Re: [ovirt-users] Multiple Data Storage Domains

[Sahina Bose]

On Mon, Nov 7, 2016 at 11:20 AM, Gary Pedretty  wrote:

> As a storage domain, this gluster volume will not work whether it is
> preallocated or thin provision.   It will work as a straight gluster volume
> mounted directly to any VM on the ovirt Cluster, or any physical machine,
> just not as a data storage domain in the Data Center.
>
> Are there restrictions to having more than one data storage domain that
> has it gluster volumes on the same hosts that are also part of the Data
> Center and Cluster?
>

There are no such restrictions.

However your volume configuration seems suspect -"stripe 2 replica 2". Can
you provide gluster volume info of your second storage domain gluster
volume? The mount logs of the volume (under
/var/log/glusterfs/rhev-datacenter...log) from the host where the
volume is being mounted will also help.


>
>
> Gary
>
>
> 
> Gary Pedrettyg...@ravnalaska.net
> 
> Systems Manager  www.flyravn.com
> Ravn Alaska   /\907-450-7251
> 5245 Airport Industrial Road /  \/\ 907-450-7238 fax
> Fairbanks, Alaska  99709/\  /\ \ Second greatest commandment
> Serving All of Alaska  /  \/  /\  \ \/\   “Love your neighbor as
> Really loving the record green up date! Summmer!!   yourself” Matt 22:39
> 
>
>
>
>
>
>
>
>
>
>
>
>
> On Nov 6, 2016, at 6:28 AM, Maor Lipchuk  wrote:
>
> Hi Gary,
>
> Do you have other disks on this storage domain?
> Have you tried to use other VMs with disks on this storage domain?
> Is this disk is preallocated? If not can you try to create a pre-allocate
> disk and re-try
>
> Regards,
> Maor
>
>
>
> On Sat, Nov 5, 2016 at 2:28 AM, Gary Pedretty  wrote:
>
>> I am having an issue in a Hosted Engine GlusterFS setup.   I have 4 hosts
>> in a cluster, with the Engined being hosted on the Cluster.  This follows
>> the pattern shown in the docs for a glusterized setup, except that I have 4
>> hosts.   I have engine, data, iso and export storage domains all as
>> glusterfs on a replica 3 glusterfs on the first 3 hosts.  These gluster
>> volumes are running on an SSD Hardware Raid 6, which is identical on all
>> the hosts.  All the hosts have a second Raid 6 Array with Physical Hard
>> Drives and I have created a second data storage domain as a glusterfs
>> across all 4 hosts as a stripe 2 replica 2 and have added it to the Data
>> Center.  However if I use this second Storage Domain as the boot disk for a
>> VM, or as second disk for a VM that is already running, the VM will become
>> non-responsive as soon as the VM starts using this disk.   Happens during
>> the OS install if the VM is using this storage domain for its boot disk, or
>> if I try copying anything large to it when it is a second disk for a VM
>> that has its boot drive on the Master Data Storage Domain.
>>
>> If I mount the gluster volume that is this second storage domain on one
>> of the hosts directly or any other machine on my local network, the gluster
>> volume works fine.  It is only when it is used as a storage domain (second
>> data domain) on VMs in the cluster.
>>
>> Once the vm becomes non-responsive it cannot be stopped, removed or
>> destroyed without restarting the host machine that the VM is currently
>> running on.   The 4 hosts are connected via 10gig ethernet, so should not
>> be a network issue.
>>
>>
>> Any ideas?
>>
>> Gary
>>
>
>
> ___
> 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] Multiple Data Storage Domains

[Gary Pedretty]

As a storage domain, this gluster volume will not work whether it is 
preallocated or thin provision.   It will work as a straight gluster volume 
mounted directly to any VM on the ovirt Cluster, or any physical machine, just 
not as a data storage domain in the Data Center.

Are there restrictions to having more than one data storage domain that has it 
gluster volumes on the same hosts that are also part of the Data Center and 
Cluster?


Gary



Gary Pedrettyg...@ravnalaska.net 

Systems Manager  www.flyravn.com 

Ravn Alaska   /\907-450-7251
5245 Airport Industrial Road /  \/\ 907-450-7238 fax
Fairbanks, Alaska  99709/\  /\ \ Second greatest commandment
Serving All of Alaska  /  \/  /\  \ \/\   “Love your neighbor as
Really loving the record green up date! Summmer!!   yourself” Matt 22:39













> On Nov 6, 2016, at 6:28 AM, Maor Lipchuk  wrote:
> 
> Hi Gary,
> 
> Do you have other disks on this storage domain?
> Have you tried to use other VMs with disks on this storage domain?
> Is this disk is preallocated? If not can you try to create a pre-allocate 
> disk and re-try
> 
> Regards,
> Maor
> 
> 
> 
> On Sat, Nov 5, 2016 at 2:28 AM, Gary Pedretty  > wrote:
> I am having an issue in a Hosted Engine GlusterFS setup.   I have 4 hosts in 
> a cluster, with the Engined being hosted on the Cluster.  This follows the 
> pattern shown in the docs for a glusterized setup, except that I have 4 
> hosts.   I have engine, data, iso and export storage domains all as glusterfs 
> on a replica 3 glusterfs on the first 3 hosts.  These gluster volumes are 
> running on an SSD Hardware Raid 6, which is identical on all the hosts.  All 
> the hosts have a second Raid 6 Array with Physical Hard Drives and I have 
> created a second data storage domain as a glusterfs across all 4 hosts as a 
> stripe 2 replica 2 and have added it to the Data Center.  However if I use 
> this second Storage Domain as the boot disk for a VM, or as second disk for a 
> VM that is already running, the VM will become non-responsive as soon as the 
> VM starts using this disk.   Happens during the OS install if the VM is using 
> this storage domain for its boot disk, or if I try copying anything large to 
> it when it is a second disk for a VM that has its boot drive on the Master 
> Data Storage Domain. 
> 
> If I mount the gluster volume that is this second storage domain on one of 
> the hosts directly or any other machine on my local network, the gluster 
> volume works fine.  It is only when it is used as a storage domain (second 
> data domain) on VMs in the cluster.
> 
> Once the vm becomes non-responsive it cannot be stopped, removed or destroyed 
> without restarting the host machine that the VM is currently running on.   
> The 4 hosts are connected via 10gig ethernet, so should not be a network 
> issue.
> 
> 
> Any ideas?
> 
> Gary

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

Re: [ovirt-users] Import of OVA fails in 4.0.4

[Derek Atkins]

Hi,

On Sun, November 6, 2016 4:07 pm, Richard W.M. Jones wrote:
> On Sun, Nov 06, 2016 at 10:11:43PM +0100, Tomáš Golembiovský wrote:
>> Hi,
>>
>> On Sat, 5 Nov 2016 17:34:25 -0400
>> "Derek Atkins"  wrote:
>>
>> > Hi again,
>> >
>> > After a bit more research I found this message in libguestfs[0] that
>> seems
>> > to imply the issue is related to Fedora 23 in that dracut was moved
>> from
>> > /sbin to /usr/bin.  Apparently this was fixed in libguestfs on January
>> 28,
>> > 2016 by a change to convert_linux.ml [1].  But apparently this change
>> > isn't in the centos-7.2/ovirt-4.0.4 packages?
>>
>> Yes you're right. That change was not back-ported to the version of
>> virt-v2v on CentOS 7.2. But if I understand things correctly it should
>> appear in CentOS 7.3. Is waiting for Centos 7.3 an option for you?

t depends.  When is 7.3 due?  If the answer is "next week", then yes, I
can wait.  If the answer is "in a month or four" then no.  ;-)

>> > I just looked through my ovirt host and canot find "convert_linux.ml"
>> > anywhere, so I have no idea how to actually apply this patch locally.
>>
>> You can't just patch the code localy (it's not Python). You'd need to
>> recompile the virt-v2v binary.
>>
>>
>> > Short term I suppose I can fix this by adding a symlink in the VM and
>> > repackage the OVA, but it would be nice to actually get this bugfix
>> > propagated; it's been almost a year?
>>
>> Adding Richard to the loop.
>
> I need the additional information as described here:
>
> https://bugzilla.redhat.com/show_bug.cgi?id=1392228
>
> Rich.

Thanks, Rich.  I'm honestly not sure what the full log told you that my
snippets didn't, but upgrading to your EL7.3 pre-release packages solved
my problem.

Thanks,

Now back to continuing my setup and testing (and backups).

> --
> Richard Jones, Virtualization Group, Red Hat
> http://people.redhat.com/~rjones
> Read my programming and virtualization blog: http://rwmj.wordpress.com
> libguestfs lets you edit virtual machines.  Supports shell scripting,
> bindings from many languages.  http://libguestfs.org

-derek

-- 
   Derek Atkins 617-623-3745
   de...@ihtfp.com www.ihtfp.com
   Computer and Internet Security Consultant

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

Re: [ovirt-users] Import of OVA fails in 4.0.4

[Richard W.M. Jones]

On Sun, Nov 06, 2016 at 10:11:43PM +0100, Tomáš Golembiovský wrote:
> Hi,
> 
> On Sat, 5 Nov 2016 17:34:25 -0400
> "Derek Atkins"  wrote:
> 
> > Hi again,
> > 
> > After a bit more research I found this message in libguestfs[0] that seems
> > to imply the issue is related to Fedora 23 in that dracut was moved from
> > /sbin to /usr/bin.  Apparently this was fixed in libguestfs on January 28,
> > 2016 by a change to convert_linux.ml [1].  But apparently this change
> > isn't in the centos-7.2/ovirt-4.0.4 packages?
> 
> Yes you're right. That change was not back-ported to the version of
> virt-v2v on CentOS 7.2. But if I understand things correctly it should
> appear in CentOS 7.3. Is waiting for Centos 7.3 an option for you?
> 
> 
> > I just looked through my ovirt host and canot find "convert_linux.ml"
> > anywhere, so I have no idea how to actually apply this patch locally.
> 
> You can't just patch the code localy (it's not Python). You'd need to
> recompile the virt-v2v binary.
> 
> 
> > Short term I suppose I can fix this by adding a symlink in the VM and
> > repackage the OVA, but it would be nice to actually get this bugfix
> > propagated; it's been almost a year?
> 
> Adding Richard to the loop.

I need the additional information as described here:

https://bugzilla.redhat.com/show_bug.cgi?id=1392228

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
libguestfs lets you edit virtual machines.  Supports shell scripting,
bindings from many languages.  http://libguestfs.org
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] Import of OVA fails in 4.0.4

[Tomáš Golembiovský]

Hi,

On Sat, 5 Nov 2016 17:34:25 -0400
"Derek Atkins"  wrote:

> Hi again,
> 
> After a bit more research I found this message in libguestfs[0] that seems
> to imply the issue is related to Fedora 23 in that dracut was moved from
> /sbin to /usr/bin.  Apparently this was fixed in libguestfs on January 28,
> 2016 by a change to convert_linux.ml [1].  But apparently this change
> isn't in the centos-7.2/ovirt-4.0.4 packages?

Yes you're right. That change was not back-ported to the version of
virt-v2v on CentOS 7.2. But if I understand things correctly it should
appear in CentOS 7.3. Is waiting for Centos 7.3 an option for you?


> I just looked through my ovirt host and canot find "convert_linux.ml"
> anywhere, so I have no idea how to actually apply this patch locally.

You can't just patch the code localy (it's not Python). You'd need to
recompile the virt-v2v binary.


> Short term I suppose I can fix this by adding a symlink in the VM and
> repackage the OVA, but it would be nice to actually get this bugfix
> propagated; it's been almost a year?

Adding Richard to the loop.


> Thanks,
> 
> -derek
> 
> [0] https://www.redhat.com/archives/libguestfs/2016-January/msg00138.html
> [1]
> https://github.com/libguestfs/libguestfs/commit/569ad259b4726fbdabf380bdca34da00ca262e1e
> 
> On Sat, November 5, 2016 5:02 pm, Derek Atkins wrote:
> > Hi,
> > I just tried using the OVA import feature in 4.0.4 to import an OVA file
> > created from VMware-Workstation-12.  I packaged the OVA, got it onto my
> > ovirt host, and then go the Admin GUI and under the Virtual Machines tab
> > clicked on Import.  From there I chose Source:  VMware Virtual Appliance
> > (OVA), and selected the (local-to-the-host) path where I put the OVA
> > file.  After clicking Load it displayed the vm name in the window.  I
> > click on that and click the right arrow to and then Next.  Then I select
> > my storage domain to use and continue.  It loads the VM and starts to
> > process it internally with virt-v2v, but then it inexplicably fails.
> > The only message in the GUI is:
> >Failed to import Vm  to Data Center Default, Cluster Default
> >
> > I tried to look into /var/log/vdsm/vdsm.log looking for any errors.  I
> > found this:
> >
> > Thread-88166::ERROR::2016-11-05 16:36:09,219::v2v::674::root::(_run) Job
> > u'cdb1e
> > eb7-63f3-46e8-abd5-5292dcc8809a' failed
> > Traceback (most recent call last):
> >   File "/usr/lib/python2.7/site-packages/vdsm/v2v.py", line 669, in _run
> > self._import()
> >   File "/usr/lib/python2.7/site-packages/vdsm/v2v.py", line 696, in
> > _import
> > self._proc.stderr.read(1024)))
> > V2VProcessError: Job u'cdb1eeb7-63f3-46e8-abd5-5292dcc8809a' process
> > failed exit
> > -code: 1, stderr: virt-v2v: error: unable to rebuild initrd
> > (/boot/initramfs-4.2.6-301.fc23.x86_64.img) because mkinitrd or dracut was
> > not found in the guest
> >
> > If reporting bugs, run virt-v2v with debugging enabled and include the
> > complete output:
> >
> >   virt-v2v -v -x [...]
> >
> > I don't understand this error because /usr/bin/dracut certainly exists
> > in the VM.
> >
> > Anyone have any suggestions on how to get around this?
> >
> > Thanks,
> >
> > -derek
> > --
> >Derek Atkins, SB '93 MIT EE, SM '95 MIT Media Laboratory
> >Member, MIT Student Information Processing Board  (SIPB)
> >URL: http://web.mit.edu/warlord/PP-ASEL-IA N1NWH
> >warl...@mit.eduPGP key available
> > ___
> > Users mailing list
> > Users@ovirt.org
> > http://lists.ovirt.org/mailman/listinfo/users
> >  
> 
> 
> -- 
>Derek Atkins 617-623-3745
>de...@ihtfp.com www.ihtfp.com
>Computer and Internet Security Consultant
> 
> ___
> Users mailing list
> Users@ovirt.org
> http://lists.ovirt.org/mailman/listinfo/users


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

Re: [ovirt-users] Live merge speed

[Kai Wagner]

Hi Markus,

can you send an update to the list if this does help as well?

Thanks

Kai

Am 06.11.2016 um 20:32 schrieb Markus Stockhausen:

Hi Maor,

from our observation it MIGHT come from the xfs filesystem on the NFS storage.
qemu thin provisioned raw disks of linux database VMs tend to fragment heavily
during growth. Some of the disk images have > 1.000.000 extends.

To sort it out we have fixed the extendsize of the OVirt mount point to 1M
(xfs_io -c 'extsize 1m' /var/nas/OVirt). If this does not help I will send you 
some
update.

Best regards.

Markus


Von: Maor Lipchuk [mlipc...@redhat.com]
Gesendet: Sonntag, 6. November 2016 16:33
An: Markus Stockhausen
Cc: Ovirt Users
Betreff: Re: [ovirt-users] Live merge speed

Hi Markus,

Can you try to use dd on a random file in that storage domain, What is the 
performance it reflects?
Just want to make sure what is the origin of the problem, if  it is related to 
the storage domain or the qemu process.

Regards,
Maor

On Thu, Nov 3, 2016 at 8:09 PM, Markus Stockhausen
 wrote:

Hi there,

in the past we already observed that live merge operations may take some time.
Currently another of those operations is running on one of our VMs on a
CentOS 7/qemu 2.3 node. Storage is (async - yeah I know about potential data
loss) NFS over 10GBit network.

The operation is running for 50 minutes now with an average throughput of
5MByte/sec. The disk (defined at 400GB) and its snapshots does not show much
thin provisioning:

ls -al
-rw-rw.   1 vdsm kvm 429496729600  7. Apr 2016  
a204865e-ef57-41e0-a2ac-465a6c9c3d60
-rw-rw.   1 vdsm kvm 203334418432  3. Nov 2016  
aeda833f-1f4b-4b6f-b553-578f1f3f06ee
-rw-rw.   1 vdsm kvm 192091389952  3. Nov 2016  
c8acdbc7-af24-4c5c-94c5-ae7262d98f5c
du -m a204865e-ef57-41e0-a2ac-465a6c9c3d60
383070  a204865e-ef57-41e0-a2ac-465a6c9c3d60
du -m aeda833f-1f4b-4b6f-b553-578f1f3f06ee
193953  aeda833f-1f4b-4b6f-b553-578f1f3f06ee
du -m c8acdbc7-af24-4c5c-94c5-ae7262d98f5c
183222  c8acdbc7-af24-4c5c-94c5-ae7262d98f5c

Usually at some point the process is gaining speed and we see >100MByte/sec
speed. Can anyone explain what might be going on.

Best regards.

Markus Stockhausen


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



it-novum GmbH 
i. A. Kai Wagner  -  Team Lead Support & Presales openATTIC 
_


Tel: +49 661 103-762 
Fax: +49 661 10317762   
Mail: kai.wag...@it-novum.com


it-novum GmbH • Edelzeller Straße 44 • 36043 Fulda • http://www.it-novum.com/ 
Handelsregister Amtsgericht Fulda, HRB 1934 • Geschäftsführer: Michael Kienle • Sitz der Gesellschaft: Fulda 

Der Inhalt dieser E-Mail ist vertraulich. Wenn Sie nicht der eigentliche Empfänger sein sollten, informieren Sie bitte sofort den Absender oder vernichten umgehend diese Mail. Jegliche unerlaubte Vervielfältigung oder Weiterleitung dieser Mail ist strengstens verboten. 
This e-mail may contain confidential and/or priviledged information. If you are not the intended recepient (or have received this e-mail in error) please notify the sender immediately and destroy this e-mail. Any unauthorised copying, disclosure or distribution of material in this e-mail is strictly forbidden. 

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

Re: [ovirt-users] Live merge speed

[Markus Stockhausen]

Hi Maor,

from our observation it MIGHT come from the xfs filesystem on the NFS storage.
qemu thin provisioned raw disks of linux database VMs tend to fragment heavily
during growth. Some of the disk images have > 1.000.000 extends.

To sort it out we have fixed the extendsize of the OVirt mount point to 1M
(xfs_io -c 'extsize 1m' /var/nas/OVirt). If this does not help I will send you 
some
update.

Best regards.

Markus


>>>
Von: Maor Lipchuk [mlipc...@redhat.com]
Gesendet: Sonntag, 6. November 2016 16:33
An: Markus Stockhausen
Cc: Ovirt Users
Betreff: Re: [ovirt-users] Live merge speed

Hi Markus,

Can you try to use dd on a random file in that storage domain, What is the 
performance it reflects? 
Just want to make sure what is the origin of the problem, if  it is related to 
the storage domain or the qemu process.

Regards,
Maor

On Thu, Nov 3, 2016 at 8:09 PM, Markus Stockhausen 
 wrote:

Hi there,

in the past we already observed that live merge operations may take some time.
Currently another of those operations is running on one of our VMs on a
CentOS 7/qemu 2.3 node. Storage is (async - yeah I know about potential data
loss) NFS over 10GBit network.

The operation is running for 50 minutes now with an average throughput of
5MByte/sec. The disk (defined at 400GB) and its snapshots does not show much
thin provisioning:

ls -al
-rw-rw.   1 vdsm kvm 429496729600  7. Apr 2016  
a204865e-ef57-41e0-a2ac-465a6c9c3d60
-rw-rw.   1 vdsm kvm 203334418432  3. Nov 2016  
aeda833f-1f4b-4b6f-b553-578f1f3f06ee
-rw-rw.   1 vdsm kvm 192091389952  3. Nov 2016  
c8acdbc7-af24-4c5c-94c5-ae7262d98f5c
du -m a204865e-ef57-41e0-a2ac-465a6c9c3d60
383070  a204865e-ef57-41e0-a2ac-465a6c9c3d60
du -m aeda833f-1f4b-4b6f-b553-578f1f3f06ee
193953  aeda833f-1f4b-4b6f-b553-578f1f3f06ee
du -m c8acdbc7-af24-4c5c-94c5-ae7262d98f5c
183222  c8acdbc7-af24-4c5c-94c5-ae7262d98f5c

Usually at some point the process is gaining speed and we see >100MByte/sec
speed. Can anyone explain what might be going on.

Best regards.

Markus Stockhausen

Diese E-Mail enthält vertrauliche und/oder rechtlich geschützte
Informationen. Wenn Sie nicht der richtige Adressat sind oder diese E-Mail
irrtümlich erhalten haben, informieren Sie bitte sofort den Absender und
vernichten Sie diese Mail. Das unerlaubte Kopieren sowie die unbefugte
Weitergabe dieser Mail ist nicht gestattet.

Über das Internet versandte E-Mails können unter fremden Namen erstellt oder
manipuliert werden. Deshalb ist diese als E-Mail verschickte Nachricht keine
rechtsverbindliche Willenserklärung.

Collogia
Unternehmensberatung AG
Ubierring 11
D-50678 Köln

Vorstand:
Kadir Akin
Dr. Michael Höhnerbach

Vorsitzender des Aufsichtsrates:
Hans Kristian Langva

Registergericht: Amtsgericht Köln
Registernummer: HRB 52 497

This e-mail may contain confidential and/or privileged information. If you
are not the intended recipient (or have received this e-mail in error)
please notify the sender immediately and destroy this e-mail. Any
unauthorized copying, disclosure or distribution of the material in this
e-mail is strictly forbidden.

e-mails sent over the internet may have been written under a wrong name or
been manipulated. That is why this message sent as an e-mail is not a
legally binding declaration of intention.

Collogia
Unternehmensberatung AG
Ubierring 11
D-50678 Köln

executive board:
Kadir Akin
Dr. Michael Höhnerbach

President of the supervisory board:
Hans Kristian Langva

Registry office: district court Cologne
Register number: HRB 52 497


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

Re: [ovirt-users] Live merge speed

[Maor Lipchuk]

Hi Markus,

Can you try to use dd on a random file in that storage domain, What is the
performance it reflects?
Just want to make sure what is the origin of the problem, if  it is related
to the storage domain or the qemu process.

Regards,
Maor

On Thu, Nov 3, 2016 at 8:09 PM, Markus Stockhausen 
wrote:

> Hi there,
>
> in the past we already observed that live merge operations may take some
> time.
> Currently another of those operations is running on one of our VMs on a
> CentOS 7/qemu 2.3 node. Storage is (async - yeah I know about potential
> data
> loss) NFS over 10GBit network.
>
> The operation is running for 50 minutes now with an average throughput of
> 5MByte/sec. The disk (defined at 400GB) and its snapshots does not show
> much
> thin provisioning:
>
> ls -al
> -rw-rw.   1 vdsm kvm 429496729600  7. Apr 2016
> a204865e-ef57-41e0-a2ac-465a6c9c3d60
> -rw-rw.   1 vdsm kvm 203334418432  3. Nov 2016
> aeda833f-1f4b-4b6f-b553-578f1f3f06ee
> -rw-rw.   1 vdsm kvm 192091389952  3. Nov 2016
> c8acdbc7-af24-4c5c-94c5-ae7262d98f5c
>
> du -m a204865e-ef57-41e0-a2ac-465a6c9c3d60
> 383070  a204865e-ef57-41e0-a2ac-465a6c9c3d60
> du -m aeda833f-1f4b-4b6f-b553-578f1f3f06ee
> 193953  aeda833f-1f4b-4b6f-b553-578f1f3f06ee
> du -m c8acdbc7-af24-4c5c-94c5-ae7262d98f5c
> 183222  c8acdbc7-af24-4c5c-94c5-ae7262d98f5c
>
> Usually at some point the process is gaining speed and we see >100MByte/sec
> speed. Can anyone explain what might be going on.
>
> Best regards.
>
> Markus Stockhausen
>
>
> ___
> 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] Multiple Data Storage Domains

[Maor Lipchuk]

Hi Gary,

Do you have other disks on this storage domain?
Have you tried to use other VMs with disks on this storage domain?
Is this disk is preallocated? If not can you try to create a pre-allocate
disk and re-try

Regards,
Maor



On Sat, Nov 5, 2016 at 2:28 AM, Gary Pedretty  wrote:

> I am having an issue in a Hosted Engine GlusterFS setup.   I have 4 hosts
> in a cluster, with the Engined being hosted on the Cluster.  This follows
> the pattern shown in the docs for a glusterized setup, except that I have 4
> hosts.   I have engine, data, iso and export storage domains all as
> glusterfs on a replica 3 glusterfs on the first 3 hosts.  These gluster
> volumes are running on an SSD Hardware Raid 6, which is identical on all
> the hosts.  All the hosts have a second Raid 6 Array with Physical Hard
> Drives and I have created a second data storage domain as a glusterfs
> across all 4 hosts as a stripe 2 replica 2 and have added it to the Data
> Center.  However if I use this second Storage Domain as the boot disk for a
> VM, or as second disk for a VM that is already running, the VM will become
> non-responsive as soon as the VM starts using this disk.   Happens during
> the OS install if the VM is using this storage domain for its boot disk, or
> if I try copying anything large to it when it is a second disk for a VM
> that has its boot drive on the Master Data Storage Domain.
>
> If I mount the gluster volume that is this second storage domain on one of
> the hosts directly or any other machine on my local network, the gluster
> volume works fine.  It is only when it is used as a storage domain (second
> data domain) on VMs in the cluster.
>
> Once the vm becomes non-responsive it cannot be stopped, removed or
> destroyed without restarting the host machine that the VM is currently
> running on.   The 4 hosts are connected via 10gig ethernet, so should not
> be a network issue.
>
>
> Any ideas?
>
> Gary
>
> 
> Gary Pedrettyg...@ravnalaska.net
> 
> Systems Manager  www.flyravn.com
> Ravn Alaska   /\907-450-7251
> 5245 Airport Industrial Road /  \/\ 907-450-7238 fax
> Fairbanks, Alaska  99709/\  /\ \ Second greatest commandment
> Serving All of Alaska  /  \/  /\  \ \/\   “Love your neighbor as
> Really loving the record green up date! Summmer!!   yourself” Matt 22:39
> 
>
>
>
>
>
>
>
>
>
>
>
>
>
> ___
> 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] check_rhev3 1.6 released

[René Koch]

I'm happy to announce version 1.6 of check_rhev3.

check_rhev3 is a monitoring plugin for Icinga/Nagios and it's forks, 
which is
used to monitor datacenters, clusters, hosts, vms, vm pools and storage 
domains
of Red Hat Enterprise Virtualization (RHEV) and oVirt virtualization 
environments.


Download this plugin from: 
https://github.com/ovido/check_rhev3/releases/check_rhev3-1.6


For further information on how to install this plugin visit:
https://github.com/ovido/check_rhev3/wiki/Installation-Documentation

A detailed usage documentation can be found here:
https://github.com/ovido/check_rhev3/wiki/Usage-Documentation


Changelog:

- New features:
-   Add support for oVirt 4 and RHEV 4 (#42)
-   Add option to ignore xml warnings (#49)
-   Components in state maintenance result in warning state now instead 
of critical


Bugs fixed:
-   Critical/Warning thresholds for -l vms/hosts are ignored (#44)


Please note that API path changed in RHEV 4 and oVirt 4.
It's /ovirt-engine/api instead of /api. This plugin will still use the 
old path for compatibility
reasons with older installations. If you use RHEV 4 or oVirt 4 please 
use option -A:

./check_rhev3 -A "/ovirt-engine/api" ...


If you have any questions or ideas, please drop me an email:
rk...@rk-it.at.

Thank you for using check_rhev3.


Regards,
René




smime.p7s
Description: S/MIME Cryptographic Signature
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] 3.6 to 4.0 Functionality Elevation

[Maor Lipchuk]

Hi Clint,
See my comments inline

Regards,
Maor

On Fri, Nov 4, 2016 at 1:51 PM, Clint Boggio 
wrote:

> Greetings OVirt Community
>
> I have a production system that started life as a OV 3.6 system, and I've
> recently upgraded the engine to 4.0. I see in the cluster config, a
> configuration element that indicates the cluster functionality level. It
> has a drop down that allows me to toggle between 3.6, and 4.0.
>
> If I select 4.0 and save, what potential problems could I expect ?
>
> Setup;
>
> 1 Dedicated engine CentOS7 OV version 4
>
> 6 hosts running CentOS7 and OV version 4.0 and 3.6 repos.
>

If your hosts are already EL7, you can take a look at the upgrade manager
https://www.ovirt.org/develop/release-management/features/engine/
upgrademanager/
see also the following demo video:
https://www.youtube.com/watch?v=fDzBNKu5pyQ=youtu.be

Otherwise you'll have to reinstall hosts with EL7 one by one.

After all the hosts were upgraded you simply need to change all your
clusters' compatibility level to 4.0 and finally the Data Center's
compatibility level to 4.0.


>
> Main data domain is iscsi over IB.
>
> ISO and all other data domains are  NFS over IB
>
>
Regarding your Storage Domains (ISO, NFS, iSCSI) there shoul not be any
additional changes as far as I know since all of them will use the same
version (V3) for engine 3.6 and 4.0.


>
> ___
> 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] Migrate oVirtBackup to sdk v4: How to create a snapshot

[Ondra Machacek]

On 11/04/2016 09:42 PM, gregor wrote:

Hi,

I'm started the migration of my backup tool [1] from sdk v3 to v4.
Whats the correct REST URI to create a snapshost from a given VM and how
can this be done with the SnapshotsService.add(...) method?


Please take a look at following example:


https://github.com/oVirt/ovirt-engine-sdk/blob/master/sdk/examples/add_vm_snapshot.py#L45



This method needs a "snapshot" object, but currently I don't know how to
obtain one from a given VM or initialize one which works with the
SnapshotsService class.

A REST way will also be interesting because the Python SDK is horrible
documented and I plan to move to basic REST. Also the Python SDK has
lots of bug's which consumes many of the migration time. For example a
obtained VmService object get by the VmsService.list(with filter) method
has no information about the ID but iterating over all objects from
list() without filter result in usable VmService objects. This will
later be documented, meanwhile I wrote a workaround.


VmsService.list doesn't return list of VmService objects, but list of 
types.Vm object.


So what you will see in above add_vm_snapshot example is follwoing:

# Obtain types.Vm object which is actually a Python representation of
# XML, which oVirt API returns, when you send request like this:
# ovirt-engine/api/vms?search=name=myvm
vm = vms_service.list(search='name=myvm')[0]

# Locate the service that manages the snapshots of the virtual machine,
# which we have found in previous step:
snapshots_service = vms_service.vm_service(vm.id).snapshots_service()

# Add the new snapshot to that VM:
snapshots_service.add(
  types.Snapshot(
description='My snapshot',
  ),
)

Everything works here as expected. Can you please tell us what is
confusing and what should be better documented?

The documentation of SnapshotsService class is here:


https://machacekondra.github.io/ovirt-engine-sdk/v4.0/latest/services.m.html#ovirtsdk4.services.SnapshotsService

We want to improve, so the pages will contain also Python examples and 
maybe some tutorial.


 https://machacekondra.github.io/ovirt-engine-sdk/



cheers
gregor

[1] https://github.com/wefixit-AT/oVirtBackup
___
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] RHEVM3.3 java SDK on RHEV3.6

[Tejesh M]

Hi All,

I'm facing an issue in invoking Cloudinit while customization in RHEV 3.6
using RHEV Java SDK 3.6.

Below link says we need to explicitly call *use_cloud_init=True*, but i'm
not sure how to pass this value as parameter while creating VM.

http://users.ovirt.narkive.com/BGUkanAk/ovirt-users-cloud-init-not-apply-when-v-create-from-api


*My Code:*
String password = "password"; //"password"

String templateName = "centos6.7-final";
String clusterName = "Default";
String vmName = "testVM3";

//Prepare the data to create the VM from the template:
org.ovirt.engine.sdk.entities.Template templateData = new Template();
templateData.setName(templateName);
org.ovirt.engine.sdk.entities.Cluster clusterData = new Cluster();
clusterData.setName(clusterName);
org.ovirt.engine.sdk.entities.VM vmDataForCreate = new VM();
vmDataForCreate.setName(vmName);
vmDataForCreate.setCluster(clusterData);
vmDataForCreate.setTemplate(templateData);

api.getVMs().add(vmDataForCreate);
String state;
for (;;) {
 state = api.getVMs().get(vmName).getStatus().getState();
if ("down".equals(state)) {
  break;
}
Thread.sleep(1000);
}
System.out.println("System Status:"+state);
org.ovirt.engine.sdk.entities.User userData = new User();
userData.setUserName("root");
userData.setPassword(password);
Users usersData = new Users();
usersData.getUsers().add(userData);
CloudInit cloudData = new CloudInit();


cloudData.setUsers(usersData);
Host hostData = new Host();
hostData.setAddress(vmName);
cloudData.setHost(hostData);

org.ovirt.engine.sdk.entities.NetworkConfiguration networkConfiguration=new
NetworkConfiguration();

networkConfiguration.setNics(new Nics());

Nics nics = networkConfiguration.getNics();
nics.getNics().add(createNic("eth0", "STATIC",
createNetwork("192.168.1.102", "255.255.0.0", "192.168.2.1"), true));

networkConfiguration.setNics(nics);

cloudData.setNetworkConfiguration(networkConfiguration);

Initialization initData = new Initialization();

initData.setCloudInit(cloudData);

VM vmDataForStart = new VM();
vmDataForStart.setInitialization(initData);
Action actionData = new Action();

actionData.setVm(vmDataForStart);

// Send the request to start the VM to the server:
api.getVMs().get(vmName).start(actionData);

-- 
Thanks & Regards
Tejesh
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] Ubuntu - Cloud-init - Routed Networking Problem

[Barak Korren]

On 5 November 2016 at 20:19, Erick Vogeler  wrote:
>
> Is there a way to have cloud-init inject these commands?

You can run arbitrary commands by using the YAML configuration snippet  [1].

[1]: http://cloudinit.readthedocs.io/en/latest/topics/modules.html#runcmd


-- 
Barak Korren
bkor...@redhat.com
RHEV-CI Team
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users