- Test - lists.nanog.org

Re: How relable does the Internet need to be? (Was: Re: Converged Network Threat)
by W.D.McKinney 26 Feb '04

26 Feb '04

>-----Original Message----- >From: Steve Gibbard [mailto:scg@gibbard.org] >Sent: Thursday, February 26, 2004 12:30 AM >To: nanog(a)merit.edu >Subject: How relable does the Internet need to be? (Was: Re: Converged Network Threat) <snipped> >>So, it appears that among general infrastructure we depend on, there are >probably the following reliability thresholds: > >Employees not being able to get to work due to snow: two to three days per >year. >Berkeley storm sewers: overflow two to three days per year. >Residential Electricity: out two to three hours per year. >Cell phone service: Somewhat better than nine fives of reliability ;) >Landline phone service: I haven't noticed an outage on my home lines in a >few years. >Natural gas: I've never noticed an outage. > >How Internet service fits into that of course depends on how you're >accessing the Net. The T-Mobile GPRS card I got recently seems >significantly less reliable than my cell phone. My SBC DSL line is almost >to the reliability level of my landline phone or natural gas service, >except that the DSL router in my basement doesn't work when electric power >is out. I'm probably poorly qualified to talk about the end-user >experience on the networks I actually work on, even if I had permission >to. Like pretty much everybody else here, I'm always interested in doing >better on reliability. And, like many of my neighbors, I'd like to be >able to store stuff on my basement floor. In comparison to a lot of other >infrastructure we depend on, it seems to me the Internet is already doing >pretty well. > >-Steve > > With BPL on the horizon and the Electric Utils looking to de-regulate in some areas, it will be interesting to watch infrastructure adapt accordingly. I think the Internet is doing pretty well save some IOS code problems from time to time, and the typical root server hicups. Dee

3 2

Re: How relable does the Internet need to be? (Was: Re: Converged Network Threat)
by W.D.McKinney 26 Feb '04

26 Feb '04

Thanks for pointing that out. That was the wrong way to describe my standpoint. Frequent changes in DNS across the board, including edge servers make connections seem non-working, when in reality it is a mis-configured DNS zone. So whether Dee >-----Original Message----- >From: Joe Abley [mailto:jabley@isc.org] >Sent: Thursday, February 26, 2004 12:57 AM >To: 'W.D.McKinney' >Cc: nanog(a)merit.edu >Subject: Re: How reliable does the Internet need to be? (Was: Re: Converged Network Threat) > > > >On 26 Feb 2004, at 08:46, W.D.McKinney wrote: > >> I think the Internet is doing pretty well save some IOS code problems >> from time to time, and the typical root server hicups. > >I'm interested to know what you mean by "typical root server hicups". >I'm trying to think of an incident which left the Internet generally >unable to receive answers to queries on the root zone, but I can't >think of one. > >By "typical", do you mean "non-existent"? > > >Joe > >

1 0

Re: New Draft Document: De-boganising New Address Blocks
by Randy Bush 25 Feb '04

25 Feb '04

[ nick has trouble posting, so ... ] Date: Wed, 25 Feb 2004 00:27:00 -0500 From: Nick Feamster <feamster(a)lcs.mit.edu> Subject: Re: New Draft Document: De-boganising New Address Blocks To: nanog(a)nanog.org User-Agent: Mutt/1.4.1i On Tue, Feb 24, 2004 at 06:28:48PM +0100, Daniel Karrenberg wrote: > > Why can't ISPs subscribe to a feed of all new > > RIPE allocations in near real-time? > > Personally I think this is a great idea and if we hear from a lot of > operators actually willing to take such feeds it may become reality > beyond volunteer efforts like the Team CYMRU one. However there are a > number of serious issues with something like this, not the least of > which are the liability issues in case this goes wrong very dynamically > and semi-automatedly. > > It is certainly something to progress if there is enough interest. > > However I think the current proposal shold go ahead too because the false > positives are a real problem that needs to be addressed quickly. > fyi, I have written a configuration checking tool that checks for a configuration's conformance to the Cymru bogon list. See: http://nms.lcs.mit.edu/bgp/rolex/ for more information. The tool also checks for various other errors (summarized at http://nms.lcs.mit.edu/bgp/rolex/tests.html) I also have a writeup that describes the tool in further detail, as well as an empirical evaluation that supports these observations about bogon filtering practices (based on results of running the tool on several ASes). Let me know if you'd either (1) like a copy of this writeup or (2) want to help me generate more empirical data (i.e., want to run the tool on your configs, let me do so, etc.) Cheers, -Nick

1 0

Sprint midwest/NW backbone issues?
by Dave O'Shea 25 Feb '04

25 Feb '04

Anyone at sprint care to shed some light on a latency issue that's been going on since December? (While the SLA credits are always nice, there's something to be said for actually getting the traffic from A to B!) sl-bb20-fw>ping 144.228.241.75 Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 144.228.241.75, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 84/115/172 ms sl-bb20-fw> ... and Seattle to Dallas: sl-bb20-sea>ping 144.228.241.51 Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 144.228.241.51, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 76/76/80 ms sl-bb20-sea>

1 0

Re: T1 Customer CPE Replacement?
by Curtis Maurand 25 Feb '04

25 Feb '04

They're still in business. They've been bought out, but they're still there. Curtis On Tue, 24 Feb 2004, Sameer Khosla wrote: > Just to add my 2 cents, I have installed a lot of Openroute routers over the > years, and have had virtually no problems with them. There is a GTX 1000 > Model that is modular, for which there are T1 CSU modules available. I > believe there were DSL modules as well. There was also a GTX1500 model > which had the encryption hardware for VPN's. > > If anyone is interested, drop me a line and I'll do some digging with one of > their former SE's. He has access to a fair supply of them. > > Sameer > > ----- Original Message ----- > From: "Curtis Maurand" <curtis(a)maurand.com> > To: "Brian Bruns" <bruns(a)2mbit.com> > Cc: "Claydon, Tom" <Tom.Claydon(a)DobsonTelco.net>; <nanog(a)merit.edu> > Sent: Tuesday, February 24, 2004 1:35 PM > Subject: Re: T1 Customer CPE Replacement? > > > > > > > > I had excellent luck with OpenRoute (formerly Proteon) GT90's. They > > handle dual ethernet or T1/E1. They need an external CSU/DSU, but they > > get the job done and they're very stable. They will do NAT and most of > > the other goodies that you can think about. They also have gt900 firewall > > and they have a model with a hardware accelerator to handle cryptographic > > calculations. I installed one of the latter into a customer about 4 years > > ago and I've not had any trouble with it, except to upgrade the OS once to > > fix some wierd packet length issues surrounding IPSEC tunnels. > > > > http://www.openroute.com > > > > Curtis > > > > > > On Mon, 23 Feb 2004, Brian Bruns wrote: > > > > > > > > On Monday, February 23, 2004 3:37 PM [EST], Claydon, Tom > > > <Tom.Claydon(a)DobsonTelco.net> wrote: > > > > > > > Hello, > > > > > > > > We're looking for a good replacement for fractional T1 customers with > Cisco > > > > 1600- & 1700-series routers as their CPE. They are good routers, but > the > > > > ongoing support costs are an issue, and we need to replace them ASAP. > > > > > > > > Someone had mentioned several CPE vendors, such as Adtran and Netopia. > Are > > > > there any others, and does anyone have any pros/cons of what they're > > > > familiar with? > > > > > > > > > > > > > > I'm quite familiar with the Netopia R53xx series T1 routers. Excellent > little > > > routers for deplyoing to customers. Very reliable, and if you are > familiar > > > with the DSL routers, you'll be right at home. They have built in > > > PPTP/ATMP/IPSec VPN support (both client and server), basic routing > features, > > > filtering, NAT, one-to-one IP mapping, remote syslog logging, as well as > > > everything you'd expect in a T1 router (fractional T1 support, HDLC, > PPP, > > > FrameRelay, etc). Theres also a 56k dialup backup module which is > handy. > > > > > > > > > > -- > > -- > > Curtis Maurand > > mailto:curtis@maurand.com > > http://www.maurand.com > > > > > -- -- Curtis Maurand mailto:curtis@maurand.com http://www.maurand.com

1 0

Re: Converged Networks Threat (Was: Level3 Outage)
by Matthew Crocker 25 Feb '04

25 Feb '04

> > Yesterday we witnessed a large scale failure that has yet to be > attributed to configuration, software, or hardware; however one need > look no further than the 168.0.0.0/6 thread, or the GBLX customer who > leaked several tens of thousands of their peers' routes to GBLX shortly > This should be rewritten 'Or GLBX who LET one of their customers leak several tens of thousands of the peers routes...'. I'm sorry, a network should be able to protect itself from its users and customers. BGP filters are not that hard to figure out and peer prefix limits should be part of every config. Don't trust the guy at the other end of the pipe to do the right thing. -Matt

1 0

RE: Converged Networks Threat (Was: Level3 Outage)
by Sean Crandall 25 Feb '04

25 Feb '04

>From Jared: > I keep hear of Frame-Relay and ATM signaling that is > going to happen in large providers MPLS cores. That's right, > your "safe" TDM based services, will be transported over > someones IP backbone first. > This means if they don't protect their IP network, the TDM > services could fail. These types of CES services are not > just limited to Frame and ATM. > (Did anyone with frame/atm/vpn services from Level3 > experience the same outage?) We use Level3 for IP transit and transport (both DS-3 and Ethernet over MPLS (via Martini)) all over the country. As with everyone else, we saw the problems with the transit traffic out of SJC and ATL. However, our transport services were not affected at all by the problems. In fact, I just ended up sending my Level3-SJC bound traffic to LAX via Level3 which was going through the same equipment as the transit traffic which was having problems. >From Pete: > From this, it can be deduced that reducing unneccessary > system complexity and shortening the strings of pearls that > make up the system contribute to better availablity and > resiliency of the system. Diversity works both ways in this > equation. It lessens the probablity of same failure hitting > majority of your boxes but at the same time increases the > knowledge needed to understand and maintain the whole system. > > I would vote for the KISS principle if in doubt. I agree. Granted the string of pearls is always going to be pretty long, but there are definitely is a trend from what I have seen with customers to make the string longer than it needs to be. -Sean Sean P. Crandall VP Engineering Operations MegaPath Networks Inc. 6691 Owens Drive Pleasanton, CA 94588 (925) 201-2530 (office) (925) 201-2550 (fax)

1 0

RE: possible L3 issues
by Arjan Lugtenberg 25 Feb '04

25 Feb '04

Here at planet (AS8737) we also having problems reaching msn/hotmail/messenger. Seems that C&W are also having problems reaching microsoft?? regards, Arjan -----Original Message----- From: owner-nanog(a)merit.edu [mailto:owner-nanog@merit.edu]On Behalf Of kwallace(a)pcconnection.com Sent: maandag 23 februari 2004 23:53 To: nanog(a)merit.edu Subject: possible L3 issues anyone else seeing high latency via L3 , especially the west coast ? - Keith

8 7

RE: New Draft Document: De-boganising New Address Blocks
by Michael.Dillon＠radianz.com 25 Feb '04

25 Feb '04

>> Timothy Brown wrote: >> I disagree with the view that it is a hack. >> It's no more a hack than using a DNS feed; >I concur with this. Besides, from the pragmatic side of the "consumer", >if it does solve a problem (albeit short or medium term) I don't care >much if it's a "hack". Then we all agree. The Cymru bogon feed is a hack. The completewhois feed is a hack. In fact, any third party feed that puports to identify authentic IP allocations is a hack. The only fix for this is to get the addressing authorities to provide an authoritative feed. That probably means first getting the RIRs to do it and then ICANN to fill in the gaps. As to whether the mechanisms used by bogon feeds are hacks or carefully crafted technology, well, my rule of thumb is that if it can completely automate the process while still allowing human intervention to make judgement calls on all changes to network configuration, then it is carefully crafted technology. I'm afraid that by this measure, using BGP feed as the mechanism is a hack because it involves plugging a 3rd party directly into your routing architecture. A directory service like DNS or LDAP is closer to the carefully crafted technology because it can be plugged into OSS systems that allow humans to validate and release any network changes. --Michael Dillon

1 0

RE: New Draft Document: De-boganising New Address Blocks
by Michel Py 25 Feb '04

25 Feb '04

William, > william(at)elan.net wrote: > [http://www.cymru.com/BGP/bogon-rs.html] > Unfortunetly this is kind-of a bgp hack and as has > been already mentioned it needs very carefull > implemention This is not positive thinking. I don't consider this a hack (if it is, then the draft that proposes to re-use ORF mechanisms is even worse). I'm not interested in what things were designed to do, I'm interested in what things _can_ do. I use [Cymru] myself as I am within the applicability domain. I use your stuff too. However, you might be borderline to abusing co-author privileges here; Team Cymru has lead the route-server project for long, and although the bogon list is by nature a lot less controversial than what you do the bottom line is the acceptance of Cymru's feed is far greater than Completewhois. Please don't use draft-py-idr-redisfilter-01.txt as a platform to compete with other co-authors; collaborative competition is sound, please stay within its borders. Thanks Michel.

2 2