
2) VLAN for live migration (can I defined more that 1 of this kind?)
3) VLAN for iSCSI against an EQL system

EQL guys keep saying that the only supported way to connect is through 2
different adapters and using multipath then. The iSCSI lan is on VLAN100
and I have to use it.

This brings in a problem in the way I can configure logical networks in
oVirt and a scenario where I create bond on top of vlan and not the reverse
as I normally do

The desiderata would be something like this, having the two 10Gbit nics the
names eth2 and eth3 and vlan100 the one for iSCSI and vlan65 the one for
VMs, for example:


eth2  --- eth2.100 -- iSCSI IP 1
\ eth2.65 ---
  \ __ bond65 -- IP
eth3  --- eth3.100 -- iSCSI IP 2  /
\ eth3.65 ---

Questions:
- does this work at all in Linux in general? I saw something with bonding
option use_carrier=1
- in that case, can I then configure oVirt to use this configuration for
networks?
- is LACP supporto on switch an enabler or a stopper...?

Other suggestions?

Thanks in advance,
Gianluca
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] oVirt 4 and 10GbE NICs

[Yaniv Kaul]

On Tue, Mar 14, 2017 at 12:57 PM Davide Ferrari  wrote:

> Hello
>
> the oVirt hosts in my cluster have all 10GbE interfaces (bonded 2 by 2),
> and I was wondering if a single Linux VM with a virtio interface can reach
> for example 8Gbps.
>

Are you able from the host, first of all?


>
> I'm asking because I'm seeing a 100% network usage in some VMs in ovirt VM
> dashboard but the real traffic in that VM is between 1-2Gbps.
>

You might be able to, but it's not easy. You'll need to use NUMA and set
correctly the vCPUs pinning, perform correct tuning (multiple queues) and
so on.  With SR-IOV I assume there's a better chance (lower latency), btw.

Traffic shown in the dashboard is averaged, is it a sustained 8Gbps or
bursty?
Y.


> Thanks
>
> --
> Davide Ferrari
> Senior Systems Engineer
> ___
> 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

[ovirt-users] oVirt 4 and 10GbE NICs

[Davide Ferrari]

Hello

the oVirt hosts in my cluster have all 10GbE interfaces (bonded 2 by 2),
and I was wondering if a single Linux VM with a virtio interface can reach
for example 8Gbps.

I'm asking because I'm seeing a 100% network usage in some VMs in ovirt VM
dashboard but the real traffic in that VM is between 1-2Gbps.

Thanks

-- 
Davide Ferrari
Senior Systems Engineer
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] Maintenance GUI vs CLI

[Yedidyah Bar David]

On Tue, Mar 14, 2017 at 12:12 PM, Marcin Kruk  wrote:
> What is the difference between:

There are two different notions of host maintenance:

1. in the engine, meaning the engine will migrate away VMs from
this host, not start new ones on it, etc. This applies to all
hosts, not just hosted-engine ones

2. in ovirt-hosted-engine-ha, the high availability daemons.
Here it means similar things, but applies only to the hosted
engine vm, and is maintained in the HE shared storage (not in
the engine db).

> host CLI command:  hosted-engine --vm-maintenance --mode=local

This one does (2.).

> and
> RHVEM gui click action:  Hosts ->  -> Maintenance?

In the past, this did only (1.), but now it does both.

See also:

https://www.ovirt.org/develop/release-management/features/engine/self-hosted-engine-maintenance-flows/
https://bugzilla.redhat.com/show_bug.cgi?id=1047649
https://bugzilla.redhat.com/show_bug.cgi?id=1277646

Best,
-- 
Didi
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

[ovirt-users] Maintenance GUI vs CLI

[Marcin Kruk]

What is the difference between:
host CLI command:  hosted-engine --vm-maintenance --mode=local
and
RHVEM gui click action:  Hosts ->  -> Maintenance?
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] assigned Host was not available

[Bill James]

I'm not sure if its related or not but I noticed 3 other hosts are doing 
similar things.
They all have in common that they are earlier hardware nodes from the 
other 3 nodes in the cluster.
All are HP DL360's, but that later 3 have 256 GB ram, older ones have 
128GB ram.

The other 3 nodes are still running ovirt 3.6.4.


On 3/12/17 11:57 PM, Bill James wrote:
only thing different on the VM config is that I says Deploy to host 
ovirt4.test, which ovirt then ignores because "host is not available".


I attached logs from ovirt engine and ovirt4.test.
VM I tried to start is called "billj4-2".
I also tried to migrate another VM that is suppose to run on ovirt4 
called "billj4".

This time it successfully migrated and is running on ovrit4.test.

Thank you for your help!


[root@ovirt4 test vdsm]# virsh -r list --all
 IdName   State

 2 billj4 running


On 3/12/17 6:16 AM, Yaniv Kaul wrote:



On Tue, Mar 7, 2017 at 9:50 PM Bill James > wrote:


I have a hardware node that I recently upgraded from 3.6.4 to
4.1.0.4-1.
It is having issues that certain VMs I cannot deploy to it
because ovirt
says host is not available.
However that host is status up and has 14 running VMs on it.

How do I tell why ovirt thinks the host is not available?


Anything different on the VM's configuration?
If some VMs do run on it, look like the issue is with the VM 
configuration. Can you provide full logs?

Y.


Attached is output from "vdsClient -s ovirt4.test.j2noc.com
 getVdsStats"


vdsm-4.19.4-1.el7.centos.x86_64
ovirt-engine-4.1.0.4-1.el7.centos.noarch


engine.log:
2017-03-07 11:06:49,269-08 WARN
[org.ovirt.engine.core.dal.dbbroker.auditloghandling.AuditLogDirector]
(ForkJo
inPool-1-worker-0) [] EVENT_ID: USER_RUN_VM_ON_NON_DEFAULT_VDS(152),
Correlation ID: 81490f5b-3029-44d4-b223-df
6bb506efb3, Job ID: 76900dc9-ed9a-4fda-a17a-31825aa515c9, Call Stack:
null, Custom Event ID: -1, Message: Guest
  billj4 started on Host ovirt7.test.j2noc.com
. (Default Host parameter
was ignored - assigned Host was not available).


___
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] Serious Trouble - Lost Domains

[Arman Khalatyan]

What kind of storage are you using?
If you check images with "qemu-img info" are you able to see the
filesystems?
Can you simply import the domain unto the new ovirt?

Am 14.03.2017 9:59 vorm. schrieb "JC Clark" :

> Dear Fellows and Fellettes,
>
> I am having serious disaster problems.  After a power transformer outside
> the building literally exploded,  surged and fried the motherboard to the
> main SPM domain computer in my Ovirt 4.0 system. After getting the computer
> working again with a new mother board.  I managed to basically loose
> integrity of the old engine.  I have rebuilt the engine.
>
> I have backups of the old engine and 4 data domains an ISO domain and an
> export domain which appear to not be damaged. They are all accessible from
> the CL.
>  I had to create a new host and SPM.  How do I get the floating domains
> into the new Engine?
>
> I have 1 storage container (85GB) I must get back.  MUST GET BACK!!
>
> I really appreciate you reading my sob story.  Hope you can help..
>
> Thank You much
>
>
> ___
> 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

[ovirt-users] Serious Trouble - Lost Domains

[JC Clark]

Dear Fellows and Fellettes,

I am having serious disaster problems.  After a power transformer 
outside the building literally exploded,  surged and fried the 
motherboard to the main SPM domain computer in my Ovirt 4.0 system. 
After getting the computer working again with a new mother board.  I 
managed to basically loose integrity of the old engine.  I have rebuilt 
the engine.


I have backups of the old engine and 4 data domains an ISO domain and an 
export domain which appear to not be damaged. They are all accessible 
from the CL.
 I had to create a new host and SPM.  How do I get the floating domains 
into the new Engine?


I have 1 storage container (85GB) I must get back.  MUST GET BACK!!

I really appreciate you reading my sob story.  Hope you can help..

Thank You much


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

[ovirt-users] upgrade 3.5 to 3.6 auto import storage domain failed

[Резников А . А .]

Hi list!

During the upgrade ovirt from 3.5 to 3.6, I get errors:
"VDSM command failed: Can not acquire host id: 
(u'8905c9ac-d892-478d-8346-63b8fa1c5763 ', SanlockException (22,' 
Sanlock lockspace add failure ',' Invalid argument '))",
"Failed to attach Storage Domain hosted_storage to Data Center Default. 
(User: SYSTEM)",

"Failed to import the Hosted Engine Storage Domain".

Thus in my data storage nfs OVF_STORE disks do not appear.

The following messages appear in vdsm.log:

Thread-2806 :: DEBUG :: 2017-03-14 10: 40: 29,569 :: task :: 827 :: 
Storage.TaskManager.Task: :( resourceAcquired) Task = 
`cd294427-d6fa-454b-9289-6660ad228c8f`: : _resourcesAcquired: 
Storage.8905c9ac-d8

92-478d-8346-63b8fa1c5763 (shared)
Thread-2806 :: DEBUG :: 2017-03-14 10: 40: 29,569 :: task :: 993 :: 
Storage.TaskManager.Task :: (_ decref) Task = 
`cd294427-d6fa-454b-9289-6660ad228c8f`: : Ref 1 aborting False
Thread-2806 :: ERROR :: 2017-03-14 10: 40: 29,569 :: task :: 866 :: 
Storage.TaskManager.Task :: (_ setError) Task = 
`cd294427-d6fa-454b-9289-6660ad228c8f`: : Unexpected error

Traceback (most recent last call last):
  File "/usr/share/vdsm/storage/task.py", line 873, in _run
Return fn (* args, ** kargs)
  File "/usr/share/vdsm/logUtils.py", line 49, in wrapper
Res = f (* args, ** kwargs)
  File "/usr/share/vdsm/storage/hsm.py", line 3315, in getImagesList
Images = dom.getAllImages ()
  File "/usr/share/vdsm/storage/fileSD.py", line 373, in getAllImages
Self.getPools () [0],
IndexError: list index out of range
Thread-2806 :: DEBUG :: 2017-03-14 10: 40: 29,569 :: task :: 885 :: 
Storage.TaskManager.Task :: (_ run) Task = 
`cd294427-d6fa-454b-9289-6660ad228c8f`: : Task._run: 
cd294427-d6fa-454b-9289-6660ad228c8f ('8

905c9ac-d892-478d-8346-63b8fa1c5763 ',) {} failed - stopping task
Thread-2806 :: DEBUG :: 2017-03-14 10: 40: 29,570 :: task :: 1246 :: 
Storage.TaskManager.Task: :( stop) Task = 
`cd294427-d6fa-454b-9289-6660ad228c8f`: : Stopping in state preparing 
(force False)
Thread-2806 :: DEBUG :: 2017-03-14 10: 40: 29,570 :: task :: 993 :: 
Storage.TaskManager.Task :: (_ decref) Task = 
`cd294427-d6fa-454b-9289-6660ad228c8f`: : Ref 1 aborting True
Thread-2806 :: INFO :: 2017-03-14 10: 40: 29,570 :: task :: 1171 :: 
Storage.TaskManager.Task: :( prepare) Task = 
`cd294427-d6fa-454b-9289-6660ad228c8f`: : Aborting: Task is aborted: 
u'list index out of r

Ange '- code 100
Thread-2806 :: DEBUG :: 2017-03-14 10: 40: 29,570 :: task :: 1176 :: 
Storage.TaskManager.Task: :( prepare) Task = 
`cd294427-d6fa-454b-9289-6660ad228c8f`: : Prepare: aborted: list index 
out of range
Thread-2806 :: DEBUG :: 2017-03-14 10: 40: 29,570 :: task :: 993 :: 
Storage.TaskManager.Task :: (_ decref) Task = 
`cd294427-d6fa-454b-9289-6660ad228c8f`: : Ref 0 aborting True
Thread-2806 :: DEBUG :: 2017-03-14 10: 40: 29,570 :: task :: 928 :: 
Storage.TaskManager.Task :: (_ doAbort) Task = 
`cd294427-d6fa-454b-9289-6660ad228c8f`: : Task._doAbort: force False
Thread-2806 :: DEBUG :: 2017-03-14 10: 40: 29,571 :: resourceManager :: 
980 :: Storage.ResourceManager.Owner: :( cancelAll) Owner.cancelAll 
requests {}
Thread-2806 :: DEBUG :: 2017-03-14 10: 40: 29,571 :: task :: 595 :: 
Storage.TaskManager.Task :: (_ updateState) Task = 
`cd294427-d6fa-454b-9289-6660ad228c8f`: : Moving from state preparing -> 
state aborti

Ng
Thread-2806 :: DEBUG :: 2017-03-14 10: 40: 29,571 :: task :: 550 :: 
Storage.TaskManager.Task :: (__ state_aborting) Task = 
`cd294427-d6fa-454b-9289-6660ad228c8f`: : _aborting: recover policy none
Thread-2806 :: DEBUG :: 2017-03-14 10: 40: 29,571 :: task :: 595 :: 
Storage.TaskManager.Task :: (_ updateState) Task = 
`cd294427-d6fa-454b-9289-6660ad228c8f`: : Moving from state aborting -> 
state failed
Thread-2806 :: DEBUG :: 2017-03-14 10: 40: 29,571 :: resourceManager :: 
943 :: Storage.ResourceManager.Owner: :( releaseAll) Owner.releaseAll 
requests {} resources {'Storage.8905c9ac-d892 -478d-8346-6
3b8fa1c5763 ': }
Thread-2806 :: DEBUG :: 2017-03-14 10: 40: 29,571 :: resourceManager :: 
980 :: Storage.ResourceManager.Owner: :( cancelAll) Owner.cancelAll 
requests {}
Thread-2806 :: DEBUG :: 2017-03-14 10: 40: 29,572 :: resourceManager :: 
619 :: Storage.ResourceManager: :( releaseResource) Trying to release 
resource 'Storage.8905c9ac-d892-478d-8346-63b8fa1c5763 '
Thread-2806 :: DEBUG :: 2017-03-14 10: 40: 29,572 :: resourceManager :: 
638 :: Storage.ResourceManager: :( releaseResource) Released resource 
'Storage.8905c9ac-d892-478d-8346-63b8fa1c5763' ( 0 active

Users)
Thread-2806 :: DEBUG :: 2017-03-14 10: 40: 29,572 :: resourceManager :: 
644 :: Storage.ResourceManager: :( releaseResource) Resource 
'Storage.8905c9ac-d892-478d-8346-63b8fa1c5763' is free , Finding ou

T if anyone is waiting for it.
Thread-2806 :: DEBUG :: 2017-03-14 10: 40: 29,572 :: resourceManager :: 
652 :: Storage.ResourceManager: :( releaseResource) 

Re: [ovirt-users] Can I post email with screen shots as attachment ?

[Yaniv Dary]

It should work. What error are you getting?

Yaniv Dary
Technical Product Manager
Red Hat Israel Ltd.
34 Jerusalem Road
Building A, 4th floor
Ra'anana, Israel 4350109

Tel : +972 (9) 7692306
8272306
Email: yd...@redhat.com
IRC : ydary


On Mon, Mar 13, 2017 at 7:44 AM, John Joseph  wrote:

> Hi,
> Like to introduce myself, I am Joseph John some years back I had tried
> oVirt, now after very long time I am trying to setup oVirt and trying to
> prepare for demo.
> Was not active in the mailing list for very long time, now I am facing
> issues with ovirt, like to post the experience, eg I would like to show
> that my connect to CD drive option drop down is not working, so Can I send
> a screen shot attached along with the email
> Thanks
> Joseph John
>
>
> ___
> 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] idempiere

[Martin Perina]

Thanks for reporting this in https://bugzilla.redhat.com/1431882

Until this is fixed please use suggested workaround by adding '*' prefix:

Vms: *idempiere

Thanks

Martin Perina

On Tue, Mar 14, 2017 at 7:53 AM, Yedidyah Bar David  wrote:

> On Mon, Mar 13, 2017 at 8:22 PM, Logan Kuhn 
> wrote:
> > We have 4 idempiere-* VMs in an older cluster and we can search for VMs
> > based on all sorts of strings, but if it contains idempiere it shows:
> >
> > 2017-03-13 11:27:33,268 INFO  [org.ovirt.engine.core.bll.SearchQuery]
> > (default task-19) [] ResourceManager::searchBusinessObjects - erroneous
> > search text - ''Vms: idempiere''
> >
> > Can anyone else duplicate this problem in their environment?
>
> Duplicated here.
>
> It seems like entering "Vms: id" triggers the syntax-based search
> (expecting
> that you'll search by "id"). So if you decide to open a bug, it can be
> something
> like: "Free text search fails if starting with a keyword". The same
> happens if
> I search e.g. "Vms: ipa" - it seems to expect me to search by ip address.
>
> Adding Ravi.
>
> For your current needs, you can use: "Vms: name=*idempiere*".
>
> Best,
>
> >
> > Logan
> >
> > ___
> > Users mailing list
> > Users@ovirt.org
> > http://lists.ovirt.org/mailman/listinfo/users
> >
>
>
>
> --
> Didi
> ___
> 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] idempiere

[Yedidyah Bar David]

On Mon, Mar 13, 2017 at 8:22 PM, Logan Kuhn  wrote:
> We have 4 idempiere-* VMs in an older cluster and we can search for VMs
> based on all sorts of strings, but if it contains idempiere it shows:
>
> 2017-03-13 11:27:33,268 INFO  [org.ovirt.engine.core.bll.SearchQuery]
> (default task-19) [] ResourceManager::searchBusinessObjects - erroneous
> search text - ''Vms: idempiere''
>
> Can anyone else duplicate this problem in their environment?

Duplicated here.

It seems like entering "Vms: id" triggers the syntax-based search (expecting
that you'll search by "id"). So if you decide to open a bug, it can be something
like: "Free text search fails if starting with a keyword". The same happens if
I search e.g. "Vms: ipa" - it seems to expect me to search by ip address.

Adding Ravi.

For your current needs, you can use: "Vms: name=*idempiere*".

Best,

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



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