Re: [openstack-dev] [Neutron][Magnum] Container + OpenStack Integration

2015-05-21 Thread Kevin Benton
Hi,

I wasn't able to make it to the session but from the etherpad I saw
that the VLAN-aware VMs mentioned there. Would that be enough to cover
this particular use-case?

On Wed, May 20, 2015 at 3:28 AM, Adrian Otto adrian.o...@rackspace.com wrote:
 Magnum and Neutron Teams,

 If you are in Vancouver today and can help design or implement native
 networking support to obviate the need for overlay networks when using
 containers, please join this design session:

 http://sched.co/3CjK (3:30 PM, Room 303)

 This will be based on feedback gathered at the Fishbowl session earlier in
 the day: http://sched.co/3B0z (11:00 AM, Room 306)

 Thanks,

 Adrian

 __
 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




-- 
Kevin Benton

__
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][Magnum] Container + OpenStack Integration

2015-05-21 Thread Fox, Kevin M
Since docker 1.7 is suposed to support advanced networking, could that just be 
plugged into neutron?

Thanks,
Kevin


From: Brenden Blanco
Sent: Thursday, May 21, 2015 6:40:37 AM
To: OpenStack Development Mailing List (not for usage questions)
Subject: Re: [openstack-dev] [Neutron][Magnum] Container + OpenStack Integration

Hi,

Just brainstorming here, but IMHO it would be good to keep it extensible and 
not to lock in VLAN as the sole container VIF mechanism. For instance, there 
is the ipvlan [1] technology that was purpose-built to be more appropriate for 
container systems. This tech actually comes from some of the Google folks that 
deploy containers at scale, if my memory serves correctly. So, we may want to 
start with VLAN as the reference implementation but should leave it open to try 
out an alternative such as the above in the near future. Perhaps in practice 
this implies that the vid field of the neutron side isn't strictly an int 
field, not sure.

Cheers,
Brenden Blanco

[1] https://lwn.net/Articles/620087/

On Wed, May 20, 2015 at 11:21 PM, Kevin Benton 
blak...@gmail.commailto:blak...@gmail.com wrote:
Hi,

I wasn't able to make it to the session but from the etherpad I saw
that the VLAN-aware VMs mentioned there. Would that be enough to cover
this particular use-case?

On Wed, May 20, 2015 at 3:28 AM, Adrian Otto 
adrian.o...@rackspace.commailto:adrian.o...@rackspace.com wrote:
 Magnum and Neutron Teams,

 If you are in Vancouver today and can help design or implement native
 networking support to obviate the need for overlay networks when using
 containers, please join this design session:

 http://sched.co/3CjK (3:30 PM, Room 303)

 This will be based on feedback gathered at the Fishbowl session earlier in
 the day: http://sched.co/3B0z (11:00 AM, Room 306)

 Thanks,

 Adrian

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




--
Kevin Benton

__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: 
openstack-dev-requ...@lists.openstack.org?subject:unsubscribehttp://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][Magnum] Container + OpenStack Integration

2015-05-21 Thread Brenden Blanco
Hi,

Just brainstorming here, but IMHO it would be good to keep it extensible
and not to lock in VLAN as the sole container VIF mechanism. For
instance, there is the ipvlan [1] technology that was purpose-built to be
more appropriate for container systems. This tech actually comes from some
of the Google folks that deploy containers at scale, if my memory serves
correctly. So, we may want to start with VLAN as the reference
implementation but should leave it open to try out an alternative such as
the above in the near future. Perhaps in practice this implies that the vid
field of the neutron side isn't strictly an int field, not sure.

Cheers,
Brenden Blanco

[1] https://lwn.net/Articles/620087/

On Wed, May 20, 2015 at 11:21 PM, Kevin Benton blak...@gmail.com wrote:

 Hi,

 I wasn't able to make it to the session but from the etherpad I saw
 that the VLAN-aware VMs mentioned there. Would that be enough to cover
 this particular use-case?

 On Wed, May 20, 2015 at 3:28 AM, Adrian Otto adrian.o...@rackspace.com
 wrote:
  Magnum and Neutron Teams,
 
  If you are in Vancouver today and can help design or implement native
  networking support to obviate the need for overlay networks when using
  containers, please join this design session:
 
  http://sched.co/3CjK (3:30 PM, Room 303)
 
  This will be based on feedback gathered at the Fishbowl session earlier
 in
  the day: http://sched.co/3B0z (11:00 AM, Room 306)
 
  Thanks,
 
  Adrian
 
 
 __
  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
 



 --
 Kevin Benton

 __
 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][Magnum] Container + OpenStack Integration

2015-05-21 Thread Antoni Segura Puimedon
On Thu, May 21, 2015 at 8:04 AM, Fox, Kevin M kevin@pnnl.gov wrote:

  Since docker 1.7 is suposed to support advanced networking, could that
 just be plugged into neutron?


Yes, we'd be looking at doing a docker libnetwork driver that uses the
remote driver interface.

The remote driver would request the ports/networks to neutron and then use
that information and the port type to plug the port into the container in
the appropriate way.



 Thanks,
 Kevin

 --
 *From:* Brenden Blanco
 *Sent:* Thursday, May 21, 2015 6:40:37 AM
 *To:* OpenStack Development Mailing List (not for usage questions)
 *Subject:* Re: [openstack-dev] [Neutron][Magnum] Container + OpenStack
 Integration

  Hi,

  Just brainstorming here, but IMHO it would be good to keep it extensible
 and not to lock in VLAN as the sole container VIF mechanism. For
 instance, there is the ipvlan [1] technology that was purpose-built to be
 more appropriate for container systems. This tech actually comes from some
 of the Google folks that deploy containers at scale, if my memory serves
 correctly. So, we may want to start with VLAN as the reference
 implementation but should leave it open to try out an alternative such as
 the above in the near future. Perhaps in practice this implies that the vid
 field of the neutron side isn't strictly an int field, not sure.

  Cheers,
 Brenden Blanco

  [1] https://lwn.net/Articles/620087/

 On Wed, May 20, 2015 at 11:21 PM, Kevin Benton blak...@gmail.com wrote:

 Hi,

 I wasn't able to make it to the session but from the etherpad I saw
 that the VLAN-aware VMs mentioned there. Would that be enough to cover
 this particular use-case?

 On Wed, May 20, 2015 at 3:28 AM, Adrian Otto adrian.o...@rackspace.com
 wrote:
  Magnum and Neutron Teams,
 
  If you are in Vancouver today and can help design or implement native
  networking support to obviate the need for overlay networks when using
  containers, please join this design session:
 
  http://sched.co/3CjK (3:30 PM, Room 303)
 
  This will be based on feedback gathered at the Fishbowl session earlier
 in
  the day: http://sched.co/3B0z (11:00 AM, Room 306)
 
  Thanks,
 
  Adrian
 
 
 __
  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
 



 --
 Kevin Benton

 __
 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 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