[123478] in North American Network Operators' Group
Re: Alcatel-Lucent
daemon@ATHENA.MIT.EDU (Phil Bedard)
Tue Mar 9 15:58:03 2010
In-Reply-To: <4B968D78.8010105@gmail.com>
From: Phil Bedard <bedard.phil@gmail.com>
Date: Tue, 9 Mar 2010 15:57:28 -0500
To: piotr sawicki <piotr.sawicki.pl@gmail.com>
Cc: NANOG list <nanog@nanog.org>
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
I've done some recent testing and while the BGP download time isn't =
blazing fast, it can load 400k routes and propagate them to 20 other =
peers in a few minutes. Certainly not 2 hours. :) I've also done =
quite a bit of interop testing with the other main vendors as well and =
have yet to run into anything major. =20
Phil=20
On Mar 9, 2010, at 1:03 PM, piotr sawicki wrote:
>>=20
>> The worst thing in it was bgp proto .. Router was unable to withstand =
20+ peering sessions , most of that outgoing bgp session to customers , =
a few peerings , and only 1v2 incoming upstream providers
>> When there was instability/surge in bgp updates , router was able to =
break itself tcp sess. Dwnld bgp table (150,000prefix) took 2h or more =
...
>> Things done in hardware should be working although ( bridge .. , =
maybe label switching , vpls ) Tech support is very weak. Expect =
problems with interoperability .
>> Sorry to say that . It was 3+ years time ago , maybe they improved =
themself .. :)
>>=20
>=20
> Hi,
>=20
> I must say we had very old timos 2.0R17 , as you now use 8.0 I guess =
..
> As most of my complains is against software , this may be improved - i =
hope it is :)
> Hardware is solid, 'hard to kill'
> Sorry for a bit preliminary assumptions about 7750SR platform .
>=20
> // regards PiotrSawicki.
>=20
