suggestions about load-balance/cluster
Jorge Pereira
jpereiran at gmail.com
Thu Oct 22 03:11:15 CEST 2015
Hi Matthew,
Good to know that more people are working with it. It would be
great to us share ideas. I am looking for another solutions to after
research can choose the best approach to be applied.
I think that use of keyed-balance will be the best way. But, I really
would like to hear more people talking about that.
About the ways to based the requests, I have used the Acct-Session-Id
which is the key of storage. below a example.
"%{redis: HSET %{Acct-Session-Id} Client-Shortname %{client:shortname}}"
My gTalk is jpereiran at gmail.com, feel free to add and talk about some ideas.
--
Jorge Pereira
On Wed, Oct 21, 2015 at 8:59 PM, Matthew Newton <mcn4 at leicester.ac.uk> wrote:
> On Wed, Oct 21, 2015 at 03:48:53PM -0200, Jorge Pereira wrote:
>> home_server_pool auth_loadbalance {
>> type = load-balance
>> home_server = auth_node_1
>> home_server = auth_node_N
>> virtual_server = handle-node-auth
>> }
> ...
>
>> Doubts: What the best approach based in the "type" of the pool? Maybe
>> the "keyed-balance" sounds better? I really appreciate
>> all suggestions and tips about my approach.
>
> I'm looking at doing the same at the moment.
>
> load-balance should be OK with e.g. plain PAP etc.
>
> As soon as you move to EAP then it'll break, so keyed-balance
> should be the only real option AFAICT, though I have considered
> ideas using unlang to choose the proxy server and storing it based
> on Calling-Station-Id and User-Name in a local redis server with a
> timeout. But I usually look for the complicated solutions...
>
> My likely plan if I go ahead with this is to use keyed-balance
> probably with Calling-Station-Id (and possibly also User-Name) as
> the key source.
>
> Matthew
>
>
> --
> Matthew Newton, Ph.D. <mcn4 at le.ac.uk>
>
> Systems Specialist, Infrastructure Services,
> I.T. Services, University of Leicester, Leicester LE1 7RH, United Kingdom
>
> For IT help contact helpdesk extn. 2253, <ithelp at le.ac.uk>
> -
> List info/subscribe/unsubscribe? See http://www.freeradius.org/list/users.html
More information about the Freeradius-Users
mailing list