>
>
> The actual number would be considerably smaller as there were large
> (for some definition of large) block assignments of ASNs <~1000 or so
> to various academic networking entities such as NSFNet and regional
> networks as well as other Federal/Military networking organisations.
>
> -dorian
>
>
Well, for one data point, I was issued 3492 around Spring of 1994.
Does no one remember EGP? ASNs are MUCH older than BGP. And we were
using BGPv3 prior to the existence of V4. We used BGPv4 back in the days
when Tony Li would chastise us for reporting a bug in a 10 day old Cisco
build saying that we could not expect BGPv4 code over a week old to
work. He felt that we should deploy new code daily.
The big push was to have v4 available before the old PRDB was frozen by
Merit/NSFnet. (And, who remembers the PRDB?)
Does no one remember EGP? ASNs are MUCH older than BGP. And we were
using BGPv3 prior to the existence of V4. We used BGPv4 back in the days
when Tony Li would chastise us for reporting a bug in a 10 day old Cisco
build saying that we could not expect BGPv4 code over a week old to
work. He felt that we should deploy new code daily.
The big push was to have v4 available before the old PRDB was frozen by
Merit/NSFnet. (And, who remembers the PRDB?)
--
I caught the trailing edge of V3. I remember announcing 129 prefixes to
Sprintlink, one for our B, and 128 for our /17 from C-space
My name is Cathy and I used to EGP. wow that's a blast from the past.
In the late 80s the NSFNet used EGP. That's why the RADB came to be. EGP
only tolerated one route to any destination. Otherwise lots of fun loops.
Kevin, when I configured ESNet to go from EGP to BGP (in the early 90s) we
hit what I call a Yakov limit. We were like the 64th BGP peer and the
NSFNET shut down completely. It was a total riot.
To be fair, that was for folks on the isp-geeks mailing list, who were effectively doing alpha test with me. I was fixing about 1 significant bug per day and doing at least one release per day. 10 day old code was missing at least 10 fixes... And that was BGP3. BGP4 was the next developer.
To be fair, that was for folks on the isp-geeks mailing list, who were effectively doing alpha test with me. I was fixing about 1 significant bug per day and doing at least one release per day. 10 day old code was missing at least 10 fixes... And that was BGP3. BGP4 was the next developer.
I always liked seeing the string "tli" in the IOS bundle in those days.
To be fair, that was for folks on the isp-geeks mailing list, who were
effectively doing alpha test with me. I was fixing about 1 significant
bug per day and doing at least one release per day. 10 day old code was
missing at least 10 fixes... And that was BGP3. BGP4 was the next
developer.
I always liked seeing the string "tli" in the IOS bundle in those days.
Yes, I remember EGP every well. When we built the NSFNET T1 backbone in 1987, EGP was the only available routing protocol for exterior routing. We deployed it and used EGP to exchange routing information with the connected regional networks. Initially, it worked fine but then when the routing table and traffic grew, we observed that every 3 minutes, the network performance got a hit. After some investigation, we discovered that it was due to the fact that EGP did routing updates every 3 minutes by flooding the whole routing table to the peer and the process overwhelmed router processor. At that time, the processor on the router did both routing and forwarding.
Fortunately, Yakov of IBM, Kirk from Cisco and we Merit were working on the development and testing of BGP, which was intended to replace EGP. BGP does incremental routing updates i.e. it sends its peer the delta whenever routing topology changes rather than flooding its peer with the whole routing table every 3 minutes. It saved a lot of processing power. In addition, it reduces routing convergence time since BGP sends its neighbors the updates whenever changes occurs. In the case of EGP, it may take as long as close to 3 minutes after a route change before the routing table got updated. In addition, BGP has loop detection capability due to its inclusion of AS path information. These were the technical reasons to replace EGP with BGP at the time.
We worked with regional network reps and started to convert NSFNET to regional peers from EGP to BGP in early 1990s. I also created the BGP Deployment Work Group at IETF to push the deployment of BGP in the whole Internet.
And that was BGP3. BGP4 was the next developer.