Open Stack

I have the feeling that the discussion is diverging a bit.

John is correctly stating that with pluggable IPAM it should be possible,
in theory to write a driver which interfaces to any DHCP server, providing
it the information it needs to distribute addresses assigned by Neutron.
This is possible in theory even if the DHCP server does not expose any sort
of API - as an example you might write a driver which writes into
configuration files and then restarts or reloads the DHCP server as needed,
which is not very different from what we do today with Dnsmasq.

However, it seems to me that the gist of Padmanabhan's question is whether
neutron should allow completely out of band IP address management. Neutron
is evolving in a direction where this would be possible only if one
considers it as a simple service for building virtualised layer-2 segments.
Unless some form of interaction with the entity which manages IPs is
envisioned, it would not be possible to use neutron's security and routing
services.
Also, another aspect to factor in is nova. I have not checked this in a
while, but I seem to recall it was not able to boot a VM with network
interfaces if it did not obtain also IP addresses for such interfaces - and
those IPs are indeed fetched from neutron's port info.
If I got Padmanabhan's point right then we need to discuss if and how out
of band address mgmt and distribution should be supported in Neutron.

Regards,
Salvatore

On 5 February 2015 at 15:20, John Belamaric <jbelamaric at infoblox.com> wrote:

>  The pluggable IPAM [1] is intended to support the scenario you described
> below. That is, a fixed IP is provided for the port by IPAM, and then the
> DHCP server is programmed to return that IP for that MAC or DUID (for
> IPv6). If I understand correctly, your concern then is:
>
>  1) The DHCP server may not allow such programming;
> 2) The DHCP server may not be able to provide an IP allocation via any
> means other than DHCP (ie, no API to pre-allocate the IP)
>
>  I would think then, what you are asking for is the ability to attach an
> interface to a layer 2 network without also giving it an IP in any
> subnet(s) that are on that network. Seems like a reasonable request, but I
> am not sure what would be involved in implementing that.
>
>  Additionally, we would need some mechanism to update the port with the
> IP after it is allocated via the DHCP server. Otherwise, I don't think all
> the iptables rules and such will be properly configured, so you probably
> won't be able to pass traffic beyond the local subnet. Not sure on that
> though.
>
>
>   [1] -
> http://specs.openstack.org/openstack/neutron-specs/specs/kilo/neutron-ipam.html
>
>
>
>   From: Padmanabhan Krishnan <kprad1 at yahoo.com>
> Reply-To: Padmanabhan Krishnan <kprad1 at yahoo.com>
> Date: Thursday, February 5, 2015 at 1:47 AM
> To: John Belamaric <jbelamaric at infoblox.com>, "OpenStack Development
> Mailing List (not for usage questions)" <openstack-dev at lists.openstack.org
> >
> Subject: Re: [openstack-dev] [Neutron] fixed ip info shown for port even
> when dhcp is disabled
>
>    Hi John,
> Sure, this may not be in the scope of IPAM as originally proposed in [1].
> Am just trying to see if there can be a solution for this scenario using
> IPAM.
>
>  What I had in mind is also a centrally managed IPAM and not each VM
> allocating its own IP address. Let me clarify.
>
> It's not an uncommon requirement to use a commercial DHCP server in an
> enterprise/DC. I am referring to it as external DHCP server from
> Openstack's perspective.
> One scenario of deployment is when IPAM is done through Openstack and in
> which case, the <MAC, IP, optionally segment> can be programmed in the
> external DHCP server, assuming they have standard API's for that. Then, the
> VM will get the openstack assigned IP address from the DHCP server when it
> boots up. You also suggested this and sure, it's not a problem here.
>
>  Now, another scenario and that's of interest is when IPAM itself is done
> directly using the commercial DHCP server for various reasons. I am just
> trying to see how this model will work (or can be extended to work) for
> this case. What's the role of pluggable IPAM [1] in this scenario? If at
> all there's a way to tell the external DHCP server, do your own allocation
> and return an IP address for me for this <MAC, segment>, then this IP
> address can be returned during the allocate_ip API by the pluggable IPAM?
> But, I am not sure, an external DHCP server will have this flexibility.
> Then, in such scenarios, the only way is to not allocate an IP address
> during create_port. When the VM comes up and sends a DHCP request the
> commercial DHCP server responds with an address. The pluggable IPAM then
> can use its own method, and when it discovers the IP address of the VM, it
> can call to update_port with the IP address.
> Any other way?
>
>  [1] -
> http://specs.openstack.org/openstack/neutron-specs/specs/kilo/neutron-ipam.html
>
>  Thanks,
> Paddu
>
>
>   On Tuesday, February 3, 2015 8:57 AM, John Belamaric <
> jbelamaric at infoblox.com> wrote:
>
>
>   Hi Paddu,
>
>  I think this is less an issue of the pluggable IPAM than it is the
> Neutron management layer, which requires an IP for a port, as far as I
> know. If the management layer is updated to allow a port to exist without a
> known IP, then an additional IP request type could be added to represent
> the placeholder you describing.
>
>  However, doing so leaves IPAM out of the hands of Neutron and out of the
> hands of the external (presumably authoritative) IPAM system. This could
> lead to duplicate IP issues since each VM is deciding its own IP without
> any centralized coordination. I wouldn't recommend this approach to
> managing your IP space.
>
>  John
>
>   From: Padmanabhan Krishnan <kprad1 at yahoo.com>
> Reply-To: Padmanabhan Krishnan <kprad1 at yahoo.com>
> Date: Wednesday, January 28, 2015 at 4:58 PM
> To: John Belamaric <jbelamaric at infoblox.com>, "OpenStack Development
> Mailing List (not for usage questions)" <openstack-dev at lists.openstack.org
> >
> Subject: Re: [openstack-dev] [Neutron] fixed ip info shown for port even
> when dhcp is disabled
>
>    Some follow up questions on this.
>
>   In the specs, i see that during a "create_port",  there's provisions to
> query the external source by  "Pluggable IPAM" to return the IP.
>  This works fine for cases where the external source (say, DHCP server)
> can be queried for the IP address when a launch happens.
>
>  Is there a provision to have the flexibility of a "late IP assignment"?
>
>   I am thinking of cases, like the temporary unavailability of external
> IP source or lack of standard interfaces in which case data packet snooping
> is used to find the IP address of a VM after launch. Something similar to
> late binding of IP addresses.
>  This means the create_port  may not get the IP address from the
> pluggable IPAM. In that case, launch of a VM (or create_port) shouldn't
> fail. The Pluggable IPAM should have some provision to return something
> equivalent to "unavailable" during create_port and be able to do an
> update_port when the IP address becomes available.
>
>  I don't see that option. Please correct me if I am wrong.
>
>  Thanks,
>  Paddu
>
>
>   On Thursday, December 18, 2014 7:59 AM, Padmanabhan Krishnan <
> kprad1 at yahoo.com> wrote:
>
>
>   Hi John,
> Thanks for the pointers. I shall take a look and get back.
>
>  Regards,
> Paddu
>
>
>    On Thursday, December 18, 2014 6:23 AM, John Belamaric <
> jbelamaric at infoblox.com> wrote:
>
>
>   Hi Paddu,
>
>  Take a look at what we are working on in Kilo [1] for external IPAM.
> While this does not address DHCP specifically, it does allow you to use an
> external source to allocate the IP that OpenStack uses, which may solve
> your problem.
>
>  Another solution to your question is to invert the logic - you need to
> take the IP allocated by OpenStack and program the DHCP server to provide a
> fixed IP for that MAC.
>
>  You may be interested in looking at this Etherpad [2] that Don Kehn put
> together gathering all the various DHCP blueprints and related info, and
> also at this BP [3] for including a DHCP relay so we can utilize external
> DHCP more easily.
>
>  [1] https://blueprints.launchpad.net/neutron/+spec/neutron-ipam
>  [2] https://etherpad.openstack.org/p/neutron-dhcp-org
> [3] https://blueprints.launchpad.net/neutron/+spec/dhcp-relay
>
>  John
>
>   From: Padmanabhan Krishnan <kprad1 at yahoo.com>
> Reply-To: Padmanabhan Krishnan <kprad1 at yahoo.com>, "OpenStack Development
> Mailing List (not for usage questions)" <openstack-dev at lists.openstack.org
> >
> Date: Wednesday, December 17, 2014 at 6:06 PM
> To: "openstack-dev at lists.openstack.org" <openstack-dev at lists.openstack.org
> >
> Subject: Re: [openstack-dev] [Neutron] fixed ip info shown for port even
> when dhcp is disabled
>
>   This means whatever tools the operators are using, it need to make sure
> the IP address assigned inside the VM matches with Openstack has assigned
> to the port.
> Bringing the question that i had in another thread on the same topic:
>
>  If one wants to use the provider DHCP server and not have Openstack's
> DHCP or L3 agent/DVR, it may not be possible to do so even with DHCP
> disabled in Openstack network. Even if the provider DHCP server is
> configured with the same start/end range in the same subnet, there's no
> guarantee that it will match with Openstack assigned IP address for bulk VM
> launches or  when there's a failure case.
> So, how does one deploy external DHCP with Openstack?
>
>  If Openstack hasn't assigned a IP address when DHCP is disabled for a
> network, can't port_update be done with the provider DHCP specified IP
> address to put the anti-spoofing and security rules?
>  With Openstack assigned IP address, port_update cannot be done since IP
> address aren't in sync and can overlap.
>
>  Thanks,
> Paddu
>
>
>
> On 12/16/14 4:30 AM, "Pasquale Porreca" <pasquale.porreca at dektech.com.au>
> wrote:
>
> >I understood and I agree that assigning the ip address to the port is
> >not a bug, however showing it to the user, at least in Horizon dashboard
> >where it pops up in the main instance screen without a specific search,
> >can be very confusing.
> >
> >On 12/16/14 12:25, Salvatore Orlando wrote:
> >> In Neutron IP address management and distribution are separated
> >>concepts.
> >> IP addresses are assigned to ports even when DHCP is disabled. That IP
> >> address is indeed used to configure anti-spoofing rules and security
> >>groups.
> >>
> >> It is however understandable that one wonders why an IP address is
> >>assigned
> >> to a port if there is no DHCP server to communicate that address.
> >>Operators
> >> might decide to use different tools to ensure the IP address is then
> >> assigned to the instance's ports. On XenServer for instance one could
> >>use a
> >> guest agent reading network configuration from XenStore; as another
> >> example, older versions of Openstack used to inject network
> >>configuration
> >> into the instance file system; I reckon that today's configdrive might
> >>also
> >> be used to configure instance's networking.
> >>
> >> Summarising I don't think this is a bug. Nevertheless if you have any
> >>idea
> >> regarding improvements on the API UX feel free to file a bug report.
> >>
> >> Salvatore
> >>
> >> On 16 December 2014 at 10:41, Pasquale Porreca <
> >> pasquale.porreca at dektech.com.au> wrote:
> >>>
> >>> Is there a specific reason for which a fixed ip is bound to a port on a
> >>> subnet where dhcp is disabled? it is confusing to have this info shown
> >>> when the instance doesn't have actually an ip on that port.
> >>> Should I fill a bug report, or is this a wanted behavior?
> >>>
> >>> --
> >>> Pasquale Porreca
> >>>
> >>> DEK Technologies
> >>> Via dei Castelli Romani, 22
> >>> 00040 Pomezia (Roma)
> >>>
> >>> Mobile +39 3394823805
> >>> Skype paskporr
> >>>
> >>> _______________________________________________
> >>> OpenStack-dev mailing list
> >>> OpenStack-dev at lists.openstack.org
> >>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
> >>>
> >>
> >>
> >>
> >> _______________________________________________
> >> OpenStack-dev mailing list
> >> OpenStack-dev at lists.openstack.org
> >> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
> >>
> >
> >--
> >Pasquale Porreca
> >
> >DEK Technologies
> >Via dei Castelli Romani, 22
> >00040 Pomezia (Roma)
> >
> >Mobile +39 3394823805
> >Skype paskporr
> >
> >_______________________________________________
> >OpenStack-dev mailing list
> >OpenStack-dev at lists.openstack.org
> >http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>
>
>
>
>
>
>
>
>
> __________________________________________________________________________
> OpenStack Development Mailing List (not for usage questions)
> Unsubscribe: OpenStack-dev-request at lists.openstack.org?subject:unsubscribe
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20150205/5aee0efd/attachment.html>