Re: When did OSPF become The IGP? (was Re: BGP on dialup

John Storms (jstorms@livingston.com)
Thu, 24 Jul 1997 14:56:34 -0700

At 01:38 PM 7/24/97 -0700, Daniel Simms wrote:
>>>>>> "JS" == John Storms <jstorms@livingston.com> writes:
>
> JS> Using BGP as and IGP is like swatting flys with a sledge hammer,
>
>Elaborate, please. This seems to get right to the heart of what (and
>why) Livingston will and will not do WRT IGP.

At 01:38 PM 7/24/97 -0700, you wrote:
>>>>>> "JS" == John Storms <jstorms@livingston.com> writes:
> JS> Using BGP as and IGP is like swatting flys with a sledge hammer,
>
>Elaborate, please. This seems to get right to the heart of what (and
>why) Livingston will and will not do WRT IGP.

First, forgive my ignorance, but I don't the acronym 'WRT IGP' as I'm only
familiar with Livingston's implementation via SQA testing.

Livingston's implementation of BGP was engineered according to RFC1771 and
the latest IETF draft and accordingly will peer both internally and
externally. We support confederations (RFC1965), Route Reflection
(RFC1966), Communities (RFC1997), plus all the fixins (policies,
assumed-defaults, easy-multihome, summarizations, igp-lockstep, yadda yadda
yadda). BGP connections will go over dialup network connections (just a
TCP connection, nothing to stop it). I like the additional applications
for it that people have volunteered.

What I meant by...
> JS> Using BGP as and IGP is like swatting flys with a sledge hammer,
...is that most cases, not all, OSPF makes a better choice for an internal
gateway protocol and depends on their system needs. If you want to use BGP
as an IGP feel free to do so, we've tested that knowing that some people
will implement it in this fashion.

* * * (Scenery Change) * * *

Now, to the OSPF issue being thrown around.

--This can be easily fixed with a packet filter blocking syslog
messages--Just Kidding, sorry, I couldn't resist. Starting to get punchie.

Seriously, Livingston is checking this out. From the customers we have
talked to we've noticed that the Cisco will generate the syslog message
when simple passwords are being used. I've directed support to glean off
the valuable information from this list to added to the characteristics of
this issue. We've also been brainstorming with our OSPF engineer about the
issue.

We are still in fact-finding mode to figure out what is going on. We find
troubleshooting works efficiently if we focus on the facts and if we don't
make rash speculations. It might be us, it might not, either way its worth
our while to work the problem.

The thing that seems odd to me, and please correct me if I'm wrong, is why
does the Cisco neighbor with the Livingston if the password checksum was
incorrect?

---
jstorms@livingston.com
Diplomacy:  The art of saying good doggie
while seaching for a big rock.