Re: [openstack-dev] [neutron] OVSDB native interface as default in gate jobs

2016-06-09 Thread IWAMOTO Toshihiro 
At Wed, 08 Jun 2016 03:26:59 -0700,
Clark Boylan wrote:
> 
> On Tue, Apr 5, 2016, at 08:32 PM, IWAMOTO Toshihiro wrote:
> > At Tue, 5 Apr 2016 12:57:33 -0400,
> > Assaf Muller wrote:
> > > 
> > > On Tue, Apr 5, 2016 at 12:35 PM, Sean M. Collins  
> > > wrote:
> > > > Russell Bryant wrote:
> > > >> because they are related to two different command line utilities
> > > >> (ovs-vsctl vs ovs-ofctl) that speak two different protocols (OVSDB vs
> > > >> OpenFlow) that talk to two different daemons on the system 
> > > >> (ovsdb-server vs
> > > >> ovs-vswitchd) ?
> > > >
> > > > True, they influence two different daemons - but it's really two options
> > > > that both have two settings:
> > > >
> > > > * "talk to it via the CLI tool"
> > > > * "talk to it via a native interface"
> > > >
> > > > How likely is it to have one talking via native interface and the other
> > > > via CLI?
> > > 
> > > The ovsdb native interface is a couple of cycles more mature than the
> > > openflow one, I see how some users would use one but not the other.
> > 
> > The native of_interface has been tested by periodic jobs and seems
> > pretty stable.
> > 
> > http://graphite.openstack.org/dashboard/#neutron-ovs-native
> > 
> > > > Also, if the native interface is faster, I think we should consider
> > > > making it the default.
> > > 
> > > Definitely. I'd prefer to deprecate and delete the cli interfaces and
> > > keep only the native interfaces in the long run.
> > > 
> > > >
> > > > --
> > > > Sean M. Collins
> > 
> > The native of_interface is definitely faster than the CLI alternative,
> > but (un?)fortunately that's not a performance bottleneck.
> > 
> > The transition would be a gain, but it comes with uncovering a few
> > unidentified bugs etc.
> > 
> > Anyway, I'll post an updated version of performance comparison shortly.
> 
> Going to resurrect this thread to see where we have gotten. Did an
> updated comparison ever get posted? If so I mised it. Looks like Neutron
> does have functional tests now that use the native interfaces for ovsdb
> and openflow. The change to update the default interface for ovsdb is
> still in review though.
> 

It happened on a different mail thread, and I haven't done any
comparison with ovsdb drivers.

http://lists.openstack.org/pipermail/openstack-dev/2016-April/091639.html
http://lists.openstack.org/pipermail/openstack-dev/2016-April/091642.html



> Any chance that we document these choices yet?

I've put a patch that switches to the native of_interface.
If we cannot switch the default, better documentation might help.

> Clark
> 

--
IWAMOTO Toshihiro

__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [openstack-dev] [neutron] OVSDB native interface as default in gate jobs

2016-06-08 Thread Clark Boylan 
On Tue, Apr 5, 2016, at 08:32 PM, IWAMOTO Toshihiro wrote:
> At Tue, 5 Apr 2016 12:57:33 -0400,
> Assaf Muller wrote:
> > 
> > On Tue, Apr 5, 2016 at 12:35 PM, Sean M. Collins  wrote:
> > > Russell Bryant wrote:
> > >> because they are related to two different command line utilities
> > >> (ovs-vsctl vs ovs-ofctl) that speak two different protocols (OVSDB vs
> > >> OpenFlow) that talk to two different daemons on the system (ovsdb-server 
> > >> vs
> > >> ovs-vswitchd) ?
> > >
> > > True, they influence two different daemons - but it's really two options
> > > that both have two settings:
> > >
> > > * "talk to it via the CLI tool"
> > > * "talk to it via a native interface"
> > >
> > > How likely is it to have one talking via native interface and the other
> > > via CLI?
> > 
> > The ovsdb native interface is a couple of cycles more mature than the
> > openflow one, I see how some users would use one but not the other.
> 
> The native of_interface has been tested by periodic jobs and seems
> pretty stable.
> 
> http://graphite.openstack.org/dashboard/#neutron-ovs-native
> 
> > > Also, if the native interface is faster, I think we should consider
> > > making it the default.
> > 
> > Definitely. I'd prefer to deprecate and delete the cli interfaces and
> > keep only the native interfaces in the long run.
> > 
> > >
> > > --
> > > Sean M. Collins
> 
> The native of_interface is definitely faster than the CLI alternative,
> but (un?)fortunately that's not a performance bottleneck.
> 
> The transition would be a gain, but it comes with uncovering a few
> unidentified bugs etc.
> 
> Anyway, I'll post an updated version of performance comparison shortly.

Going to resurrect this thread to see where we have gotten. Did an
updated comparison ever get posted? If so I mised it. Looks like Neutron
does have functional tests now that use the native interfaces for ovsdb
and openflow. The change to update the default interface for ovsdb is
still in review though.

Any chance that we document these choices yet?

Clark

__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [openstack-dev] [neutron] OVSDB native interface as default in gate jobs

2016-04-05 Thread IWAMOTO Toshihiro 
At Tue, 5 Apr 2016 12:57:33 -0400,
Assaf Muller wrote:
> 
> On Tue, Apr 5, 2016 at 12:35 PM, Sean M. Collins  wrote:
> > Russell Bryant wrote:
> >> because they are related to two different command line utilities
> >> (ovs-vsctl vs ovs-ofctl) that speak two different protocols (OVSDB vs
> >> OpenFlow) that talk to two different daemons on the system (ovsdb-server vs
> >> ovs-vswitchd) ?
> >
> > True, they influence two different daemons - but it's really two options
> > that both have two settings:
> >
> > * "talk to it via the CLI tool"
> > * "talk to it via a native interface"
> >
> > How likely is it to have one talking via native interface and the other
> > via CLI?
> 
> The ovsdb native interface is a couple of cycles more mature than the
> openflow one, I see how some users would use one but not the other.

The native of_interface has been tested by periodic jobs and seems
pretty stable.

http://graphite.openstack.org/dashboard/#neutron-ovs-native

> > Also, if the native interface is faster, I think we should consider
> > making it the default.
> 
> Definitely. I'd prefer to deprecate and delete the cli interfaces and
> keep only the native interfaces in the long run.
> 
> >
> > --
> > Sean M. Collins

The native of_interface is definitely faster than the CLI alternative,
but (un?)fortunately that's not a performance bottleneck.

The transition would be a gain, but it comes with uncovering a few
unidentified bugs etc.

Anyway, I'll post an updated version of performance comparison shortly.

--
IWAMOTO Toshihiro

__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [openstack-dev] [neutron] OVSDB native interface as default in gate jobs

2016-04-05 Thread Russell Bryant 
On Tue, Apr 5, 2016 at 12:57 PM, Assaf Muller  wrote:

> On Tue, Apr 5, 2016 at 12:35 PM, Sean M. Collins 
> wrote:
> > Russell Bryant wrote:
> >> because they are related to two different command line utilities
> >> (ovs-vsctl vs ovs-ofctl) that speak two different protocols (OVSDB vs
> >> OpenFlow) that talk to two different daemons on the system
> (ovsdb-server vs
> >> ovs-vswitchd) ?
> >
> > True, they influence two different daemons - but it's really two options
> > that both have two settings:
> >
> > * "talk to it via the CLI tool"
> > * "talk to it via a native interface"
> >
> > How likely is it to have one talking via native interface and the other
> > via CLI?
>
> The ovsdb native interface is a couple of cycles more mature than the
> openflow one, I see how some users would use one but not the other.


​and they use separate Python libraries, as well (ovs vs ryu).



>
> >
> > Also, if the native interface is faster, I think we should consider
> > making it the default.
>
> Definitely. I'd prefer to deprecate and delete the cli interfaces and
> keep only the native interfaces in the long run.
>

​+1.

-- 
Russell Bryant​
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [openstack-dev] [neutron] OVSDB native interface as default in gate jobs

2016-04-05 Thread Assaf Muller 
On Tue, Apr 5, 2016 at 12:35 PM, Sean M. Collins  wrote:
> Russell Bryant wrote:
>> because they are related to two different command line utilities
>> (ovs-vsctl vs ovs-ofctl) that speak two different protocols (OVSDB vs
>> OpenFlow) that talk to two different daemons on the system (ovsdb-server vs
>> ovs-vswitchd) ?
>
> True, they influence two different daemons - but it's really two options
> that both have two settings:
>
> * "talk to it via the CLI tool"
> * "talk to it via a native interface"
>
> How likely is it to have one talking via native interface and the other
> via CLI?

The ovsdb native interface is a couple of cycles more mature than the
openflow one, I see how some users would use one but not the other.

>
> Also, if the native interface is faster, I think we should consider
> making it the default.

Definitely. I'd prefer to deprecate and delete the cli interfaces and
keep only the native interfaces in the long run.

>
> --
> Sean M. Collins
>
> __
> OpenStack Development Mailing List (not for usage questions)
> Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [openstack-dev] [neutron] OVSDB native interface as default in gate jobs

2016-04-05 Thread Sean M. Collins 
Russell Bryant wrote:
> because they are related to two different command line utilities
> (ovs-vsctl vs ovs-ofctl) that speak two different protocols (OVSDB vs
> OpenFlow) that talk to two different daemons on the system (ovsdb-server vs
> ovs-vswitchd) ?

True, they influence two different daemons - but it's really two options
that both have two settings:

* "talk to it via the CLI tool"
* "talk to it via a native interface"

How likely is it to have one talking via native interface and the other
via CLI? 

Also, if the native interface is faster, I think we should consider
making it the default.

-- 
Sean M. Collins

__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [openstack-dev] [neutron] OVSDB native interface as default in gate jobs

2016-04-04 Thread Russell Bryant 
On Mon, Apr 4, 2016 at 5:32 PM, Sean M. Collins  wrote:

> Inessa Vasilevskaya wrote:
> > different configurations of of_interface and ovsdb_interface options
> > (dsvm-fullstack [2] and rally tests are by now all I can think of).
>
> Wait, we have *two* different configuration options???
>
> WHY WHY WHY
>

​because they are related to two different command line utilities
(ovs-vsctl vs ovs-ofctl) that speak two different protocols (OVSDB vs
OpenFlow) that talk to two different daemons on the system (ovsdb-server vs
ovs-vswitchd) ?

With that said, I see no reason to keep two methods if one is clearly
better.  I just don't think combining them into one config option makes
much sense.
​

-- 
Russell Bryant
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [openstack-dev] [neutron] OVSDB native interface as default in gate jobs

2016-04-04 Thread Eugene Nikanorov 
May be you just don't have enough pain, Sean :)

I'd agree that these should be coalesced, with a deprecation period then...

E.

On Mon, Apr 4, 2016 at 2:32 PM, Sean M. Collins  wrote:

> Inessa Vasilevskaya wrote:
> > different configurations of of_interface and ovsdb_interface options
> > (dsvm-fullstack [2] and rally tests are by now all I can think of).
>
> Wait, we have *two* different configuration options???
>
> WHY WHY WHY
>
> --
> Sean M. Collins
>
> __
> OpenStack Development Mailing List (not for usage questions)
> Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [openstack-dev] [neutron] OVSDB native interface as default in gate jobs

2016-04-04 Thread Sean M. Collins 
Inessa Vasilevskaya wrote:
> different configurations of of_interface and ovsdb_interface options
> (dsvm-fullstack [2] and rally tests are by now all I can think of).

Wait, we have *two* different configuration options??? 

WHY WHY WHY

-- 
Sean M. Collins

__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [openstack-dev] [neutron] OVSDB native interface as default in gate jobs

2016-04-04 Thread Inessa Vasilevskaya 
Yes, of_interface has 'native' option as well.
I have plans to dig into the matter further by looking at native ovsdb
interface state (a naive change of config didn't pass some functional
tests, need investigation [1]) and running some performance tests in
different configurations of of_interface and ovsdb_interface options
(dsvm-fullstack [2] and rally tests are by now all I can think of).
And yes, I have plans to document my findings once I'll become more
acquainted with the subject.

[1] https://review.openstack.org/#/c/299655/
[2] https://review.openstack.org/#/c/299380/

Inessa

On Thu, Mar 31, 2016 at 9:47 PM, Clark Boylan  wrote:

> On Tue, Mar 29, 2016, at 07:34 PM, Kevin Benton wrote:
> > I'm not aware of any issues. Perhaps you can propose a patch to just
> > change
> > the default in Neutron to that interface and people can -1 if there are
> > any
> > concerns.
>
> Yes, please do. This is one of the nice things about pre merge testing,
> we can all see how it works upfront before committing to it.
>
> As far as concerns go I got curious about this and did some digging
> around documentation and found basically zero docs. It is possible I
> don't know where to look, but I would expect that we should be
> documenting the behavior of the various interface options? However, this
> is a small concern and can be addressed if/when the changes are made to
> Neutron.
>
> In my digging I did find that there is also an of_interface that
> defaults to ovs-ofctl, does this also have a native interface that could
> be used?
>
> Clark
>
> __
> OpenStack Development Mailing List (not for usage questions)
> Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [openstack-dev] [neutron] OVSDB native interface as default in gate jobs

2016-03-31 Thread Clark Boylan 
On Tue, Mar 29, 2016, at 07:34 PM, Kevin Benton wrote:
> I'm not aware of any issues. Perhaps you can propose a patch to just
> change
> the default in Neutron to that interface and people can -1 if there are
> any
> concerns.

Yes, please do. This is one of the nice things about pre merge testing,
we can all see how it works upfront before committing to it.

As far as concerns go I got curious about this and did some digging
around documentation and found basically zero docs. It is possible I
don't know where to look, but I would expect that we should be
documenting the behavior of the various interface options? However, this
is a small concern and can be addressed if/when the changes are made to
Neutron.

In my digging I did find that there is also an of_interface that
defaults to ovs-ofctl, does this also have a native interface that could
be used?

Clark

__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [openstack-dev] [neutron] OVSDB native interface as default in gate jobs

2016-03-30 Thread Russell Bryant 
On Tue, Mar 29, 2016 at 10:34 PM, Kevin Benton  wrote:

> I'm not aware of any issues. Perhaps you can propose a patch to just
> change the default in Neutron to that interface and people can -1 if there
> are any concerns.
>

FWIW, the ovs Python library lacked Python 3 support until very recently.
I finished it and have it all merged into ovs master, but it won't be in an
official release until just before the Newton release (based on the current
rough plan).  In the meantime, there's a dev snapshot on PyPI that includes
it (2.6.0.dev1).​


​I'm not aware of any other problems.​

-- 
Russell Bryant
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [openstack-dev] [neutron] OVSDB native interface as default in gate jobs

2016-03-29 Thread Kevin Benton 
I'm not aware of any issues. Perhaps you can propose a patch to just change
the default in Neutron to that interface and people can -1 if there are any
concerns.

On Tue, Mar 29, 2016 at 4:32 PM, Inessa Vasilevskaya <
ivasilevsk...@mirantis.com> wrote:

> Hi all,
>
> I spent some time researching the current state of native ovsdb interface
> feature, which has been fully implemented and merged since Liberty [1][2].
> I was pretty much impressed by the expected performance improvement (as per
> spec [1], some interesting research on ovs-vsctl + rootwrap also done here
> [3]). Preliminary test results also showed that native interface does quite
> well.
>
> So my question is - why don’t we make native interface the default option
> for voting gate jobs? Are there any caveats or security issues that I am
> unaware of?
>
> Regards,
>
> Inessa
>
> [1]
> https://specs.openstack.org/openstack/neutron-specs/specs/kilo/vsctl-to-ovsdb.html
>
> [2] https://blueprints.launchpad.net/neutron/+spec/vsctl-to-ovsd
> 
> [3]
> http://blog.otherwiseguy.com/replacing-ovs-vsctl-calls-with-native-ovsdb-in-neutron/
>
> __
> OpenStack Development Mailing List (not for usage questions)
> Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>
>
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

[openstack-dev] [neutron] OVSDB native interface as default in gate jobs

2016-03-29 Thread Inessa Vasilevskaya 
Hi all,

I spent some time researching the current state of native ovsdb interface
feature, which has been fully implemented and merged since Liberty [1][2].
I was pretty much impressed by the expected performance improvement (as per
spec [1], some interesting research on ovs-vsctl + rootwrap also done here
[3]). Preliminary test results also showed that native interface does quite
well.

So my question is - why don’t we make native interface the default option
for voting gate jobs? Are there any caveats or security issues that I am
unaware of?

Regards,

Inessa

[1]
https://specs.openstack.org/openstack/neutron-specs/specs/kilo/vsctl-to-ovsdb.html

[2] https://blueprints.launchpad.net/neutron/+spec/vsctl-to-ovsd

[3]
http://blog.otherwiseguy.com/replacing-ovs-vsctl-calls-with-native-ovsdb-in-neutron/
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev