Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-11 Thread Jonathan D. Proulx 
On Wed, May 11, 2016 at 07:30:45PM -0400, Emilien Macchi wrote:
:On Tue, May 3, 2016 at 11:43 AM, Matt Riedemann
: wrote:

:> Another thought is that deployment tools are just copying what devstack
:> does, or what shows up in the configs in our dsvm gate jobs, and those are
:> using qemu, so they assume that's what should be used since that's what we
:> gate on.
:
:In the case of Puppet OpenStack, kvm is default.
:We set the parameter to qemu only in our gate, like devstack does but
:people using puppet-nova to deploy will have KVM driver.
:
:https://github.com/openstack/puppet-nova/blob/master/manifests/compute/libvirt.pp#L110
:
:Should we send a warning if qemu is set?

Since the defautl is KVM a deployer would have to make a conscious
decision to to change that so I wouldn't expect a warning.

If you think additional warning is needed maybe an extra comment line
in libvirt.pp or somewhere?

Though I don't think I'd oppose one either.  I can imagine a scenario
in which someone builds an environment in nested virtualization where
they need non-accelerated qemu then forget to switch it back for
hardware deploy, but that's pretty edge casey.

-Jon

___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-11 Thread Emilien Macchi 
On Tue, May 3, 2016 at 11:43 AM, Matt Riedemann
 wrote:
> On 5/3/2016 10:01 AM, Daniel P. Berrange wrote:
>>
>> Hello Operators,
>>
>> One of the things that constantly puzzles me when reading the user
>> survey results wrt hypervisor is the high number of respondants
>> claiming to be using QEMU (as distinct from KVM).
>>
>> As a reminder, in Nova saying virt_type=qemu causes Nova to use
>> plain QEMU with pure CPU emulation which is many many times slower
>> to than native CPU performance, while virt_type=kvm causes Nova to
>> use QEMU with KVM hardware CPU acceleration which is close to native
>> performance.
>>
>> IOW, virt_type=qemu is not something you'd ever really want to use
>> unless you had no other options due to the terrible performance it
>> would show. The only reasons to use QEMU are if you need non-native
>> architecture support (ie running arm/ppc on x86_64 host), or if you
>> can't do KVM due to hardware restrictions (ie ancient hardware, or
>> running compute hosts inside virtual machines)
>>
>> Despite this, in the 2016 survey 10% claimed to be using QEMU in
>> production & 3% in PoC and dev, in 2014 it was even higher at 15%
>> in prod & 12% in PoC and 28% in dev.
>>
>> Personally my gut feeling says that QEMU usage ought to be in very
>> low single figures, so I'm curious as to the apparent anomoly.
>>
>> I can think of a few reasons
>>
>>  1. Respondants are confused as to the difference between QEMU
>> and KVM, so are saying QEMU, despite fact they are using KVM.
>>
>>  2. Respondants are confused as to the difference between QEMU
>> and KVM, so have mistakenly configured their nova hosts to
>> use QEMU instead of KVM and suffering poor performance without
>> realizing their mistake.
>>
>>  3. There are more people than I expect who are running their
>> cloud compute hosts inside virtual machines, and thus are
>> unable to use KVM.
>>
>>  4. There are more people than I expect who are providing cloud
>> hosting for non-native architectures. eg ability to run an
>> arm7/ppc guest image on an x86_64 host and so genuinely must
>> use QEMU
>>
>> If items 1 / 2 are the cause, then by implication the user survey
>> is likely under-reporting the (already huge) scale of the KVM usage.
>>
>> I can see 3. being a likely explanation for high usage of QEMU in a
>> dev or PoC scenario, but it feels unlikely for a production deployment.
>>
>> While 4 is technically possible, Nova doesn't really do a very good
>> job at mixed guest arch hosting - I'm pretty sure there are broken
>> pieces waiting to bite people who try it.
>>
>> Does anyone have any thoughts on this topic ?
>>
>> Indeed, is there anyone here who genuinely use virt_type=qemu in a
>> production deployment of OpenStack who might have other reasons that
>> I've missed ?
>>
>> Regards,
>> Daniel
>>
>
> Another thought is that deployment tools are just copying what devstack
> does, or what shows up in the configs in our dsvm gate jobs, and those are
> using qemu, so they assume that's what should be used since that's what we
> gate on.

In the case of Puppet OpenStack, kvm is default.
We set the parameter to qemu only in our gate, like devstack does but
people using puppet-nova to deploy will have KVM driver.

https://github.com/openstack/puppet-nova/blob/master/manifests/compute/libvirt.pp#L110

Should we send a warning if qemu is set?

> We should be more clear in our help text for the virt_type config option
> between using kvm vs qemu. Today it just says:
>
> # Libvirt domain type (string value)
> # Allowed values: kvm, lxc, qemu, uml, xen, parallels
> #virt_type = kvm
>
> It'd be good to point out the performance impacts and limitations of kvm vs
> qemu in that help text. There might already be a patch up for review that
> makes this better.
>
> --
>
> Thanks,
>
> Matt Riedemann
>
>
>
> ___
> OpenStack-operators mailing list
> OpenStack-operators@lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators



-- 
Emilien Macchi

___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-11 Thread Sergio Cuellar Valdes 
On 11 May 2016 at 14:32, Maish Saidel-Keesing  wrote:

>
> On 11/05/16 22:18, Chris Friesen wrote:
>
> On 05/11/2016 11:46 AM, Ronald Bradford wrote:
>
> I have been curious as to why as mentioned in the thread virt_type=kvm,
> but
> os-hypervisors API call states QEMU.
>
>
> Arguably in both cases the hypervisor is qemu.  When virt_type=kvm we
> simply enable some additional acceleration.
>
> So rather than asking "Are you using qemu or kvm?", it would be more
> accurate to ask "Are you using hardware-accelerated qemu or just software
> emulation?".
>
> And how would OpenStack present the difference to the Operator?
>
> In the meantime - I have opened up this bug [1]
>
> Chris
>
>
> [1] https://bugs.launchpad.net/horizon/+bug/1580746
>

​There is a similar bug:  ​https://bugs.launchpad.net/nova/+bug/1195361




>
> --
> Best Regards,
> Maish Saidel-Keesing
>
> ___
> OpenStack-operators mailing list
> OpenStack-operators@lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
>
>


-- 
* Sergio Cuéllar │DevOps Engineer*
 KIO NETWORKS
 Mexico City Phone (52) 55 8503 2600 ext. 4335
 Mobile: 5544844298
 www.kionetworks.com
___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-11 Thread Chris Friesen 

On 05/11/2016 01:29 PM, Robert Starmer wrote:

I don't disagree, what we're really getting at is that any lookup (ask the
system what it's using on a particular instance, look at the config, look at the
output of a nova CLI request, querry via Horizon), should all return the same
answer.  So one is a bug (Horizon), the other requires looking up information in
the system itself.  As I suggested, the config is one path, and I still believe
will provide the current correct answer for the hypervisor node (Linux QEMU/KVM
or QEMU/QEMU) regardless of other issues, and the Horizon path is a bug that
should be fixed.


I think the problem is poor modeling.  We specify "virt_type", but export the 
hypervisor.


The "virt_type" option does not have a 1:1 mapping to hypervisor.  Both kvm and 
qemu will use the "qemu" hypervisor but kvm will enable hardware acceleration.


Perhaps we should change it to export "virt_type" instead via a microversion.

Chris


___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-11 Thread Maish Saidel-Keesing 

On 11/05/16 22:18, Chris Friesen wrote:
> On 05/11/2016 11:46 AM, Ronald Bradford wrote:
>> I have been curious as to why as mentioned in the thread
>> virt_type=kvm, but
>> os-hypervisors API call states QEMU.
>
> Arguably in both cases the hypervisor is qemu.  When virt_type=kvm we
> simply enable some additional acceleration.
>
> So rather than asking "Are you using qemu or kvm?", it would be more
> accurate to ask "Are you using hardware-accelerated qemu or just
> software emulation?".

And how would OpenStack present the difference to the Operator?

In the meantime - I have opened up this bug [1]

> Chris
>
>
[1] https://bugs.launchpad.net/horizon/+bug/1580746
-- 
Best Regards,
Maish Saidel-Keesing
___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-11 Thread Robert Starmer 
I don't disagree, what we're really getting at is that any lookup (ask the
system what it's using on a particular instance, look at the config, look
at the output of a nova CLI request, querry via Horizon), should all return
the same answer.  So one is a bug (Horizon), the other requires looking up
information in the system itself.  As I suggested, the config is one path,
and I still believe will provide the current correct answer for the
hypervisor node (Linux QEMU/KVM or QEMU/QEMU) regardless of other issues,
and the Horizon path is a bug that should be fixed.

R

On Wed, May 11, 2016 at 9:20 AM, Maish Saidel-Keesing 
wrote:

> Or we could just fix the problem within OpenStack to report the correct
> Hypervisor in the first place.
>
> This kind of reminds me of a story.
>
> Someone was trying to drive down a path on his bicycle, but there were
> some tacks on the path.
> So his wheel kept on getting full of holes. So they thought and thought of
> how to overcome the problem.
>
> They built small jet engines into the bike so that they could hover over
> the ground, instead of riding over
> the tacks.
>
> They were happy because they could now go down the path.
>
> The simple answer should have been - get off the bike and pick up the
> tacks - instead of finding ways to
> over-engineer the problem
>
> Either show the right thing - or don't show it at all.
>
> My 0.02 Shekels.
> On 11/05/16 22:06, Robert Starmer wrote:
>
> You could just ask for the value of virt_type parameter from a compute
> host (or the output of something like grep 'virt_type' /etc/nova/nova*) if
> you are using qemu or kvm.  I believe that's how nova figures out what
> parameters to use when launching an instance.
>
> On Wed, May 11, 2016 at 8:39 AM, Kris G. Lindgren 
> wrote:
>
>> In the next user survey - could we clarify that qemu == full software cpu
>> emulation and kvm (qemu/kvm) = hardware accelerated virtualization or some
>> similar phrasing.  It's totally possible that people are like: I run both
>> qemu and kvm (thinking that’s qemu/kvm) - when in fact they only run kvm
>> (qemu/kvm).
>>
>> ___
>> Kris Lindgren
>> Senior Linux Systems Engineer
>> GoDaddy
>>
>>
>>
>>
>>
>>
>>
>> On 5/11/16, 11:58 AM, "Tim Bell" < tim.b...@cern.ch>
>> wrote:
>>
>> >Does anyone see a good way to fix this to report KVM or QEMU/KVM ?
>> >
>> >I guess the worry is whether this would count as a bug fix or an
>> incompatible change.
>> >
>> >Tim
>> >
>> >On 11/05/16 17:51, "Kashyap Chamarthy" < 
>> kcham...@redhat.com> wrote:
>> >
>> >>On Tue, May 03, 2016 at 02:27:00PM -0500, Sergio Cuellar Valdes wrote:
>> >>
>> >>[...]
>> >>
>> >>> I'm confused too about the use of KVM or QEMU In the computes the
>> >>> file​/etc/nova/nova-compute.conf has:
>> >>>
>> >>> virt_type=kvm
>> >>>
>> >>> The output of:
>> >>>
>> >>> nova hypervisor-show  | grep hypervisor_type
>> >>>
>> >>> is:
>> >>>
>> >>> hypervisor_type   | QEMU
>> >>
>> >>As Dan noted in his response, it's because it is reporting the libvirt
>> driver
>> >>name (which is reported as QEMU).
>> >>
>> >>Refer below if you want to double-confirm if your instances are using
>> KVM.
>> >>
>> >>>
>> >>> The virsh dumpxml of the instances shows:
>> >>>
>> >>> 
>> >>
>> >>That means, yes, you using KVM.  You can confirm that by checking your
>> QEMU
>> >>command-line of the Nova instance, you'll see something like
>> "accel=kvm":
>> >>
>> >>  # This is on Fedora 23 system
>> >>  $ ps -ef | grep -i qemu-system-x86_64
>> >>  [...] /usr/bin/qemu-system-x86_64 -machine accel=kvm [...]
>> >>
>> >>> 
>> >>> /usr/bin/qemu-system-x86_64
>> >>>
>> >>> ​But according to ​this document [1], it is using QEMU emulator
>> instead of
>> >>> KVM, because it is not using /usr/bin/qemu-kvm
>> >>>
>> >>>
>> >>> So I really don't know if it's using KVM or QEMU.
>> >>
>> >>As noted above, a sure-fire way to know is to see if the instance's QEMU
>> >>command-line has "accel=kvm".
>> >>
>> >>A related useful tool is `virt-host-validate` (which is part of
>> libvirt-client
>> >>package, at least on Fedora-based systems):
>> >>
>> >>   $ virt-host-validate | egrep -i 'kvm'
>> >>QEMU: Checking if device /dev/kvm exists
>>: PASS
>> >>QEMU: Checking if device /dev/kvm is accessible
>> : PASS
>> >>
>> >>
>> >>> [1] https://libvirt.org/drvqemu.html
>> >>>
>> >>
>> >>
>> >>--
>> >>/kashyap
>> >>
>> >>___
>> >>OpenStack-operators mailing list
>> >>OpenStack-operators@lists.openstack.org
>> >>http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
>> >
>> >___
>> >OpenStack-operators mailing list
>> >OpenStack-operators@lists.openstack.org
>> >http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
>>

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-11 Thread Maish Saidel-Keesing 
Or we could just fix the problem within OpenStack to report the correct
Hypervisor in the first place.

This kind of reminds me of a story.

Someone was trying to drive down a path on his bicycle, but there were
some tacks on the path.
So his wheel kept on getting full of holes. So they thought and thought
of how to overcome the problem.

They built small jet engines into the bike so that they could hover over
the ground, instead of riding over
the tacks.

They were happy because they could now go down the path.

The simple answer should have been - get off the bike and pick up the
tacks - instead of finding ways to
over-engineer the problem

Either show the right thing - or don't show it at all.

My 0.02 Shekels.

On 11/05/16 22:06, Robert Starmer wrote:
> You could just ask for the value of virt_type parameter from a compute
> host (or the output of something like grep 'virt_type'
> /etc/nova/nova*) if you are using qemu or kvm.  I believe that's how
> nova figures out what parameters to use when launching an instance.
>
> On Wed, May 11, 2016 at 8:39 AM, Kris G. Lindgren
> > wrote:
>
> In the next user survey - could we clarify that qemu == full
> software cpu emulation and kvm (qemu/kvm) = hardware accelerated
> virtualization or some similar phrasing.  It's totally possible
> that people are like: I run both qemu and kvm (thinking that’s
> qemu/kvm) - when in fact they only run kvm (qemu/kvm).
>
> ___
> Kris Lindgren
> Senior Linux Systems Engineer
> GoDaddy
>
>
>
>
>
>
>
> On 5/11/16, 11:58 AM, "Tim Bell"  > wrote:
>
> >Does anyone see a good way to fix this to report KVM or QEMU/KVM ?
> >
> >I guess the worry is whether this would count as a bug fix or an
> incompatible change.
> >
> >Tim
> >
> >On 11/05/16 17:51, "Kashyap Chamarthy"  > wrote:
> >
> >>On Tue, May 03, 2016 at 02:27:00PM -0500, Sergio Cuellar Valdes
> wrote:
> >>
> >>[...]
> >>
> >>> I'm confused too about the use of KVM or QEMU In the computes the
> >>> file​/etc/nova/nova-compute.conf has:
> >>>
> >>> virt_type=kvm
> >>>
> >>> The output of:
> >>>
> >>> nova hypervisor-show  | grep hypervisor_type
> >>>
> >>> is:
> >>>
> >>> hypervisor_type   | QEMU
> >>
> >>As Dan noted in his response, it's because it is reporting the
> libvirt driver
> >>name (which is reported as QEMU).
> >>
> >>Refer below if you want to double-confirm if your instances are
> using KVM.
> >>
> >>>
> >>> The virsh dumpxml of the instances shows:
> >>>
> >>> 
> >>
> >>That means, yes, you using KVM.  You can confirm that by
> checking your QEMU
> >>command-line of the Nova instance, you'll see something like
> "accel=kvm":
> >>
> >>  # This is on Fedora 23 system
> >>  $ ps -ef | grep -i qemu-system-x86_64
> >>  [...] /usr/bin/qemu-system-x86_64 -machine accel=kvm [...]
> >>
> >>> 
> >>> /usr/bin/qemu-system-x86_64
> >>>
> >>> ​But according to ​this document [1], it is using QEMU
> emulator instead of
> >>> KVM, because it is not using /usr/bin/qemu-kvm
> >>>
> >>>
> >>> So I really don't know if it's using KVM or QEMU.
> >>
> >>As noted above, a sure-fire way to know is to see if the
> instance's QEMU
> >>command-line has "accel=kvm".
> >>
> >>A related useful tool is `virt-host-validate` (which is part of
> libvirt-client
> >>package, at least on Fedora-based systems):
> >>
> >>   $ virt-host-validate | egrep -i 'kvm'
> >>QEMU: Checking if device /dev/kvm exists   
>: PASS
> >>QEMU: Checking if device /dev/kvm is accessible 
>   : PASS
> >>
> >>
> >>> [1] https://libvirt.org/drvqemu.html
> >>>
> >>
> >>
> >>--
> >>/kashyap
> >>
> >>___
> >>OpenStack-operators mailing list
> >>OpenStack-operators@lists.openstack.org
> 
> >>http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
> >
> >___
> >OpenStack-operators mailing list
> >OpenStack-operators@lists.openstack.org
> 
> >http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
> ___
> OpenStack-operators mailing list
> OpenStack-operators@lists.openstack.org
> 
>

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-11 Thread Chris Friesen 

On 05/11/2016 11:46 AM, Ronald Bradford wrote:

I have been curious as to why as mentioned in the thread virt_type=kvm, but
os-hypervisors API call states QEMU.


Arguably in both cases the hypervisor is qemu.  When virt_type=kvm we simply 
enable some additional acceleration.


So rather than asking "Are you using qemu or kvm?", it would be more accurate to 
ask "Are you using hardware-accelerated qemu or just software emulation?".


Chris


___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-11 Thread Robert Starmer 
You could just ask for the value of virt_type parameter from a compute host
(or the output of something like grep 'virt_type' /etc/nova/nova*) if you
are using qemu or kvm.  I believe that's how nova figures out what
parameters to use when launching an instance.

On Wed, May 11, 2016 at 8:39 AM, Kris G. Lindgren 
wrote:

> In the next user survey - could we clarify that qemu == full software cpu
> emulation and kvm (qemu/kvm) = hardware accelerated virtualization or some
> similar phrasing.  It's totally possible that people are like: I run both
> qemu and kvm (thinking that’s qemu/kvm) - when in fact they only run kvm
> (qemu/kvm).
>
> ___
> Kris Lindgren
> Senior Linux Systems Engineer
> GoDaddy
>
>
>
>
>
>
>
> On 5/11/16, 11:58 AM, "Tim Bell"  wrote:
>
> >Does anyone see a good way to fix this to report KVM or QEMU/KVM ?
> >
> >I guess the worry is whether this would count as a bug fix or an
> incompatible change.
> >
> >Tim
> >
> >On 11/05/16 17:51, "Kashyap Chamarthy"  wrote:
> >
> >>On Tue, May 03, 2016 at 02:27:00PM -0500, Sergio Cuellar Valdes wrote:
> >>
> >>[...]
> >>
> >>> I'm confused too about the use of KVM or QEMU In the computes the
> >>> file​/etc/nova/nova-compute.conf has:
> >>>
> >>> virt_type=kvm
> >>>
> >>> The output of:
> >>>
> >>> nova hypervisor-show  | grep hypervisor_type
> >>>
> >>> is:
> >>>
> >>> hypervisor_type   | QEMU
> >>
> >>As Dan noted in his response, it's because it is reporting the libvirt
> driver
> >>name (which is reported as QEMU).
> >>
> >>Refer below if you want to double-confirm if your instances are using
> KVM.
> >>
> >>>
> >>> The virsh dumpxml of the instances shows:
> >>>
> >>> 
> >>
> >>That means, yes, you using KVM.  You can confirm that by checking your
> QEMU
> >>command-line of the Nova instance, you'll see something like "accel=kvm":
> >>
> >>  # This is on Fedora 23 system
> >>  $ ps -ef | grep -i qemu-system-x86_64
> >>  [...] /usr/bin/qemu-system-x86_64 -machine accel=kvm [...]
> >>
> >>> 
> >>> /usr/bin/qemu-system-x86_64
> >>>
> >>> ​But according to ​this document [1], it is using QEMU emulator
> instead of
> >>> KVM, because it is not using /usr/bin/qemu-kvm
> >>>
> >>>
> >>> So I really don't know if it's using KVM or QEMU.
> >>
> >>As noted above, a sure-fire way to know is to see if the instance's QEMU
> >>command-line has "accel=kvm".
> >>
> >>A related useful tool is `virt-host-validate` (which is part of
> libvirt-client
> >>package, at least on Fedora-based systems):
> >>
> >>   $ virt-host-validate | egrep -i 'kvm'
> >>QEMU: Checking if device /dev/kvm exists
>: PASS
> >>QEMU: Checking if device /dev/kvm is accessible
>   : PASS
> >>
> >>
> >>> [1] https://libvirt.org/drvqemu.html
> >>>
> >>
> >>
> >>--
> >>/kashyap
> >>
> >>___
> >>OpenStack-operators mailing list
> >>OpenStack-operators@lists.openstack.org
> >>http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
> >
> >___
> >OpenStack-operators mailing list
> >OpenStack-operators@lists.openstack.org
> >http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
> ___
> OpenStack-operators mailing list
> OpenStack-operators@lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
>
___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-11 Thread Kris G. Lindgren 
In the next user survey - could we clarify that qemu == full software cpu 
emulation and kvm (qemu/kvm) = hardware accelerated virtualization or some 
similar phrasing.  It's totally possible that people are like: I run both qemu 
and kvm (thinking that’s qemu/kvm) - when in fact they only run kvm (qemu/kvm).

___
Kris Lindgren
Senior Linux Systems Engineer
GoDaddy







On 5/11/16, 11:58 AM, "Tim Bell"  wrote:

>Does anyone see a good way to fix this to report KVM or QEMU/KVM ?
>
>I guess the worry is whether this would count as a bug fix or an incompatible 
>change.
>
>Tim
>
>On 11/05/16 17:51, "Kashyap Chamarthy"  wrote:
>
>>On Tue, May 03, 2016 at 02:27:00PM -0500, Sergio Cuellar Valdes wrote:
>>
>>[...]
>>
>>> I'm confused too about the use of KVM or QEMU In the computes the
>>> file​/etc/nova/nova-compute.conf has:
>>> 
>>> virt_type=kvm
>>> 
>>> The output of:
>>> 
>>> nova hypervisor-show  | grep hypervisor_type
>>> 
>>> is:
>>> 
>>> hypervisor_type   | QEMU
>>
>>As Dan noted in his response, it's because it is reporting the libvirt driver
>>name (which is reported as QEMU).
>>
>>Refer below if you want to double-confirm if your instances are using KVM.
>>
>>> 
>>> The virsh dumpxml of the instances shows:
>>> 
>>> 
>>
>>That means, yes, you using KVM.  You can confirm that by checking your QEMU
>>command-line of the Nova instance, you'll see something like "accel=kvm":
>>
>>  # This is on Fedora 23 system
>>  $ ps -ef | grep -i qemu-system-x86_64
>>  [...] /usr/bin/qemu-system-x86_64 -machine accel=kvm [...]
>>
>>> 
>>> /usr/bin/qemu-system-x86_64
>>> 
>>> ​But according to ​this document [1], it is using QEMU emulator instead of
>>> KVM, because it is not using /usr/bin/qemu-kvm
>>>
>>> 
>>> So I really don't know if it's using KVM or QEMU.
>>
>>As noted above, a sure-fire way to know is to see if the instance's QEMU
>>command-line has "accel=kvm".
>>
>>A related useful tool is `virt-host-validate` (which is part of libvirt-client
>>package, at least on Fedora-based systems):
>>
>>   $ virt-host-validate | egrep -i 'kvm'
>>QEMU: Checking if device /dev/kvm exists  
>>  : PASS
>>QEMU: Checking if device /dev/kvm is accessible   
>>  : PASS
>>
>>
>>> [1] https://libvirt.org/drvqemu.html
>>> 
>>
>>
>>-- 
>>/kashyap
>>
>>___
>>OpenStack-operators mailing list
>>OpenStack-operators@lists.openstack.org
>>http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
>
>___
>OpenStack-operators mailing list
>OpenStack-operators@lists.openstack.org
>http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-11 Thread Tim Bell 
Does anyone see a good way to fix this to report KVM or QEMU/KVM ?

I guess the worry is whether this would count as a bug fix or an incompatible 
change.

Tim

On 11/05/16 17:51, "Kashyap Chamarthy"  wrote:

>On Tue, May 03, 2016 at 02:27:00PM -0500, Sergio Cuellar Valdes wrote:
>
>[...]
>
>> I'm confused too about the use of KVM or QEMU In the computes the
>> file​/etc/nova/nova-compute.conf has:
>> 
>> virt_type=kvm
>> 
>> The output of:
>> 
>> nova hypervisor-show  | grep hypervisor_type
>> 
>> is:
>> 
>> hypervisor_type   | QEMU
>
>As Dan noted in his response, it's because it is reporting the libvirt driver
>name (which is reported as QEMU).
>
>Refer below if you want to double-confirm if your instances are using KVM.
>
>> 
>> The virsh dumpxml of the instances shows:
>> 
>> 
>
>That means, yes, you using KVM.  You can confirm that by checking your QEMU
>command-line of the Nova instance, you'll see something like "accel=kvm":
>
>   # This is on Fedora 23 system
>   $ ps -ef | grep -i qemu-system-x86_64
>   [...] /usr/bin/qemu-system-x86_64 -machine accel=kvm [...]
>
>> 
>> /usr/bin/qemu-system-x86_64
>> 
>> ​But according to ​this document [1], it is using QEMU emulator instead of
>> KVM, because it is not using /usr/bin/qemu-kvm
>>
>> 
>> So I really don't know if it's using KVM or QEMU.
>
>As noted above, a sure-fire way to know is to see if the instance's QEMU
>command-line has "accel=kvm".
>
>A related useful tool is `virt-host-validate` (which is part of libvirt-client
>package, at least on Fedora-based systems):
>
>   $ virt-host-validate | egrep -i 'kvm'
>QEMU: Checking if device /dev/kvm exists   
> : PASS
>QEMU: Checking if device /dev/kvm is accessible
> : PASS
>
>
>> [1] https://libvirt.org/drvqemu.html
>> 
>
>
>-- 
>/kashyap
>
>___
>OpenStack-operators mailing list
>OpenStack-operators@lists.openstack.org
>http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-11 Thread Ronald Bradford 
I have been curious as to why as mentioned in the thread virt_type=kvm, but
os-hypervisors API call states QEMU.

Interestingly this command mentioned (which works on Ubuntu) gives me a
FAIL and WARN on my home test setup that runs on physical H/W.


$ virt-host-validate
  QEMU: Checking for hardware virtualization
  : PASS
  QEMU: Checking for device /dev/kvm
  : FAIL (Check that the 'kvm-intel' or 'kvm-amd' modules are loaded & the
BIOS has enabled virtualization)
  QEMU: Checking for device /dev/vhost-net
  : WARN (Load the 'vhost_net' module to improve performance of virtio
networking)
  QEMU: Checking for device /dev/net/tun
  : PASS
   LXC: Checking for Linux >= 2.6.26
  : PASS


Ronald Bradford

Web Site: http://ronaldbradford.com
LinkedIn:  http://www.linkedin.com/in/ronaldbradford
Twitter:@RonaldBradford 
Skype: RonaldBradford
GTalk: Ronald.Bradford
IRC: rbradfor


On Wed, May 11, 2016 at 12:18 PM, Maish Saidel-Keesing 
wrote:

> Which still brings me back to the original point.
>
> Is this a bug - and should it be reported as such?
>
>
>
> On 11/05/16 18:51, Kashyap Chamarthy wrote:
>
> On Tue, May 03, 2016 at 02:27:00PM -0500, Sergio Cuellar Valdes wrote:
>
> [...]
>
>
> I'm confused too about the use of KVM or QEMU In the computes the
> file​/etc/nova/nova-compute.conf has:
>
> virt_type=kvm
>
> The output of:
>
> nova hypervisor-show  | grep hypervisor_type
>
> is:
>
> hypervisor_type   | QEMU
>
> As Dan noted in his response, it's because it is reporting the libvirt driver
> name (which is reported as QEMU).
>
> Refer below if you want to double-confirm if your instances are using KVM.
>
>
> The virsh dumpxml of the instances shows:
>
> 
>
> That means, yes, you using KVM.  You can confirm that by checking your QEMU
> command-line of the Nova instance, you'll see something like "accel=kvm":
>
>   # This is on Fedora 23 system
>   $ ps -ef | grep -i qemu-system-x86_64
>   [...] /usr/bin/qemu-system-x86_64 -machine accel=kvm [...]
>
>
> 
> /usr/bin/qemu-system-x86_64
>
> ​But according to ​this document [1], it is using QEMU emulator instead of
> KVM, because it is not using /usr/bin/qemu-kvm
>
>
> So I really don't know if it's using KVM or QEMU.
>
> As noted above, a sure-fire way to know is to see if the instance's QEMU
> command-line has "accel=kvm".
>
> A related useful tool is `virt-host-validate` (which is part of libvirt-client
> package, at least on Fedora-based systems):
>
>$ virt-host-validate | egrep -i 'kvm'
> QEMU: Checking if device /dev/kvm exists  
>  : PASS
> QEMU: Checking if device /dev/kvm is accessible   
>  : PASS
>
>
>
> [1] https://libvirt.org/drvqemu.html
>
>
> --
> Best Regards,
> Maish Saidel-Keesing
>
> ___
> OpenStack-operators mailing list
> OpenStack-operators@lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
>
>
___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-11 Thread Maish Saidel-Keesing 
Which still brings me back to the original point.

Is this a bug - and should it be reported as such?



On 11/05/16 18:51, Kashyap Chamarthy wrote:
> On Tue, May 03, 2016 at 02:27:00PM -0500, Sergio Cuellar Valdes wrote:
>
> [...]
>
>> I'm confused too about the use of KVM or QEMU In the computes the
>> file​/etc/nova/nova-compute.conf has:
>>
>> virt_type=kvm
>>
>> The output of:
>>
>> nova hypervisor-show  | grep hypervisor_type
>>
>> is:
>>
>> hypervisor_type   | QEMU
> As Dan noted in his response, it's because it is reporting the libvirt driver
> name (which is reported as QEMU).
>
> Refer below if you want to double-confirm if your instances are using KVM.
>
>> The virsh dumpxml of the instances shows:
>>
>> 
> That means, yes, you using KVM.  You can confirm that by checking your QEMU
> command-line of the Nova instance, you'll see something like "accel=kvm":
>
>   # This is on Fedora 23 system
>   $ ps -ef | grep -i qemu-system-x86_64
>   [...] /usr/bin/qemu-system-x86_64 -machine accel=kvm [...]
>
>> 
>> /usr/bin/qemu-system-x86_64
>>
>> ​But according to ​this document [1], it is using QEMU emulator instead of
>> KVM, because it is not using /usr/bin/qemu-kvm
>>
>>
>> So I really don't know if it's using KVM or QEMU.
> As noted above, a sure-fire way to know is to see if the instance's QEMU
> command-line has "accel=kvm".
>
> A related useful tool is `virt-host-validate` (which is part of libvirt-client
> package, at least on Fedora-based systems):
>
>$ virt-host-validate | egrep -i 'kvm'
> QEMU: Checking if device /dev/kvm exists  
>  : PASS
> QEMU: Checking if device /dev/kvm is accessible   
>  : PASS
>
>
>> [1] https://libvirt.org/drvqemu.html
>>
>

-- 
Best Regards,
Maish Saidel-Keesing
___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-11 Thread Kashyap Chamarthy 
On Tue, May 03, 2016 at 02:27:00PM -0500, Sergio Cuellar Valdes wrote:

[...]

> I'm confused too about the use of KVM or QEMU In the computes the
> file​/etc/nova/nova-compute.conf has:
> 
> virt_type=kvm
> 
> The output of:
> 
> nova hypervisor-show  | grep hypervisor_type
> 
> is:
> 
> hypervisor_type   | QEMU

As Dan noted in his response, it's because it is reporting the libvirt driver
name (which is reported as QEMU).

Refer below if you want to double-confirm if your instances are using KVM.

> 
> The virsh dumpxml of the instances shows:
> 
> 

That means, yes, you using KVM.  You can confirm that by checking your QEMU
command-line of the Nova instance, you'll see something like "accel=kvm":

# This is on Fedora 23 system
$ ps -ef | grep -i qemu-system-x86_64
[...] /usr/bin/qemu-system-x86_64 -machine accel=kvm [...]

> 
> /usr/bin/qemu-system-x86_64
> 
> ​But according to ​this document [1], it is using QEMU emulator instead of
> KVM, because it is not using /usr/bin/qemu-kvm
>
> 
> So I really don't know if it's using KVM or QEMU.

As noted above, a sure-fire way to know is to see if the instance's QEMU
command-line has "accel=kvm".

A related useful tool is `virt-host-validate` (which is part of libvirt-client
package, at least on Fedora-based systems):

   $ virt-host-validate | egrep -i 'kvm'
QEMU: Checking if device /dev/kvm exists   
: PASS
QEMU: Checking if device /dev/kvm is accessible
: PASS


> [1] https://libvirt.org/drvqemu.html
> 


-- 
/kashyap

___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-04 Thread Daniel P. Berrange 
On Tue, May 03, 2016 at 07:06:56PM +, Jared Wilkinson wrote:
> So forgive my lack of kvm/qemu knowledge but I couldn’t find anything
> on Google on this. If you deployed an instance of a different architecture
> than the physical CPU, wouldn’t qemu just emulate the processor (if you
> were in virt_type=kvm) mode, or would libvirt throw some error?

You would get an error from libvirt because KVM will only work if the
host and guest arches match

Regards,
Daniel
-- 
|: http://berrange.com  -o-http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org  -o- http://virt-manager.org :|
|: http://autobuild.org   -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org   -o-   http://live.gnome.org/gtk-vnc :|

___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-04 Thread Daniel P. Berrange 
On Tue, May 03, 2016 at 10:20:34PM +0300, Maish Saidel-Keesing wrote:
> I would think that the problem is that OpenStack does not really report
> back that you are using KVM - it reports that you are using QEMU.
> 
> Even when in nova.conf I have configured virt_type=kvm, when I run nova
> hypervisor-show XXX | grep hypervisor_type
> 
> I am presented with the following
> 
> | hypervisor_type   | QEMU
> 
> Bug?

Ah that's interesting - its certainly something that would be misleading
to operators.  It seems it is reporting the libvirt driver name, rather
than the actual hypervisor type. It'd be nice to fix it, but I'n not sure
if we can do so without creating upgrade incompatibility if people have
scheduler filters depending on that value :-(


Regards,
Daniel
-- 
|: http://berrange.com  -o-http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org  -o- http://virt-manager.org :|
|: http://autobuild.org   -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org   -o-   http://live.gnome.org/gtk-vnc :|

___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-03 Thread Sergio Cuellar Valdes 
On 3 May 2016 at 10:01, Daniel P. Berrange  wrote:

> Hello Operators,
>
> One of the things that constantly puzzles me when reading the user
> survey results wrt hypervisor is the high number of respondants
> claiming to be using QEMU (as distinct from KVM).
>
> As a reminder, in Nova saying virt_type=qemu causes Nova to use
> plain QEMU with pure CPU emulation which is many many times slower
> to than native CPU performance, while virt_type=kvm causes Nova to
> use QEMU with KVM hardware CPU acceleration which is close to native
> performance.
>
> IOW, virt_type=qemu is not something you'd ever really want to use
> unless you had no other options due to the terrible performance it
> would show. The only reasons to use QEMU are if you need non-native
> architecture support (ie running arm/ppc on x86_64 host), or if you
> can't do KVM due to hardware restrictions (ie ancient hardware, or
> running compute hosts inside virtual machines)
>
> Despite this, in the 2016 survey 10% claimed to be using QEMU in
> production & 3% in PoC and dev, in 2014 it was even higher at 15%
> in prod & 12% in PoC and 28% in dev.
>
> Personally my gut feeling says that QEMU usage ought to be in very
> low single figures, so I'm curious as to the apparent anomoly.
>
> I can think of a few reasons
>
>  1. Respondants are confused as to the difference between QEMU
> and KVM, so are saying QEMU, despite fact they are using KVM.
>
>  2. Respondants are confused as to the difference between QEMU
> and KVM, so have mistakenly configured their nova hosts to
> use QEMU instead of KVM and suffering poor performance without
> realizing their mistake.
>
>  3. There are more people than I expect who are running their
> cloud compute hosts inside virtual machines, and thus are
> unable to use KVM.
>
>  4. There are more people than I expect who are providing cloud
> hosting for non-native architectures. eg ability to run an
> arm7/ppc guest image on an x86_64 host and so genuinely must
> use QEMU
>
> If items 1 / 2 are the cause, then by implication the user survey
> is likely under-reporting the (already huge) scale of the KVM usage.
>
> I can see 3. being a likely explanation for high usage of QEMU in a
> dev or PoC scenario, but it feels unlikely for a production deployment.
>
> While 4 is technically possible, Nova doesn't really do a very good
> job at mixed guest arch hosting - I'm pretty sure there are broken
> pieces waiting to bite people who try it.
>
> Does anyone have any thoughts on this topic ?
>
> Indeed, is there anyone here who genuinely use virt_type=qemu in a
> production deployment of OpenStack who might have other reasons that
> I've missed ?
>
> Regards,
> Daniel
> --
> |: http://berrange.com  -o-http://www.flickr.com/photos/dberrange/
> :|
> |: http://libvirt.org  -o- http://virt-manager.org
> :|
> |: http://autobuild.org   -o- http://search.cpan.org/~danberr/
> :|
> |: http://entangle-photo.org   -o-   http://live.gnome.org/gtk-vnc
> :|
>
> ___
> OpenStack-operators mailing list
> OpenStack-operators@lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
>


​Hi everybody,

I'm confused too about the use of KVM or QEMU In the computes the
file​/etc/nova/nova-compute.conf has:

virt_type=kvm

The output of:

nova hypervisor-show  | grep hypervisor_type

is:

hypervisor_type   | QEMU

The virsh dumpxml of the instances shows:




/usr/bin/qemu-system-x86_64

​But according to ​this document [1], it is using QEMU emulator instead of
KVM, because it is not using /usr/bin/qemu-kvm

So I really don't know if it's using KVM or QEMU.

[1] https://libvirt.org/drvqemu.html

​Regards,
Sergio Cuéllar​


-- 
* Sergio Cuéllar │DevOps Engineer*
 KIO NETWORKS
 Mexico City Phone (52) 55 8503 2600 ext. 4335
 Mobile: 5544844298
 www.kionetworks.com
___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-03 Thread Silence Dogood 
what you should be looking for is hvm.

On Tue, May 3, 2016 at 3:20 PM, Maish Saidel-Keesing 
wrote:

> I would think that the problem is that OpenStack does not really report
> back that you are using KVM - it reports that you are using QEMU.
>
> Even when in nova.conf I have configured virt_type=kvm, when I run nova
> hypervisor-show XXX | grep hypervisor_type
>
> I am presented with the following
>
> | hypervisor_type   | QEMU
>
> Bug?
>
>
> On 03/05/16 18:01, Daniel P. Berrange wrote:
>
> Hello Operators,
>
> One of the things that constantly puzzles me when reading the user
> survey results wrt hypervisor is the high number of respondants
> claiming to be using QEMU (as distinct from KVM).
>
> As a reminder, in Nova saying virt_type=qemu causes Nova to use
> plain QEMU with pure CPU emulation which is many many times slower
> to than native CPU performance, while virt_type=kvm causes Nova to
> use QEMU with KVM hardware CPU acceleration which is close to native
> performance.
>
> IOW, virt_type=qemu is not something you'd ever really want to use
> unless you had no other options due to the terrible performance it
> would show. The only reasons to use QEMU are if you need non-native
> architecture support (ie running arm/ppc on x86_64 host), or if you
> can't do KVM due to hardware restrictions (ie ancient hardware, or
> running compute hosts inside virtual machines)
>
> Despite this, in the 2016 survey 10% claimed to be using QEMU in
> production & 3% in PoC and dev, in 2014 it was even higher at 15%
> in prod & 12% in PoC and 28% in dev.
>
> Personally my gut feeling says that QEMU usage ought to be in very
> low single figures, so I'm curious as to the apparent anomoly.
>
> I can think of a few reasons
>
>  1. Respondants are confused as to the difference between QEMU
> and KVM, so are saying QEMU, despite fact they are using KVM.
>
>  2. Respondants are confused as to the difference between QEMU
> and KVM, so have mistakenly configured their nova hosts to
> use QEMU instead of KVM and suffering poor performance without
> realizing their mistake.
>
>  3. There are more people than I expect who are running their
> cloud compute hosts inside virtual machines, and thus are
> unable to use KVM.
>
>  4. There are more people than I expect who are providing cloud
> hosting for non-native architectures. eg ability to run an
> arm7/ppc guest image on an x86_64 host and so genuinely must
> use QEMU
>
> If items 1 / 2 are the cause, then by implication the user survey
> is likely under-reporting the (already huge) scale of the KVM usage.
>
> I can see 3. being a likely explanation for high usage of QEMU in a
> dev or PoC scenario, but it feels unlikely for a production deployment.
>
> While 4 is technically possible, Nova doesn't really do a very good
> job at mixed guest arch hosting - I'm pretty sure there are broken
> pieces waiting to bite people who try it.
>
> Does anyone have any thoughts on this topic ?
>
> Indeed, is there anyone here who genuinely use virt_type=qemu in a
> production deployment of OpenStack who might have other reasons that
> I've missed ?
>
> Regards,
> Daniel
>
>
> --
> Best Regards,
> Maish Saidel-Keesing
>
> ___
> OpenStack-operators mailing list
> OpenStack-operators@lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
>
>
___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-03 Thread Maish Saidel-Keesing 
I would think that the problem is that OpenStack does not really report
back that you are using KVM - it reports that you are using QEMU.

Even when in nova.conf I have configured virt_type=kvm, when I run nova
hypervisor-show XXX | grep hypervisor_type

I am presented with the following

| hypervisor_type   | QEMU

Bug?


On 03/05/16 18:01, Daniel P. Berrange wrote:
> Hello Operators,
>
> One of the things that constantly puzzles me when reading the user
> survey results wrt hypervisor is the high number of respondants
> claiming to be using QEMU (as distinct from KVM).
>
> As a reminder, in Nova saying virt_type=qemu causes Nova to use
> plain QEMU with pure CPU emulation which is many many times slower
> to than native CPU performance, while virt_type=kvm causes Nova to
> use QEMU with KVM hardware CPU acceleration which is close to native
> performance.
>
> IOW, virt_type=qemu is not something you'd ever really want to use
> unless you had no other options due to the terrible performance it
> would show. The only reasons to use QEMU are if you need non-native
> architecture support (ie running arm/ppc on x86_64 host), or if you
> can't do KVM due to hardware restrictions (ie ancient hardware, or
> running compute hosts inside virtual machines)
>
> Despite this, in the 2016 survey 10% claimed to be using QEMU in
> production & 3% in PoC and dev, in 2014 it was even higher at 15%
> in prod & 12% in PoC and 28% in dev.
>
> Personally my gut feeling says that QEMU usage ought to be in very
> low single figures, so I'm curious as to the apparent anomoly.
>
> I can think of a few reasons
>
>  1. Respondants are confused as to the difference between QEMU
> and KVM, so are saying QEMU, despite fact they are using KVM.
>
>  2. Respondants are confused as to the difference between QEMU
> and KVM, so have mistakenly configured their nova hosts to
> use QEMU instead of KVM and suffering poor performance without
> realizing their mistake.
>
>  3. There are more people than I expect who are running their
> cloud compute hosts inside virtual machines, and thus are
> unable to use KVM.
>
>  4. There are more people than I expect who are providing cloud
> hosting for non-native architectures. eg ability to run an
> arm7/ppc guest image on an x86_64 host and so genuinely must
> use QEMU
>
> If items 1 / 2 are the cause, then by implication the user survey
> is likely under-reporting the (already huge) scale of the KVM usage.
>
> I can see 3. being a likely explanation for high usage of QEMU in a
> dev or PoC scenario, but it feels unlikely for a production deployment.
>
> While 4 is technically possible, Nova doesn't really do a very good
> job at mixed guest arch hosting - I'm pretty sure there are broken
> pieces waiting to bite people who try it.
>
> Does anyone have any thoughts on this topic ?
>
> Indeed, is there anyone here who genuinely use virt_type=qemu in a
> production deployment of OpenStack who might have other reasons that
> I've missed ?
>
> Regards,
> Daniel

-- 
Best Regards,
Maish Saidel-Keesing
___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-03 Thread Jared Wilkinson 
So forgive my lack of kvm/qemu knowledge but I couldn’t find anything on Google 
on this. If you deployed an instance of a different architecture than the 
physical CPU, wouldn’t qemu just emulate the processor (if you were in 
virt_type=kvm) mode, or would libvirt throw some error?

Thanks,
Jared

Jared Wilkinson | Infrastructure Engineer – Systems
jwilkin...@ebsco.com | (W) 205/981-4018 | (M) 205/259-9802
5724 US Highway 280 East, Birmingham, AL 35242, USA







On 5/3/16, 10:01 AM, "Daniel P. Berrange"  wrote:

>Hello Operators,
>
>One of the things that constantly puzzles me when reading the user
>survey results wrt hypervisor is the high number of respondants
>claiming to be using QEMU (as distinct from KVM).
>
>As a reminder, in Nova saying virt_type=qemu causes Nova to use
>plain QEMU with pure CPU emulation which is many many times slower
>to than native CPU performance, while virt_type=kvm causes Nova to
>use QEMU with KVM hardware CPU acceleration which is close to native
>performance.
>
>IOW, virt_type=qemu is not something you'd ever really want to use
>unless you had no other options due to the terrible performance it
>would show. The only reasons to use QEMU are if you need non-native
>architecture support (ie running arm/ppc on x86_64 host), or if you
>can't do KVM due to hardware restrictions (ie ancient hardware, or
>running compute hosts inside virtual machines)
>
>Despite this, in the 2016 survey 10% claimed to be using QEMU in
>production & 3% in PoC and dev, in 2014 it was even higher at 15%
>in prod & 12% in PoC and 28% in dev.
>
>Personally my gut feeling says that QEMU usage ought to be in very
>low single figures, so I'm curious as to the apparent anomoly.
>
>I can think of a few reasons
>
> 1. Respondants are confused as to the difference between QEMU
>and KVM, so are saying QEMU, despite fact they are using KVM.
>
> 2. Respondants are confused as to the difference between QEMU
>and KVM, so have mistakenly configured their nova hosts to
>use QEMU instead of KVM and suffering poor performance without
>realizing their mistake.
>
> 3. There are more people than I expect who are running their
>cloud compute hosts inside virtual machines, and thus are
>unable to use KVM.
>
> 4. There are more people than I expect who are providing cloud
>hosting for non-native architectures. eg ability to run an
>arm7/ppc guest image on an x86_64 host and so genuinely must
>use QEMU
>
>If items 1 / 2 are the cause, then by implication the user survey
>is likely under-reporting the (already huge) scale of the KVM usage.
>
>I can see 3. being a likely explanation for high usage of QEMU in a
>dev or PoC scenario, but it feels unlikely for a production deployment.
>
>While 4 is technically possible, Nova doesn't really do a very good
>job at mixed guest arch hosting - I'm pretty sure there are broken
>pieces waiting to bite people who try it.
>
>Does anyone have any thoughts on this topic ?
>
>Indeed, is there anyone here who genuinely use virt_type=qemu in a
>production deployment of OpenStack who might have other reasons that
>I've missed ?
>
>Regards,
>Daniel
>-- 
>|: http://berrange.com  -o-http://www.flickr.com/photos/dberrange/ :|
>|: http://libvirt.org  -o- http://virt-manager.org :|
>|: http://autobuild.org   -o- http://search.cpan.org/~danberr/ :|
>|: http://entangle-photo.org   -o-   http://live.gnome.org/gtk-vnc :|
>
>___
>OpenStack-operators mailing list
>OpenStack-operators@lists.openstack.org
>http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-03 Thread Matt Riedemann 

On 5/3/2016 10:01 AM, Daniel P. Berrange wrote:

Hello Operators,

One of the things that constantly puzzles me when reading the user
survey results wrt hypervisor is the high number of respondants
claiming to be using QEMU (as distinct from KVM).

As a reminder, in Nova saying virt_type=qemu causes Nova to use
plain QEMU with pure CPU emulation which is many many times slower
to than native CPU performance, while virt_type=kvm causes Nova to
use QEMU with KVM hardware CPU acceleration which is close to native
performance.

IOW, virt_type=qemu is not something you'd ever really want to use
unless you had no other options due to the terrible performance it
would show. The only reasons to use QEMU are if you need non-native
architecture support (ie running arm/ppc on x86_64 host), or if you
can't do KVM due to hardware restrictions (ie ancient hardware, or
running compute hosts inside virtual machines)

Despite this, in the 2016 survey 10% claimed to be using QEMU in
production & 3% in PoC and dev, in 2014 it was even higher at 15%
in prod & 12% in PoC and 28% in dev.

Personally my gut feeling says that QEMU usage ought to be in very
low single figures, so I'm curious as to the apparent anomoly.

I can think of a few reasons

 1. Respondants are confused as to the difference between QEMU
and KVM, so are saying QEMU, despite fact they are using KVM.

 2. Respondants are confused as to the difference between QEMU
and KVM, so have mistakenly configured their nova hosts to
use QEMU instead of KVM and suffering poor performance without
realizing their mistake.

 3. There are more people than I expect who are running their
cloud compute hosts inside virtual machines, and thus are
unable to use KVM.

 4. There are more people than I expect who are providing cloud
hosting for non-native architectures. eg ability to run an
arm7/ppc guest image on an x86_64 host and so genuinely must
use QEMU

If items 1 / 2 are the cause, then by implication the user survey
is likely under-reporting the (already huge) scale of the KVM usage.

I can see 3. being a likely explanation for high usage of QEMU in a
dev or PoC scenario, but it feels unlikely for a production deployment.

While 4 is technically possible, Nova doesn't really do a very good
job at mixed guest arch hosting - I'm pretty sure there are broken
pieces waiting to bite people who try it.

Does anyone have any thoughts on this topic ?

Indeed, is there anyone here who genuinely use virt_type=qemu in a
production deployment of OpenStack who might have other reasons that
I've missed ?

Regards,
Daniel



Another thought is that deployment tools are just copying what devstack 
does, or what shows up in the configs in our dsvm gate jobs, and those 
are using qemu, so they assume that's what should be used since that's 
what we gate on.


We should be more clear in our help text for the virt_type config option 
between using kvm vs qemu. Today it just says:


# Libvirt domain type (string value)
# Allowed values: kvm, lxc, qemu, uml, xen, parallels
#virt_type = kvm

It'd be good to point out the performance impacts and limitations of kvm 
vs qemu in that help text. There might already be a patch up for review 
that makes this better.


--

Thanks,

Matt Riedemann


___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

Re: [Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-03 Thread David Medberry 
The only reason I can think of is that they are doing nested VMs and don't
have the right nesting flag enabled in their base flag.

On Tue, May 3, 2016 at 9:01 AM, Daniel P. Berrange 
wrote:

> Hello Operators,
>
> One of the things that constantly puzzles me when reading the user
> survey results wrt hypervisor is the high number of respondants
> claiming to be using QEMU (as distinct from KVM).
>
> As a reminder, in Nova saying virt_type=qemu causes Nova to use
> plain QEMU with pure CPU emulation which is many many times slower
> to than native CPU performance, while virt_type=kvm causes Nova to
> use QEMU with KVM hardware CPU acceleration which is close to native
> performance.
>
> IOW, virt_type=qemu is not something you'd ever really want to use
> unless you had no other options due to the terrible performance it
> would show. The only reasons to use QEMU are if you need non-native
> architecture support (ie running arm/ppc on x86_64 host), or if you
> can't do KVM due to hardware restrictions (ie ancient hardware, or
> running compute hosts inside virtual machines)
>
> Despite this, in the 2016 survey 10% claimed to be using QEMU in
> production & 3% in PoC and dev, in 2014 it was even higher at 15%
> in prod & 12% in PoC and 28% in dev.
>
> Personally my gut feeling says that QEMU usage ought to be in very
> low single figures, so I'm curious as to the apparent anomoly.
>
> I can think of a few reasons
>
>  1. Respondants are confused as to the difference between QEMU
> and KVM, so are saying QEMU, despite fact they are using KVM.
>
>  2. Respondants are confused as to the difference between QEMU
> and KVM, so have mistakenly configured their nova hosts to
> use QEMU instead of KVM and suffering poor performance without
> realizing their mistake.
>
>  3. There are more people than I expect who are running their
> cloud compute hosts inside virtual machines, and thus are
> unable to use KVM.
>
>  4. There are more people than I expect who are providing cloud
> hosting for non-native architectures. eg ability to run an
> arm7/ppc guest image on an x86_64 host and so genuinely must
> use QEMU
>
> If items 1 / 2 are the cause, then by implication the user survey
> is likely under-reporting the (already huge) scale of the KVM usage.
>
> I can see 3. being a likely explanation for high usage of QEMU in a
> dev or PoC scenario, but it feels unlikely for a production deployment.
>
> While 4 is technically possible, Nova doesn't really do a very good
> job at mixed guest arch hosting - I'm pretty sure there are broken
> pieces waiting to bite people who try it.
>
> Does anyone have any thoughts on this topic ?
>
> Indeed, is there anyone here who genuinely use virt_type=qemu in a
> production deployment of OpenStack who might have other reasons that
> I've missed ?
>
> Regards,
> Daniel
> --
> |: http://berrange.com  -o-http://www.flickr.com/photos/dberrange/
> :|
> |: http://libvirt.org  -o- http://virt-manager.org
> :|
> |: http://autobuild.org   -o- http://search.cpan.org/~danberr/
> :|
> |: http://entangle-photo.org   -o-   http://live.gnome.org/gtk-vnc
> :|
>
> ___
> OpenStack-operators mailing list
> OpenStack-operators@lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
>
___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

[Openstack-operators] User Survey usage of QEMU (as opposed to KVM) ?

2016-05-03 Thread Daniel P. Berrange 
Hello Operators,

One of the things that constantly puzzles me when reading the user
survey results wrt hypervisor is the high number of respondants
claiming to be using QEMU (as distinct from KVM).

As a reminder, in Nova saying virt_type=qemu causes Nova to use
plain QEMU with pure CPU emulation which is many many times slower
to than native CPU performance, while virt_type=kvm causes Nova to
use QEMU with KVM hardware CPU acceleration which is close to native
performance.

IOW, virt_type=qemu is not something you'd ever really want to use
unless you had no other options due to the terrible performance it
would show. The only reasons to use QEMU are if you need non-native
architecture support (ie running arm/ppc on x86_64 host), or if you
can't do KVM due to hardware restrictions (ie ancient hardware, or
running compute hosts inside virtual machines)

Despite this, in the 2016 survey 10% claimed to be using QEMU in
production & 3% in PoC and dev, in 2014 it was even higher at 15%
in prod & 12% in PoC and 28% in dev.

Personally my gut feeling says that QEMU usage ought to be in very
low single figures, so I'm curious as to the apparent anomoly.

I can think of a few reasons

 1. Respondants are confused as to the difference between QEMU
and KVM, so are saying QEMU, despite fact they are using KVM.

 2. Respondants are confused as to the difference between QEMU
and KVM, so have mistakenly configured their nova hosts to
use QEMU instead of KVM and suffering poor performance without
realizing their mistake.

 3. There are more people than I expect who are running their
cloud compute hosts inside virtual machines, and thus are
unable to use KVM.

 4. There are more people than I expect who are providing cloud
hosting for non-native architectures. eg ability to run an
arm7/ppc guest image on an x86_64 host and so genuinely must
use QEMU

If items 1 / 2 are the cause, then by implication the user survey
is likely under-reporting the (already huge) scale of the KVM usage.

I can see 3. being a likely explanation for high usage of QEMU in a
dev or PoC scenario, but it feels unlikely for a production deployment.

While 4 is technically possible, Nova doesn't really do a very good
job at mixed guest arch hosting - I'm pretty sure there are broken
pieces waiting to bite people who try it.

Does anyone have any thoughts on this topic ?

Indeed, is there anyone here who genuinely use virt_type=qemu in a
production deployment of OpenStack who might have other reasons that
I've missed ?

Regards,
Daniel
-- 
|: http://berrange.com  -o-http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org  -o- http://virt-manager.org :|
|: http://autobuild.org   -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org   -o-   http://live.gnome.org/gtk-vnc :|

___
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators