Mailing List Archive

Craig, netscape couldn't find anything at the address you listed. I
found what I **think** you are talking about at http://compute.merit.edu/

What you have done looks really useful. But SPRINT is refusing to
cooperate I gather. How big a hole does this put in the usefulness of
what you are doing? With UUNET, ANS and MCI particpating, how can Sprint
justify NOT participating?

*********************************************************************
Gordon Cook, Editor & Publisher Subscriptions: Individ-ascii $85
The COOK Report on Internet Individ. hard copy $150
431 Greenway Ave, Ewing, NJ 08618 USA Small Corp & Gov't $200
(609) 882-2572 Corporate $350
Internet: cook@cookreport.com Corporate Site Lic. $650
http://pobox.com/cook/ for new COOK Report Glossary of Internet terms
*********************************************************************


On Tue, 2 Apr 1996, Craig Labovitz wrote:

>
> Using many of the suggestions and comments from the last IEPG and NANOG
> meetings, we have put together some new Internet routing statistics web
> pages. See http://www.ra.net/statistics/
>
> These pages include real-time graphs of network instability, bar charts of the
> sources of routing instability, snapshots of the Route Servers' routing
> tables, graphs of long-term trends, bogus routing reports, and IRR
> discrepancies.
>
> In improving the quality and accuracy of these reports, we need your help. The
> reports are only as good as the raw data the route servers receive in peering
> sessions. If you are not peering with the route servers, we encourage you to
> do so. Your routes will ONLY be used for statistics collection purposes
> (unless you have explicitly registered policy in the IRR and indicated that
> the RA should propagate the routes.)
>
> To peer with the route servers, send mail to rs-peer@ra.net.
>
> The RA thanks the following organizations for providing routes for statistics
> collection/analysis:
>
> ANS, Advantis, Aimnet, Alpha.Net, Alternet, Argonne, CAIS, CERFnet,
> CWInet, DIGEX, DRAnet, DREN, DXnet, Delphi, ESnet, INAP.net, INSnet,
> IOSnet, IconNet, Interpath, MCI, Nacamar, NetAxs, Netcom, Netrail,
> PIPEX, Suranet, Supernet, ThePlanet, TheWell, WIS.COM
>
>
> - Craig
>
> -------------
> Craig Labovitz labovit@merit.edu
> Merit Network, Inc. (313) 764-0252 (office)
> 4251 Plymouth Road, Suite C. (313) 747-3745 (fax)
> Ann Arbor, MI 48105-2785
>
>
>
>

No, Gordon, Sprint did not refuse to cooperate.

It is quite likely that if we were asked, it was asked of me
personally and promptly buried four hundred items deep on my
todo list.

Sean.

So sprint will coperate at least to the extent that MCI, ANS, and UUNET
are already doing so?

I'd think that is good news. Do you have any kind of time line, Sean?
June 1? July 1?

Now where is PSI??? Are they only Mae East and CIX connected? if they
don't cooperate, do they hurt the community or themselves worse by non
cooperation?

*********************************************************************
Gordon Cook, Editor & Publisher Subscriptions: Individ-ascii $85
The COOK Report on Internet Individ. hard copy $150
431 Greenway Ave, Ewing, NJ 08618 USA Small Corp & Gov't $200
(609) 882-2572 Corporate $350
Internet: cook@cookreport.com Corporate Site Lic. $650
http://pobox.com/cook/ for new COOK Report Glossary of Internet terms
*********************************************************************


On Tue, 2 Apr 1996, Sean Doran wrote:

> No, Gordon, Sprint did not refuse to cooperate.
>
> It is quite likely that if we were asked, it was asked of me
> personally and promptly buried four hundred items deep on my
> todo list.
>
> Sean.
>

| So sprint will coperate at least to the extent that MCI, ANS, and UUNET
| are already doing so?

If Craig had previously pointed out that he would like to
have gotten routes from us that would not be propagated
unless specifically requested by us, then I don't think it
would have been too difficult to have done the two line
adjustment to our already-existing peerings with the two
RSes at MAE-EAST in order for us to have shown up on the
hallowed list in his message today.

Of course, that would have also deprived you of a front-page
Cook Report scandal, and several messages here on the NANOG
list.

So, think of this as job creation at work, maybe.

| I'd think that is good news. Do you have any kind of time line, Sean?
| June 1? July 1?

Well, I went digging, and couldn't find any relevant email
on the first pass. Perhaps such a message will find its way
to the appropriate people (maybe for the second time, if we
lost things on this end), so we know what routing
information is wanted and where and when, and what exactly
it will be used for.

I see no difficulty with announcing routes to the RA
specifically for gathering the sorts of statistics they have
been thinking about. Indeed, I believe that I have made
this clear in public places a number of times in the past.

We now return you to the regularly scheduled making
of mountains out of mole-hills.

Sean.

P.S.:

SL-MAE-E uptime is 1 week, 4 days, 12 hours, 37 minutes
System restarted by power-on at 10:28:06 EST Fri Mar 22 1996

BGP neighbor is 192.41.177.166, remote AS 2885, external link
NEXT_HOP is always this router
BGP version 4, remote router ID 192.41.177.166
BGP state = Established, table version = 2186252, up for 1w0d
Last read 0:00:57, hold time is 180, keepalive interval is 60 seconds
Minimum time between advertisement runs is 30 seconds
Received 16570 messages, 0 notifications, 0 in queue
Sent 16605 messages, 0 notifications, 0 in queue
Incoming update network filter list is 112
Outgoing update network filter list is 95
Outgoing update AS path filter list is 35
Route map for incoming advertisements is RA
Connections established 4; dropped 3

[.moreover, we set the peering up more than a year ago]

Sean,

Just so you can stop searching your archives....

>Sean Doran writes:
>
> [stuff deleted]
> Well, I went digging, and couldn't find any relevant email
> on the first pass. Perhaps such a message will find its way
> to the appropriate people (maybe for the second time, if we
> lost things on this end), so we know what routing
> information is wanted and where and when, and what exactly
> it will be used for.
>
> I see no difficulty with announcing routes to the RA
> specifically for gathering the sorts of statistics they have
> been thinking about. Indeed, I believe that I have made
> this clear in public places a number of times in the past.
>

Sprintlink is already peering with the route servers at MAE-East.
Please go ahead and announce full routes to both of those route
servers at 1900 EST today, (April 3). The RA will respect your
request to not propogate the routes. The routing information
learned from Sprintlink will be used to generate reports such
as: a) the internet routing table, b) daily count of prefixes
announced and withdrawn, c) daily invalid route announcments,
and d) trends in routing information.

Since Sprintlink is not yet peering with the route servers
at the AADS, PacBell and Sprint NAPs, please fill out the
route server peering template (www.ra.net/.peering.template.html)
for each of the other three NAP locations. As soon as the
peering sessions are established, announce full routes to
those route servers. The same conditions that are described
(no propogation of routes and generation of reports) will
apply.

If you have any further questions, please send email to ra-team@merit.edu.

If there are other organizations who wish to initiate peering sessions
with the route servers, please fill out the template on the web
page or send email to ra-team@merit.edu.

This is good news, Sean.

--Elise

Elise -

Gotcha. While the timings won't be exactly exact, we
will take you up on your invitation.

Sean.

On Wed, 3 Apr 1996, Sean Doran wrote:

> Gotcha. While the timings won't be exactly exact, we
> will take you up on your invitation.

Now if NANOG had a PR flack monitoring the list he would jump up and
propose a press release...

NANOG is pleased to announce that SPRINT has agreed to provide data for
the Routing Arbiter's statistics database. This makes 5 of the 6 major
North American NSP's participating in the project to collect and analyze
statistics on traffic levels through the Internet's core. SPRINT joins
Alternet, MCI, Netcom and (hmmm was AGIS/Net99 on that list??) in
supporting the RA's project to open a window on what type and volume of
traffic is flowing through the core.

This is important because ..... (educational diatribe)

The Internet's core consists of ...... (yet another educational diatribe)

NANOG (North American Network Operators Group) is an informal forum for
NSP's and major customers to share information, problems and solutions
regarding the day to day operation of the Internet. NANOG has quarterly
meetings throughout North America and members keep in touch regularly via
the NANOG mailing list. More info is available at http://www.nanog.org
(I know the URL is wrong, but that could be changed).

Now is this an evil thing?

Find a good excuse, write a short announcement, pad it out to tow or three
pages with educational material that you know the press needs to learn
anyway, provide a URL for them to followup if they want to drill deeper
into the topic...

Michael Dillon Voice: +1-604-546-8022
Memra Software Inc. Fax: +1-604-546-3049
http://www.memra.com E-mail: michael@memra.com

> Now if NANOG had a PR flack monitoring the list he would jump up and
> propose a press release...

Thank the godess it doesn't. Well, didn't.

This is an OPERATORS' forum. I am sure there is a marketeers' forum some
place, probably lots of them. Though to judge by the performance of the net
recently, the net might use more operations and less marketing.

This list, which used to be useful to operators FOR PURPOSES OF OPERATIONS
is teetering on the verge uselessness. Can we please move the rah rah,
explanations to non-operators, ... back to com-priv? Please?

randy

On Wed, 3 Apr 1996, Randy Bush wrote:

> > Now if NANOG had a PR flack monitoring the list he would jump up and
> > propose a press release...

> This list, which used to be useful to operators FOR PURPOSES OF OPERATIONS
> is teetering on the verge uselessness. Can we please move the rah rah,
> explanations to non-operators, ... back to com-priv? Please?

Start publishing regular press releases, post them at
http://www.nanog.org and to the nanog-announce mailing list and
your problems will be solved. The non-operators will leave nanog to you.


Michael Dillon Voice: +1-604-546-8022
Memra Software Inc. Fax: +1-604-546-3049
http://www.memra.com E-mail: michael@memra.com

> > > Now if NANOG had a PR flack monitoring the list he would jump up and
> > > propose a press release...
>
> > This list, which used to be useful to operators FOR PURPOSES OF OPERATIONS
> > is teetering on the verge uselessness. Can we please move the rah rah,
> > explanations to non-operators, ... back to com-priv? Please?
>
> Start publishing regular press releases, post them at
> http://www.nanog.org and to the nanog-announce mailing list and
> your problems will be solved. The non-operators will leave nanog to you.

I'm afraid I have to throw my vote with Randy on this one.

Press Release is not the verbage we want to use.

If there were to be a www.nanog.org, it should have snapshots of important
messages in the nanog list, like

Thu Apr 4 01:05:15 MST 1996 Sprint feeds routes to RA for statistical
reasons. see URL...
ya da ya da ya da

Maybe a couple of help documents:

- Safety tips for peering at exchanges
a) how to build a filter list that won't hose everyone else
b) why you should use dampening
c) don't forget to passive-interface your interfaces...OSPF neighbors across
an exchange are bad...

- NO's guide.
a) how to tune performance on 7000 series routers
b) how important is the RADB? The RS?
c) sample routing policies currently in use and implementation pointers
d) what can I do to help with the growth of the routing table?

- vendor specific gotchas
a) Cisco bugs that will bite you in the ass every time
b) Bay Network trials and tribulations
c) gated; it looks like I could compile this config file.

- new technologies
a) caching with harvest/NS/cern; shoud I bother
b) ATM/packet shredding; what's the word?
c) muxing? SONET? what next?

- Useful URL's

so on and so forth.

Sure, it sounds like a good topic for a book, too....but these are the
operational issues which get discussed (for the most part) on this list,
and a web page should represent that.

Dave

--
Dave Siegel Sr. Network Engineer, RTD Systems & Networking
(520)623-9663 Network Consultant -- Regional/National NSPs
dsiegel@rtd.com User Tracking & Acctg -- "Written by an ISP,
http://www.rtd.com/~dsiegel/ for an ISP."

Ever taken a look at the iepg web page?


>
> If there were to be a www.nanog.org, it should have snapshots of important
> messages in the nanog list, like
>
> Thu Apr 4 01:05:15 MST 1996 Sprint feeds routes to RA for statistical
> reasons. see URL...
> ya da ya da ya da
>
> Maybe a couple of help documents:
>
> - Safety tips for peering at exchanges
> a) how to build a filter list that won't hose everyone else
> b) why you should use dampening
> c) don't forget to passive-interface your interfaces...OSPF neighbors across
> an exchange are bad...
>
> - NO's guide.
> a) how to tune performance on 7000 series routers
> b) how important is the RADB? The RS?
> c) sample routing policies currently in use and implementation pointers
> d) what can I do to help with the growth of the routing table?
>
> - vendor specific gotchas
> a) Cisco bugs that will bite you in the ass every time
> b) Bay Network trials and tribulations
> c) gated; it looks like I could compile this config file.
>
> - new technologies
> a) caching with harvest/NS/cern; shoud I bother
> b) ATM/packet shredding; what's the word?
> c) muxing? SONET? what next?
>
> - Useful URL's
>
> so on and so forth.
>
> Sure, it sounds like a good topic for a book, too....but these are the
> operational issues which get discussed (for the most part) on this list,
> and a web page should represent that.
>
> Dave
>
> --
> Dave Siegel Sr. Network Engineer, RTD Systems & Networking
> (520)623-9663 Network Consultant -- Regional/National NSPs
> dsiegel@rtd.com User Tracking & Acctg -- "Written by an ISP,
> http://www.rtd.com/~dsiegel/ for an ISP."
>


--
--bill