You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@trafficcontrol.apache.org by ocket 8888 <oc...@gmail.com> on 2020/03/23 15:42:03 UTC

Multi-interface servers (blueprint)

I have a PR open with a blueprint for a new feature, servers with multiple
network interfaces. I don't want to try to explain too much since a
blueprint exists, so if that sounds interesting go check it out:
https://github.com/apache/trafficcontrol/pull/4521

Some things - particularly regarding Traffic Monitor behavior - are about
to change a bit, but the rest should at least be pretty close to the
current vision.

Re: Multi-interface servers (blueprint)

Posted by Jeremy Mitchell <mi...@gmail.com>.
Looks really good and thorough, Brennan. Many of my comments were already
added to the PR and addressed. I added a couple more.

Thanks for putting this together.

Jeremy

On Tue, Mar 24, 2020 at 7:11 AM Dave Neuman <ne...@apache.org> wrote:

> Thanks Brennan!
>
> On Mon, Mar 23, 2020 at 9:42 AM ocket 8888 <oc...@gmail.com> wrote:
>
> > I have a PR open with a blueprint for a new feature, servers with
> multiple
> > network interfaces. I don't want to try to explain too much since a
> > blueprint exists, so if that sounds interesting go check it out:
> > https://github.com/apache/trafficcontrol/pull/4521
> >
> > Some things - particularly regarding Traffic Monitor behavior - are about
> > to change a bit, but the rest should at least be pretty close to the
> > current vision.
> >
>

Re: Multi-interface servers (blueprint)

Posted by Dave Neuman <ne...@apache.org>.
Thanks Brennan!

On Mon, Mar 23, 2020 at 9:42 AM ocket 8888 <oc...@gmail.com> wrote:

> I have a PR open with a blueprint for a new feature, servers with multiple
> network interfaces. I don't want to try to explain too much since a
> blueprint exists, so if that sounds interesting go check it out:
> https://github.com/apache/trafficcontrol/pull/4521
>
> Some things - particularly regarding Traffic Monitor behavior - are about
> to change a bit, but the rest should at least be pretty close to the
> current vision.
>