Open Stack

Egor,

I agree with what you said, but I think we need to address the problem that some clouds are lack of public IP addresses. It is not uncommon that a private cloud is running without public IP addresses, and they already figured out how to route traffics in and out. In such case, a bay doesn’t need to have floating IPs and the NodePort feature seems to work with the private IP address.

Generally speaking, I think it is useful to have a feature that allows bays to work without public IP addresses. I don’t want to end up in a situation that Magnum is unusable because the clouds don’t have enough public IP addresses.

Best regards,
Hongbin

From: Guz Egor [mailto:guz_egor at yahoo.com]
Sent: March-31-16 12:08 AM
To: OpenStack Development Mailing List (not for usage questions)
Subject: Re: [openstack-dev] [magnum] Are Floating IPs really needed for all nodes?

-1

who is going to run/support this proxy? also keep in mind that Kubernetes Service/NodePort (http://kubernetes.io/docs/user-guide/services/#type-nodeport)
functionality is not going to work without public ip and this is very handy feature.

---
Egor

________________________________
From: 王华 <wanghua.humble at gmail.com<mailto:wanghua.humble at gmail.com>>
To: OpenStack Development Mailing List (not for usage questions) <openstack-dev at lists.openstack.org<mailto:openstack-dev at lists.openstack.org>>
Sent: Wednesday, March 30, 2016 8:41 PM
Subject: Re: [openstack-dev] [magnum] Are Floating IPs really needed for all nodes?

Hi yuanying,

I agree to reduce the usage of floating IP. But as far as I know, if we need to pull
docker images from docker hub in nodes floating ips are needed. To reduce the
usage of floating ip, we can use proxy. Only some nodes have floating ips, and
other nodes can access docker hub by proxy.

Best Regards,
Wanghua

On Thu, Mar 31, 2016 at 11:19 AM, Eli Qiao <liyong.qiao at intel.com<mailto:liyong.qiao at intel.com>> wrote:

Hi Yuanying,
+1
I think we can add option on whether to using floating ip address since IP address are
kinds of resource which not wise to waste.

On 2016年03月31日 10:40, 大塚元央 wrote:
Hi team,

Previously, we had a reason why all nodes should have floating ips [1].
But now we have a LoadBalancer features for masters [2] and minions [3].
And also minions do not necessarily need to have floating ips [4].
I think it’s the time to remove floating ips from all nodes.

I know we are using floating ips in gate to get log files,
So it’s not good idea to remove floating ips entirely.

I want to introduce `disable-floating-ips-to-nodes` parameter to bay model.

Thoughts?

[1]: http://lists.openstack.org/pipermail/openstack-dev/2015-June/067213.html
[2]: https://blueprints.launchpad.net/magnum/+spec/make-master-ha
[3]: https://blueprints.launchpad.net/magnum/+spec/external-lb
[4]: http://lists.openstack.org/pipermail/openstack-dev/2015-June/067280.html

Thanks
-yuanying



__________________________________________________________________________

OpenStack Development Mailing List (not for usage questions)

Unsubscribe: OpenStack-dev-request at lists.openstack.org?subject:unsubscribe<mailto:OpenStack-dev-request at lists.openstack.org?subject:unsubscribe>

http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev



--

Best Regards, Eli Qiao (乔立勇)

Intel OTC China

__________________________________________________________________________
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: OpenStack-dev-request at lists.openstack.org?subject:unsubscribe<http://OpenStack-dev-request@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-request at lists.openstack.org<mailto: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/20160331/c48ab47b/attachment-0001.html>