- Test - lists.nanog.org

Re: What is the limit? (was RE: multi-homing fixes)
by Daniel Senie 30 Aug '01

30 Aug '01

At 11:10 PM 8/29/01, Vadim Antonov wrote: >On Wed, 29 Aug 2001, Andrew Partan wrote: > > > I have proposed to various router vendors the possibility of giving > > them a chilled water feed instead of lots of cool air. At the > > moment they seem to not need it, but I would not be surprized to > > find something like this needed at some point. > >Err. Water and electricvity make a dangerous mix. And this was not a problem in IBM Mainframe computers because? I'm not registering an opinion one way or the other at this point on whether routers should consider other forms of cooling, but using water or other liquids to cool electronics is not a new concept. Properly engineered, there is no particular danger. ----------------------------------------------------------------- Daniel Senie dts(a)senie.com Amaranth Networks Inc. http://www.amaranth.com

7 6

Re: FIBER CUT: Dallas to West Coast
by Sean Donelan 30 Aug '01

30 Aug '01

On Wed, 29 August 2001, Patrick Greenwell wrote: > On Wed, 29 Aug 2001, Brian Whalen wrote: > > Routing around the problem, what a concept. More backup for the 2xt1 shop > > to contimue doing it.. > > I'm still awaiting the list of providers that never have a circuit go > down. :-) Every carrier has had a circuit go down. The difference is the carrier's response, in particular how well they keep their customers informed. On the other hand, there is a significant statistical difference between major carriers of service affecting cable cuts and the mean time to repair. Which might be somewhat of a surprise because the physical cuts are more or less evenly distributed among the carriers, on a pro-rata basis. Among "major" carriers (i.e. with more than 20,000 route miles active) Sprint has the fewest service affecting cable cuts AT&T is the best at keeping customer's informed

5 7

Re: What is the limit? (was RE: multi-homing fixes)
by Joe Loiacono 30 Aug '01

30 Aug '01

APNICs approach (FYI): Hi James, Thanks for your mail and questions. Answers in-line below. At 07:08 PM 29/08/2001 +1000, James Spenceley wrote: >http://www.apnic.net/meetings/12/docs/proposal-multihome-assign.html > >Are companies with existing historically assigned address space eligible for >additional space via the small multi-homing assignment policy ? Anyone is eligible. >Should there be a limit of 1 of these assignments made per 'entity', with >renumbering occurring if further address space is required ? For clarification, the current proposal did not specify any limitation on the number of such assignments. What would be the motivation for this? >If so should we consider reserving the next larger block for a period of >time, to account for possible growth ? No reservations will be made. It is generally assumed that applicants will be paying a one-off service fee, rather than an ongoing membership based fee. This is in fact identical to the way that PI assignments can currently be obtained. The proposal restricts the assignments to (about to be) multihomed orgs. >If entity has a 'small multi-homing' assignment and later joins as a member >and qualifies for a /20 PA allocation, are they required to renumber from >the multi-homing assignment ? Yes. The assumption would be that the organisation whose network has grown, would then take the shorter portable prefix would then announce that prefix instead of the longer one. Maybe this should be made more explicit in the policy? >Should 'small multi-homing' assignments be made from a specific (defined) >netblock ? I think there is a choice here. We can use "swamp" space, found mostly in 202/8 which by definition does not contain large ranges of address space, or we can take a range from the less "swampy" space ie. 218/8 and use that. My feeling is that it would be better to use the 202/8 range. >There is significant risk in assigning greater than /20 space from existing >APNIC allocated blocks given that operators often use RIR minium allocation >guides for filtering, if the minimum assignment/allocation for all current >APNIC blocks becomes /23 or /24 this may encourage the deaggregation of >existing PA blocks. I'm not sure I follow the logic. As Geoff pointed out, there is already significant de-aggregation. Despite this, in de-aggregating you still run the increased risk of exposure to more severe route flap dampening parameters as well as filters. The objective in filtering is to limit the size of the routing table, so I dont see this policy encouraging ISPs to adjusting their filters to allow longer prefixes, if anything, rather the other way round. regards Anne APNIC --- * APNIC-TALK: General APNIC Discussion List * * To unsubscribe: send "unsubscribe" to apnic-talk-request(a)apnic.net *

2 1

Re: What is the limit? (was RE: multi-homing fixes)
by Larry Sheldon 30 Aug '01

30 Aug '01

> > I have proposed to various router vendors the possibility of giving > > them a chilled water feed instead of lots of cool air. At the > > moment they seem to not need it, but I would not be surprized to > > find something like this needed at some point. > > Err. Water and electricvity make a dangerous mix. That sounds a bit FUDish--lots of electricity is generated using water (most of it is, actually, given that steam is part of the cycle even in most nukes). And we have used water-cooled CPU's for a lot of years. (And as a point of interest, distilled water at neutral ph is a reasonably good insulator, if I recall correctly.)

3 2

None
by David Barak 30 Aug '01

30 Aug '01

Charles Sprickman wrote: >The internet is different now, there's no >comparison to what it was in '94 or '96. Those >episodes were more >embarrasments than anything else. Would UUNet or ATT >or Genuity think >twice about upgrading routers if there was impending >danger of a lack of >memory or cpu? In this market, with capital expenditures WAY down, I would expect non-upgrade solutions to be more likely. >Can a small ISP in Florida still clobber the whole >internet? Learning experiences, that's all... You'd be surprised at the number of substantive ISPs who can be impacted by customer mistakes... David Barak "Quis custodes ipsos custodiet?" - Juvenal __________________________________________________ Do You Yahoo!? Get email alerts & NEW webcam video instant messaging with Yahoo! Messenger http://im.yahoo.com

1 0

backhoe damage (karachi, pakistan)
by Jim Mercer 30 Aug '01

30 Aug '01

a backhoe (or potentially a group of labourers with picks and shovels) managed to sever a pile of cable near one of the CO's in Karachi, Pakistan. this outage is effecting several ISP's and voice services. ETA for fix is 14:00 GMT 8^) -- [ Jim Mercer jim(a)reptiles.org +1 416 410-5633 ] [ Now with more and longer words for your reading enjoyment. ]

1 0

Re: What is the limit? (was RE: multi-homing fixes)
by smd＠clock.org 30 Aug '01

30 Aug '01

The Internet *FAILED* in 1994. The number of prefixes carried globally exceeded the ability of the global routing system to carry it, and as a result, some parts of the world deliberately summarized away whole large-scale ISPs merely to survive. The Internet *FAILED* again in 1996, when the dynamicism in the global routing system exceeded the ability of many border routers to handle, and as a result, some networks (not just core ones) deliberately made the Internet less quick to adjust to changes in topology. Thus, the routing system FAILED twice: once because of memory, once because of processor power. If the size or the dynamicism of the global routing system grows for a sustained period faster than the price/performance curve of EITHER memory OR processing power, the Internet will FAIL again. I do mean the Internet, and not just some pieces of it. The old saw, "the Internet detects damage and routes around it" simply isn't true, when your routing system isn't working. It was unpleasant both times. Three continents were effectively isolated from one another for a couple of days while organizing a response to the memory crisis. Three of the largest ISPs at the time were crippled on and off for days during the processing power crisis, and even when mechanisms were brought into place, relatively unimportant bugs destabilized the entire Internet from time to time. Note that the processor power issue is the one that has been scariest, since it has had small-scale failures fairly regularly. Things like selective packet drop *exist* because of the price/performance curve and engineering gap for deploying and USING more processing power. So, Moore's Law, or more specifically the underlying curve which tracks the growth of useful computational power, is exactly what we should compare with the global routing system's growth curve. Note that when Moore is doing better than the Internet, it allows for either cheaper supply of dynamic connectivity, or it allows for the deployment of more complex handling of the global NLRI. The major problem, as you have pointed out, is that processing requirement is often bursty, such as when everyone is trying to do a dynamic recovery from a router crash or major line fault. We could still use 68030s in our core routers, it's just that it'd take alot longer than it used to perform a global partition repair, which means your TCP sessions or your patience will probably time out alot more frequently. | I think that what we need to do is have a fourth group, call them Internet | Engineers for lack of a better word, come in and determine what the sign | should read. Structures built according to best known engineering practices still fall down from time to time. That's the problem in anticipating unforeseen failures. Consider yourself lucky that you haven't had to experience a multi-day degredation (or complete failure!) of service due to resource constraints. And that you haven't run into a sign that says: "please note: if you try to have an automatic partition repair in the event this path towards {AS SET} fails, your local routing system will destabilize". | Finally, we have a sixth group, call them the IETF, come in | and invent a flying car that doesn't need the bridge at all. As Randy (with his IETF hat) says: "send code". Sean.

13 18

Re: What is the limit? (was RE: multi-homing fixes)
by Philip Smith 30 Aug '01

30 Aug '01

At 06:01 30/08/2001 -0400, John Fraizer wrote: >Speaking of filtering: > >AS64602 at Mae-West from Digex (AS2548) ASPATH=2548 64602 64602 64602 >64602 64602 64602 IGP > >AS65515 at Mae-West from Verio (AS2914) ASPATH=2914 10910 10910 10910 >10910 10910 10910 11908 10530 5593 65515 IGP > >AS65515 at AADS from Verio (AS2914) ASPATH=2914 10910 10910 10910 10910 >10910 10910 11908 10530 5593 65515 IGP > >AS65515 at Paix from Verio (AS2914) ASPATH=2914 10910 10910 10910 10910 >10910 10910 11908 10530 5593 65515 IGP > >Which part of (64512 - 65535 != valid ASN) do providers not understand? > >Filter! Filter! Filter! And your favourite router vendors have a useful command/knob to remove private ASes from announcements too. It's not hard to put that on all eBGP peerings... philip --

1 0

Re: What is the limit? (was RE: multi-homing fixes) (fwd)
by Larry Sheldon 30 Aug '01

30 Aug '01

> > > Err. Water and electricvity make a dangerous mix. > > > > Ya. My air handling systems seem to deal with it OK. > > They are kept in different places. Now think how are you going to combine > hot-swappability and water cooling, and not let moisture to get out :) I don't recall seeing any mops around when the CE's replaced a board on the 1100/90's. They were not "hot-swappable"--but nobody had thought of _that_ as a good marketing drool. We had "redundant" equipment and took the broken one off-line to work on it. [snip] > As soon as you do liquid cooling the price goes up an order of magnitude. > > This is as simple as that. Be interesting to know where that came from. Particularly when you consider the "liquid-cooled" versus "can't do it" costs. > My take is that the real answer is to move from heavy, sticky electrons to > photons; the trick is to get them to interact in useful ways :) Had your finger on an operating LASER lately?

1 0

RE: What is the limit? (was RE: multi-homing fixes)
by Roeland Meyer 30 Aug '01

30 Aug '01

|> From: Randy Bush [mailto:randy@psg.com] |> Sent: Wednesday, August 29, 2001 9:18 PM |> |> > there would appear to be people who assert its not unafordable to |> > compute the routes for the current, and the forseeable network. |> |> and there are those of us who think we can probably do so iff we stop |> the explosion of /24s. Okay, let's get it back out of the realm of opinion and into requirements. I'll even go down one step more, to /25. Take the entire IP addr space, break it ALL into /25's, times the BGP table entry size, how big is that? Could 4GB of RAM do it? How about 8GB? Your average 64-bit processor can handle T-Bytes and they are affordable by your average basement dual-homers. Also, I just got a price quote for 512MB ECC PC133 DIMMS at $49US, retail. Oh yeah, that same processor runs faster than 1 GHz. That's todays *computer* technology. Router technology needs to catch up. Considering what router vendors charge, with very fat margins, for their low-tech routers, affordable tech arguments don't hold water. Routers are expensive because someone is stupid enough to pay the prices. We are getting royally ripped off there. Cisco takes special RAM because Cisco designed incompatible RAM circuitry so they could charge you more for the Cisco label. It also bolsters the FUD-storm and the after-market support vendors. BTW, you really ought to check out that site at http://www.routerrevolution.com. The reason I chose /25 is because the average small business can easily use that much space, but would have a hard time justifying a /24.

2 1