North American Network Operators Group
Date Prev | Date Next | Date Index | Thread Index | Author Index | Historical

Re: Long-haul protected services: (was: Re: Bee attack, fiber cut, 7-hour outage)

  • From: Bill Stewart
  • Date: Sat Sep 29 01:16:47 2007
  • Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; bh=GUONr96Kkm7V/QrcH0pu5rf2i49k3pF11eskMV5QplY=; b=LcYBVL4N3GBZhACXcOKDkBGHCYJY6uf/zbrIztmHCWa61vrmyL0kmzKZwK38lMSrhEm1plIeR9lbveMqdCTO4EyEVc/e60vRN10IuxBiN8FAbc92xIyeD0N/aiIQsvjQMo3nX+Jm84+Aq/04pxIG0T2cJMTEvCV/YjgfRqxV/fQ=
  • Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=ZDBWMxZcTyxKlD/IRdO7eRjWANYs9wUuvmNxl/YsN+Bton5+3uu8zKFmCnfY7PxEjFMIJNeF3m0JYy2BX0a9hcNiBMC0Ue8f0n9zpz7WzTKRuJpKYfCNHDxg7hocKqjCAkGDrVJIVMeVKA0kQcnhWDX2US2UNKTVJ3qhdHS1kTw=
On 9/21/07, Deepak Jain <[email protected]> wrote:
> However, when I see "Location of Maintenance: France" and a 5 minute
> outage for a protected SONET service on a supposedly redundant, high
> quality International voice/data network... well, let's just say I'm not
> impressed -- on 36 hrs notice, no less.
>
> I can't do anything with respect to an SLA since there is advanced
> notice, but isn't it reasonable to assume that in this day-and-age
> running a properly protected T3 isn't *that hard* anymore????
> Especially in advance -- you know, shunt the traffic to one your other
> circuits because, you know, you are supposed to have this massive network.

Typically for a service like that, the carrier would have one or more
long-haul rings,
either SONET or DWDM with the SONET on top of it or something,
interconnected to access rings from a local access carrier at each end,
either close to the customer endpoint or possibly back at regional
interconnects.
The long-haul carrier might not control the local access carrier
except through an SLA,
so it could be that the local carrier messed up.  One hopes that the
interconnections
between the networks are also at diverse POPs (at least for big
countries like France),
but it is possible for the interconnections to fail clumsily.

The "shunt the traffic to other circuits" approach is naively correct,
but I've seen more than one case of "discover that one side of an
access ring has failed
by shunting the traffic from the other side to do maintenance and
having a customer call you
a couple of minutes later to report an outage at their headquarters",
which is absolutely never supposed to happen, of course :-)

What a 5-minute outage sounds like to me is "The early-90s T3-based
restoration system
recovered the circuit on an end-to-end basis using DACSs instead of
SONET restoring it",
but I'd be surprised to see than happening at a newer carrier like
Vendor L who built most of their network after SONET technology became
affordable.

----
             Thanks;     Bill

Note that this isn't my regular email account - It's still experimental so far.
And Google probably logs and indexes everything you send it.