[163233] in North American Network Operators' Group
Re: why does dail-up or pppoe access always has session-timeout ?
daemon@ATHENA.MIT.EDU (Warren Bailey)
Thu May 30 13:49:52 2013
From: Warren Bailey <wbailey@satelliteintelligencegroup.com>
To: "Valdis.Kletnieks@vt.edu" <Valdis.Kletnieks@vt.edu>, Joe
<sj_hznm@hotmail.com>
Date: Thu, 30 May 2013 17:49:29 +0000
In-Reply-To: <16280.1369935452@turing-police.cc.vt.edu>
Cc: NANOG <nanog@nanog.org>
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
It also probably has something to do with oversubscription. Providers
generally allocated trunks (most dial up providers I knew used Livingston
Port Masters), however their subscriber base was much larger than the
number of phone lines available to take incoming calls. If you time out
idle users, you have more phone lines available to take calls. Even in the
BBS days, we still had to wait our turn.. Usually it was a redial until we
got a carrier.
I'll agree with the configuration aspect, but I really think it has much
more to do with resource allocation in a telco environment (n:1 oversub).
//warren
On 5/30/13 10:37 AM, "Valdis.Kletnieks@vt.edu" <Valdis.Kletnieks@vt.edu>
wrote:
>On Thu, 30 May 2013 09:10:21 -0000, Joe said:
>> a question obsessed me for a long time. "why my pppoe connection to
>> internet has a max session time, even if every thing goes ok? "
>
>From a provider's point of view, forcing a connection to re-establish
>itself
>every few days means that if you ever have to roll out a change, you
>don't end
>up with *That* *One* *User* who stays connected for weeks or even months
>with
>the old parameters.
>