Re: IPv6 transits (Was: Cogent input)
On Jun 14, 2009, at 6:04 PM, Jeroen Massar wrote: For people trying to find the list, check: http://www.sixxs.net/faq/connectivity/?faq=ipv6transit Since when has Level3 offered native IPv6? I nag our rep SE's just about every month on when and right now AFAIK it's still just tunnels. -- Robert Blayzor, BOFH INOC, LLC rblay...@inoc.net http://www.inoc.net/~rblayzor/
Re: IPv6 transits (Was: Cogent input)
i can confirm that Level(3), at least in Madrid area is only offering tunneled IPv6. --- Nuno Vieira nfsi telecom, lda. nuno.vie...@nfsi.pt Tel. (+351) 21 949 2300 - Fax (+351) 21 949 2301 http://www.nfsi.pt/ - Robert Blayzor rblayzor.b...@inoc.net wrote: On Jun 14, 2009, at 6:04 PM, Jeroen Massar wrote: For people trying to find the list, check: http://www.sixxs.net/faq/connectivity/?faq=ipv6transit Since when has Level3 offered native IPv6? I nag our rep SE's just about every month on when and right now AFAIK it's still just tunnels. -- Robert Blayzor, BOFH INOC, LLC rblay...@inoc.net http://www.inoc.net/~rblayzor/
Re: Cogent input
Speaking of the devil:
Comcast plans to enter into broadband IPv6 technical trials later
this year and into 2010, {Barry Tishgart, VP of Internet Services for
Comcast} said. Planning for general deployment is underway.
http://tech.slashdot.org/story/09/06/18/1417201/Comcast-To-Bring-IPv6-To-Residential-US-In-2010
http://www.internetnews.com/infra/article.phpr/3825696/Comcast+Embraces+IPv6.htm
http://news.google.com/news/more?um=1ned=uscf=allncl=dsg_EPKdMw3ISjMxORbZRq061pu7M
On Thu, Jun 18, 2009 at 1:52 AM, Kevin Hodlekevin.ho...@gmail.com wrote:
Hi Justin,
Just FYI - Global Crossing can currently deliver dual stack/native v6
transit in downtown KC,MO. You can either colo with them at 1100 Main St, or
possibly have them haul a wave to one of the other major downtown carrier
hotels they have strands running through / into (1102 Grand/Bryant and 324
E. 11th St/Oak Towers come to mind, not to mention Level3's suite in 1100
Walnut right across the street).
Cheers,
Kevin Hodle
On Fri, Jun 12, 2009 at 8:13 AM, Justin Shore jus...@justinshore.comwrote:
John van Oppen wrote:
NTT (2914) and GBLX (3549) both do native v6... most everyone else on
the tier1 list does tunnels. :(
There are some nice tier2 networks who do native v6, tiscali and he.net
come to mind.
Let me rephrase that. :-) I know of no tier-Ns that offer any native v6
services here in the Midwest (central Kansas) including L3 which only has a
best effort pilot program using tunnels. There might be more options in KC
or OKC but not here that I'm aware of...
Justin
--
|| Kevin Hodle
||
|| 913-780-3959 (Primary)
|| 913-626-7197 (Mobile)
PGP KeyID [0xBBDE8ED7]
fingerprint [3E1B 1F10 938E A831 8CF2 670C 1329 0B8B BBDE 8ED7]
Re: Cogent input
On Jun 17, 2009, at 1:17 AM, Michael K. Smith wrote: On 6/11/09 7:37 AM, Steve Bertrand st...@ibctech.ca wrote: Stephen Kratzer wrote: And, they have no plans to support IPv6. Ouch! I hope this is a non-starter for a lot of folks. Steve To quote Randy, I encourage all my competitors to do this. Simply untrue, at the Peering BOF yesterday Cogent said they are rolling this out. -- kris
Re: Cogent input
On Wed, Jun 17, 2009 at 02:14:31AM -0400, kris foster wrote: Simply untrue, at the Peering BOF yesterday Cogent said they are rolling this out. They saw my How to deploy IPv6 in 30 minutes or less tutorial on Sunday and apparently it actually worked. Unfortunately I neglected to mention the important acquire connectivity to the global routing table step (I assumed it was implied, but I guess it wasn't), but if you're down with their 5 v6 routes as transit you should be golden. :) -- Richard A Steenbergen r...@e-gerbil.net http://www.e-gerbil.net/ras GPG Key ID: 0xF8B12CBC (7535 7F59 8204 ED1F CC1C 53AF 4C41 5ECA F8B1 2CBC)
Re: Cogent input
2009/6/11 Tore Anderson t...@linpro.no And, they have no plans to support IPv6. I have been promised, in writing, that they will provide us with native IPv6 transit before the end of the year. I'm based in Europe, though. Perhaps they're more flexible and customer-friendly here than in the US? We too have been informed Q3/Q4 09, in Europe and NA. Regards,
Re: Cogent input
On Wed, Jun 17, 2009 at 2:24 AM, Richard A Steenbergenr...@e-gerbil.net wrote: On Wed, Jun 17, 2009 at 02:14:31AM -0400, kris foster wrote: Simply untrue, at the Peering BOF yesterday Cogent said they are rolling this out. They saw my How to deploy IPv6 in 30 minutes or less tutorial on Sunday and apparently it actually worked. Unfortunately I neglected to mention the important acquire connectivity to the global routing table step (I assumed it was implied, but I guess it wasn't), but if you're down with their 5 v6 routes as transit you should be golden. :) -- Richard A Steenbergen r...@e-gerbil.net http://www.e-gerbil.net/ras GPG Key ID: 0xF8B12CBC (7535 7F59 8204 ED1F CC1C 53AF 4C41 5ECA F8B1 2CBC) Nary a route... dtem...@jnrt-edge01.sv1 show route table inet6.0 aspath-regex .* 174 .* inet6.0: 1914 destinations, 5528 routes (1914 active, 0 holddown, 0 hidden) dtem...@jnrt-edge01.sv1 -Dave
Re: Cogent input
Joel Jaeggli wrote: Steve Bertrand wrote: Stephen Kratzer wrote: And, they have no plans to support IPv6. Ouch! I hope this is a non-starter for a lot of folks. read the rest of the thread... ...unfortunately, my message was sent out on the 11th, but just received yesterday by the list. I had both nanog@nanog.org and na...@merit.edu Cc'd, so I don't know which one failed. I have problems sending to the NANOG list from my IPv6 server, so I'll have to find out which destination list address is breaking for me... Received: from unknown (HELO ?IPv6:2607:f118::5?) (st...@ibctech.ca@2607:f118::5) by 2607:f118::b6 with ESMTPA; 11 Jun 2009 14:39:10 - Received: from v6.ibctech.ca ([2607:f118::b6] helo=ibctech.ca) by s0.nanog.org with smtp (Exim 4.68 (FreeBSD))(envelope-from st...@ibctech.ca) id 1MGhgj-qP-Cd for nanog@nanog.org; Tue, 16 Jun 2009 23:03:33 + Well, I'm glad I had nothing better to do today other than figure out what is wrong with that particular email server ;) Steve smime.p7s Description: S/MIME Cryptographic Signature
Cogent input - no peering with Global Crossing in Europe [Re: NANOG Digest, Vol 17, Issue 46]
My main concern for European Cogent users is - no European peering with global crossing - traffic goes via NY JFK. It has been like this for at least a year and staff been giving assurances this should be sorted soon. Probably there are more bad peerings - please share. 6: so-7-0-0c0.rt1.mil.it.geant2.net (62.40.112.174) asymm 5 14.446ms 7: so-7-1-0.rt1.fra.de.geant2.net (62.40.112.61)asymm 6 27.120ms 8: TenGigabitEthernet7-3.ar1.FRA4.gblx.net (207.138.144.45) 246.852ms 9: po3-20G.ar7.NYC1.gblx.net (67.16.134.74) asymm 12 122.810ms 10: te2-1.ccr01.jfk07.atlas.cogentco.com (154.54.11.61) asymm 12 123.003ms 11: te4-1.ccr01.jfk02.atlas.cogentco.com (154.54.1.221) asymm 13 118.334ms 12: te4-3.ccr01.lon01.atlas.cogentco.com (130.117.1.106) asymm 14 198.997ms 13: te2-7.ccr02.ams03.atlas.cogentco.com (130.117.1.169) asymm 14 204.575ms 14: te2-3.ccr01.dus01.atlas.cogentco.com (130.117.3.90) asymm 15 213.653ms 15: te7-1.ccr01.muc01.atlas.cogentco.com (130.117.49.154) asymm 14 225.144ms 16: te3-1.ccr01.vie01.atlas.cogentco.com (130.117.49.30) asymm 22 254.543ms 17: te3-8.ccr01.bts01.atlas.cogentco.com (130.117.49.25) asymm 21 248.505ms 18: te1-1.ccr01.tsr01.atlas.cogentco.com (130.117.48.58) asymm 20 243.334ms 19: te1-3.ccr01.tsr01.atlas.cogentco.com (130.117.0.18) asymm 20 249.374ms 20: 149.6.112.2 (149.6.112.2)asymm 19 273.730ms 21: 149.6.112.2 (149.6.112.2)asymm 19 268.122ms 22: down-int.caucasus.net (62.168.172.205) asymm 21 268.647ms 23: down-int.caucasus.net (62.168.172.205) asymm 21 274.430ms 24: sw.caucasus.net (62.168.168.60) asymm 21 277.811ms nanog-requ...@nanog.org wrote: Send NANOG mailing list submissions to nanog@nanog.org
Re: Cogent input - no peering with Global Crossing in Europe [Re: NANOG Digest, Vol 17, Issue 46]
Ouch... latency must be awful. I suppose this is based on Cogents reputation but who knows. The whole peering aspect of the networking business is often a mystery. AKK wrote: My main concern for European Cogent users is - no European peering with global crossing - traffic goes via NY JFK. It has been like this for at least a year and staff been giving assurances this should be sorted soon. Probably there are more bad peerings - please share. 6: so-7-0-0c0.rt1.mil.it.geant2.net (62.40.112.174) asymm 5 14.446ms 7: so-7-1-0.rt1.fra.de.geant2.net (62.40.112.61)asymm 6 27.120ms 8: TenGigabitEthernet7-3.ar1.FRA4.gblx.net (207.138.144.45) 246.852ms 9: po3-20G.ar7.NYC1.gblx.net (67.16.134.74) asymm 12 122.810ms 10: te2-1.ccr01.jfk07.atlas.cogentco.com (154.54.11.61) asymm 12 123.003ms 11: te4-1.ccr01.jfk02.atlas.cogentco.com (154.54.1.221) asymm 13 118.334ms 12: te4-3.ccr01.lon01.atlas.cogentco.com (130.117.1.106) asymm 14 198.997ms 13: te2-7.ccr02.ams03.atlas.cogentco.com (130.117.1.169) asymm 14 204.575ms 14: te2-3.ccr01.dus01.atlas.cogentco.com (130.117.3.90) asymm 15 213.653ms 15: te7-1.ccr01.muc01.atlas.cogentco.com (130.117.49.154) asymm 14 225.144ms 16: te3-1.ccr01.vie01.atlas.cogentco.com (130.117.49.30) asymm 22 254.543ms 17: te3-8.ccr01.bts01.atlas.cogentco.com (130.117.49.25) asymm 21 248.505ms 18: te1-1.ccr01.tsr01.atlas.cogentco.com (130.117.48.58) asymm 20 243.334ms 19: te1-3.ccr01.tsr01.atlas.cogentco.com (130.117.0.18) asymm 20 249.374ms 20: 149.6.112.2 (149.6.112.2)asymm 19 273.730ms 21: 149.6.112.2 (149.6.112.2)asymm 19 268.122ms 22: down-int.caucasus.net (62.168.172.205) asymm 21 268.647ms 23: down-int.caucasus.net (62.168.172.205) asymm 21 274.430ms 24: sw.caucasus.net (62.168.168.60) asymm 21 277.811ms nanog-requ...@nanog.org wrote: Send NANOG mailing list submissions to nanog@nanog.org
Re: Cogent input
Hi Justin, Just FYI - Global Crossing can currently deliver dual stack/native v6 transit in downtown KC,MO. You can either colo with them at 1100 Main St, or possibly have them haul a wave to one of the other major downtown carrier hotels they have strands running through / into (1102 Grand/Bryant and 324 E. 11th St/Oak Towers come to mind, not to mention Level3's suite in 1100 Walnut right across the street). Cheers, Kevin Hodle On Fri, Jun 12, 2009 at 8:13 AM, Justin Shore jus...@justinshore.comwrote: John van Oppen wrote: NTT (2914) and GBLX (3549) both do native v6... most everyone else on the tier1 list does tunnels. :( There are some nice tier2 networks who do native v6, tiscali and he.net come to mind. Let me rephrase that. :-) I know of no tier-Ns that offer any native v6 services here in the Midwest (central Kansas) including L3 which only has a best effort pilot program using tunnels. There might be more options in KC or OKC but not here that I'm aware of... Justin -- || Kevin Hodle || || 913-780-3959 (Primary) || 913-626-7197 (Mobile) PGP KeyID [0xBBDE8ED7] fingerprint [3E1B 1F10 938E A831 8CF2 670C 1329 0B8B BBDE 8ED7]
Re: Cogent input
Steve Bertrand wrote: Stephen Kratzer wrote: And, they have no plans to support IPv6. Ouch! I hope this is a non-starter for a lot of folks. read the rest of the thread... joel Steve
Re: Cogent input
Justin Shore wrote: Paul Timmins wrote: GlobalCrossing told me today I can order native IPv6 anywhere on their network. Don't know if they count as Tier 1 on your list, though. VZB has given me tunnels for a while, hopefully they'll get their pMTU issue fixed so we can do more interesting things with it. I'd love to have GLBX but I'm positive that they aren't available here. We'd have to pay for transport to a much larger market to go get them. That may be more feasible when the state network gets built but that's a few years off. Until then I'll have to dream about GBLX... For me their local POP is apparently at capacity, so I had to take them off the list since I couldn't afford the transport costs to California. I heard somewhere that it's MPLS internally in order to present dual-stack at your port, not dual-stack on every router, thus most of the path is invisible like a tunnel. ~Seth
Re: Cogent input
On 6/11/09 7:37 AM, Steve Bertrand st...@ibctech.ca wrote: Stephen Kratzer wrote: And, they have no plans to support IPv6. Ouch! I hope this is a non-starter for a lot of folks. Steve To quote Randy, I encourage all my competitors to do this. Mike
Re: Cogent input
Justin Shore wrote: I'm in search of some information about Cogent, it's past, present and future. I've heard bits and pieces about Cogent's past over the years but by no means have I actively been keeping up. We've used cogent for the past year, 100 over GigE. - Clueful and responsive tech support. - Active monitoring of faults. - Helpful people at billing (even when trying to give them less money). We were affected by one depeering during the last year (Telia). We're multihomed and that mitigated the problem. Obviously, you'd never want to go exclusively with them. Cheers, Stef
Re: Cogent input
It's worth noting that being a v4 tier1/transit-free network doesn't necessarily mean that they're the same in the v6 world. For instance, Google appears to be a transit-free v6 network. It wouldn't surprise me if the same is true for other big v6 players like Tinet and HE. Good point. Anyway, when looking for v6 transit the following page might be useful: http://www.sixxs.net/faq/connectivity/?faq=ipv6transit Yup, but take it with a grain of salt. Level3, for instance, definitely doesn't offer native IPv6 with the same geographical distribution as IPv4. Steinar Haug, Nethelp consulting, sth...@nethelp.no
Re: Cogent input
That might be because some bigger providers in the Netherlands are throwing out transits that don't support IPv6. So there's your commercial necessity ;-) Raymond Dijkxhoorn wrote: Hi! Should have said And, they have no plans to deploy IPv6 in the immediate future. :) Cogent's official stance on IPv6 is that we will deploy IPv6 when it becomes a commercial necessity. We have tested IPv6 and we have our plan for rolling it out, but there are no commercial drivers to spend money to upgrade a network to IPv6 for no real return on investment. Thats strange they are running pilots with customers on v6 in Amsterdam. Bye, Raymond. -- Met vriendelijke groet, Jeroen Wunnink, EasyHosting B.V. Systeembeheerder systeembeh...@easyhosting.nl telefoon:+31 (035) 6285455 Postbus 48 fax: +31 (035) 6838242 3755 ZG Eemnes http://www.easyhosting.nl http://www.easycolocate.nl
Re: Cogent input
John van Oppen wrote: NTT (2914) and GBLX (3549) both do native v6... most everyone else on the tier1 list does tunnels. :( There are some nice tier2 networks who do native v6, tiscali and he.net come to mind. Let me rephrase that. :-) I know of no tier-Ns that offer any native v6 services here in the Midwest (central Kansas) including L3 which only has a best effort pilot program using tunnels. There might be more options in KC or OKC but not here that I'm aware of... Justin
Re: Cogent input
Paul Timmins wrote: GlobalCrossing told me today I can order native IPv6 anywhere on their network. Don't know if they count as Tier 1 on your list, though. VZB has given me tunnels for a while, hopefully they'll get their pMTU issue fixed so we can do more interesting things with it. I'd love to have GLBX but I'm positive that they aren't available here. We'd have to pay for transport to a much larger market to go get them. That may be more feasible when the state network gets built but that's a few years off. Until then I'll have to dream about GBLX... Justin
Re: Cogent input
In a message written on Fri, Jun 12, 2009 at 08:13:02AM -0500, Justin Shore wrote: Let me rephrase that. :-) I know of no tier-Ns that offer any native v6 services here in the Midwest (central Kansas) including L3 which only has a best effort pilot program using tunnels. There might be more options in KC or OKC but not here that I'm aware of... www.tunnelbroker.net (which is HE under the hood) is available anywhere, self serve, and works great. I bring this up knowing a tunnel is not as good as native connectivity, yadda yadda yadda. My personal experience with sales folks is telling them you need IPv6 makes them go ask questions internally, while telling them you were forced to get service (a tunnel) from a competitor generally sends the entire sales force into a tizzy internally. It really drives home the point that not only are you asking, but you need it, and will go around them if necessary. This changes the equasion for them from wanting to be second (we will offer it in the market when our competitor does) to wanting to be first (we can't let our competitor pick off customers like this who are desperate for the feature). Of course, you may not want to turn up customers on such a service, but it does provide an excellent opportunity to enable your test lab, employee personal boxes, and the like so you can hit the ground running. Hard to beat for something that's free and takes 15 minutes to set up. -- Leo Bicknell - bickn...@ufp.org - CCIE 3440 PGP keys at http://www.ufp.org/~bicknell/ pgphwsGecxnLQ.pgp Description: PGP signature
Re: Cogent input
On Thu, Jun 11, 2009 at 9:46 AM, Justin Shore jus...@justinshore.comwrote: I'm in search of some information about Cogent, it's past, present and future. I've heard bits and pieces about Cogent's past over the years but by no means have I actively been keeping up. I had a very positive interaction with the Cogent folks in Europe FWIW. It was seeded by their US colleagues and everyone worked closely to try and make things work operationally and economically. That signaled to me that they have some cohesiveness internally which I found to be a big plus*. At worst, allowing Cogent to seriously compete for your business can help you reduce your costs elsewhere if you are smart about it. At best, you could cut your costs and maintain the reliability that you are seeking with some strategic mitigation. Others may have different suggestions. YMMV. Best, Marty * Not an endorsement, just an opinion. -- Martin Hannigan mar...@theicelandguy.com p: +16178216079 Power, Network, and Costs Consulting for Iceland Datacenters and Occupants
RE: Cogent input
Our experience with them was at least one major (longer than an hour) outages PER MONTH and many of those times they were black holing our routes in their network which was the most damaging aspect. The outages were one thing but when our routes still somehow managed to get advertised in their network (even though our BGP session was down) that really created issues. I have heard from some nearby folks who still have service that it's gotten better, but we are also in the regional offering when it comes to IP Transit and have sold connections to many former Cogent customers who were fed up and left. I have found with Cogent that you will get a LOT of varying opinions on them - there are several other players (at least in our market) that are priced very similar now and have a better history behind them. The specific de-peering issues never effected us much due to enough diversity in our upstreams and a fair amount of direct/public peering... Thanks, Paul -Original Message- From: Justin Shore [mailto:jus...@justinshore.com] Sent: Thursday, June 11, 2009 9:47 AM To: NANOG Subject: Cogent input I'm in search of some information about Cogent, it's past, present and future. I've heard bits and pieces about Cogent's past over the years but by no means have I actively been keeping up. I'm aware of some (regular?) depeering issues. The NANOG archives have given me some additional insight into that (recurring?) problem. The reasoning behind the depeering events is a bit fuzzy though. I would be interested in people's opinion on whether or not they should be consider for upstream service based on this particular issue. Are there any reasonable mitigation measures available to Cogent downstreams if (when?) Cogent were to be depeered again? My understanding is that at least on previous depeering occasion, the depeering partner simply null-routed all prefixes being received via Cogent, creating a blackhole essentially. I also recall reading that this meant that prefixes being advertised and received by the depeering partner from other peers would still end up in the blackhole. The only solution I would see to this problem would be to shut down the BGP session with Cogent and rely on a 2nd upstream. Are there any other possible steps for mitigation in a depeering event? I also know that their bandwidth is extremely cheap. This of course creates an issue for technical folks when trying to justify other upstream options that cost significantly more but also don't have a damaging history of getting depeered. Does Cogent still have an issue with depeering? Are there any reasonable mitigation measures or should a downstream customer do any thing in particular to ready themselves for a depeering event? Does their low cost outweigh the risks? What are the specific risks? Thanks Justin The information transmitted is intended only for the person or entity to which it is addressed and contains confidential and/or privileged material. If you received this in error, please contact the sender immediately and then destroy this transmission, including all attachments, without copying, distributing or disclosing same. Thank you.
Re: Cogent input
At $JOB-1 we used Cogent. Lots of horror stories had been heard about them. We didn't have such problems. Had nx1Gig from them. On the few occasions where we had some slight issues, I was happy to be able to get through to some one useful on the phone quickly, and not play pass the parcel with call centre operatives. and at least in the quantities we were buying they were significantly better value than others, which was the primary reason we went with them. andrew On Thu, Jun 11, 2009 at 2:55 PM, Paul Stewartpstew...@nexicomgroup.net wrote: Our experience with them was at least one major (longer than an hour) outages PER MONTH and many of those times they were black holing our routes in their network which was the most damaging aspect. The outages were one thing but when our routes still somehow managed to get advertised in their network (even though our BGP session was down) that really created issues. I have heard from some nearby folks who still have service that it's gotten better, but we are also in the regional offering when it comes to IP Transit and have sold connections to many former Cogent customers who were fed up and left. I have found with Cogent that you will get a LOT of varying opinions on them - there are several other players (at least in our market) that are priced very similar now and have a better history behind them. The specific de-peering issues never effected us much due to enough diversity in our upstreams and a fair amount of direct/public peering... Thanks, Paul -Original Message- From: Justin Shore [mailto:jus...@justinshore.com] Sent: Thursday, June 11, 2009 9:47 AM To: NANOG Subject: Cogent input I'm in search of some information about Cogent, it's past, present and future. I've heard bits and pieces about Cogent's past over the years but by no means have I actively been keeping up. I'm aware of some (regular?) depeering issues. The NANOG archives have given me some additional insight into that (recurring?) problem. The reasoning behind the depeering events is a bit fuzzy though. I would be interested in people's opinion on whether or not they should be consider for upstream service based on this particular issue. Are there any reasonable mitigation measures available to Cogent downstreams if (when?) Cogent were to be depeered again? My understanding is that at least on previous depeering occasion, the depeering partner simply null-routed all prefixes being received via Cogent, creating a blackhole essentially. I also recall reading that this meant that prefixes being advertised and received by the depeering partner from other peers would still end up in the blackhole. The only solution I would see to this problem would be to shut down the BGP session with Cogent and rely on a 2nd upstream. Are there any other possible steps for mitigation in a depeering event? I also know that their bandwidth is extremely cheap. This of course creates an issue for technical folks when trying to justify other upstream options that cost significantly more but also don't have a damaging history of getting depeered. Does Cogent still have an issue with depeering? Are there any reasonable mitigation measures or should a downstream customer do any thing in particular to ready themselves for a depeering event? Does their low cost outweigh the risks? What are the specific risks? Thanks Justin The information transmitted is intended only for the person or entity to which it is addressed and contains confidential and/or privileged material. If you received this in error, please contact the sender immediately and then destroy this transmission, including all attachments, without copying, distributing or disclosing same. Thank you.
Re: Cogent input
At 10:01 AM 6/11/2009, Andrew Mulholland wrote: We didn't have such problems. Had nx1Gig from them. On the few occasions where we had some slight issues, I was happy to be able to get through to some one useful on the phone quickly, and not play pass the parcel with call centre operatives. This matches our experience as well. When there are issues, they are EASY to get a hold of and the people who answer the phone clueful and dedicated to dealing with IP issues, not can I help you with your long distance bill Also, they are pretty good about keeping us informed about maintenance issues. I would not use them as a sole provider (why run your own AS if you only have one transit provider?) but certainly I am happy keeping them in the mix to date. We havent seen the same level of issues as some people in YYZ have seen, but I think that seems to be more on their 100Mb connections for some reason. On the gig service we are on, they are fairly reliable. Not quite as good as TATA/Teleglobe has been for us however. ---Mike
Re: Cogent input
On Jun 11, 2009, at 10:13 AM, Mike Tancsa wrote: At 10:01 AM 6/11/2009, Andrew Mulholland wrote: We didn't have such problems. Had nx1Gig from them. On the few occasions where we had some slight issues, I was happy to be able to get through to some one useful on the phone quickly, and not play pass the parcel with call centre operatives. This matches our experience as well. When there are issues, they are EASY to get a hold of and the people who answer the phone clueful and dedicated to dealing with IP issues, not can I help you with your long distance bill Also, they are pretty good about keeping us informed about maintenance issues. I would not use them as a sole provider (why run your own AS if you only have one transit provider?) but certainly I am happy keeping them in the mix to date. +1 from here. Marshall We havent seen the same level of issues as some people in YYZ have seen, but I think that seems to be more on their 100Mb connections for some reason. On the gig service we are on, they are fairly reliable. Not quite as good as TATA/Teleglobe has been for us however. ---Mike
Re: Cogent input
Should have said And, they have no plans to deploy IPv6 in the immediate future. On Thursday 11 June 2009 10:33:25 Stephen Kratzer wrote: We've only recently started using Cogent transit, but it's been stable since its introduction 6 months ago. Turn-up was a bit rocky since we never received engineering details, and engineering was atypical in that two eBGP sessions were established, one just to advertise loopbacks, and another for the actual feed. The biggest issue we have with them is that they don't allow deaggregation. If you've been allocated a prefix of length yy, they'll accept only x.x.x.x/yy, not x.x.x.x/yy le 24. Yes, sometimes deaggregation is necessary or desirable even if only temporarily. And, they have no plans to support IPv6. Cogent's official stance on IPv6 is that we will deploy IPv6 when it becomes a commercial necessity. We have tested IPv6 and we have our plan for rolling it out, but there are no commercial drivers to spend money to upgrade a network to IPv6 for no real return on investment. Stephen Kratzer Network Engineer CTI Networks, Inc. On Thursday 11 June 2009 09:46:45 Justin Shore wrote: I'm in search of some information about Cogent, it's past, present and future. I've heard bits and pieces about Cogent's past over the years but by no means have I actively been keeping up. I'm aware of some (regular?) depeering issues. The NANOG archives have given me some additional insight into that (recurring?) problem. The reasoning behind the depeering events is a bit fuzzy though. I would be interested in people's opinion on whether or not they should be consider for upstream service based on this particular issue. Are there any reasonable mitigation measures available to Cogent downstreams if (when?) Cogent were to be depeered again? My understanding is that at least on previous depeering occasion, the depeering partner simply null-routed all prefixes being received via Cogent, creating a blackhole essentially. I also recall reading that this meant that prefixes being advertised and received by the depeering partner from other peers would still end up in the blackhole. The only solution I would see to this problem would be to shut down the BGP session with Cogent and rely on a 2nd upstream. Are there any other possible steps for mitigation in a depeering event? I also know that their bandwidth is extremely cheap. This of course creates an issue for technical folks when trying to justify other upstream options that cost significantly more but also don't have a damaging history of getting depeered. Does Cogent still have an issue with depeering? Are there any reasonable mitigation measures or should a downstream customer do any thing in particular to ready themselves for a depeering event? Does their low cost outweigh the risks? What are the specific risks? Thanks Justin
Re: Cogent input
Stephen Kratzer wrote: And, they have no plans to support IPv6. Ouch! I hope this is a non-starter for a lot of folks. Steve smime.p7s Description: S/MIME Cryptographic Signature
Re: Cogent input
Hi Justin, I'm in search of some information about Cogent, it's past, present and future. I've heard bits and pieces about Cogent's past over the years but by no means have I actively been keeping up. We recently got a 10-gig port in Oslo from them. Price-wise they were competitive but absolutely not in a leauge of their own - a couple of other large providers matched their offers. In the end the main differencing factor for us was that their PoP happened to be in the same building as our data centre, so no local access was required (unlike the others). The link hasn't been up for very long so I can't comment on long-term reliability issues, but so far I've been _very_ happy with them, everything was up and running just a couple of days after we ordered, and the staff we've had contact with have been knowledgeable and helpful. The service has performed as expected: latency has been low and I haven't noticed any sub-optimal routing (trampolines) or packet loss. We multihome, so we're not too concerned about potential de-peerings. I would not single-home to any of the transit-free networks anyway, as any of them could end up on the receiving end of a de-peering. Best regards, -- Tore Anderson Redpill Linpro AS - http://www.redpill-linpro.com/ Tel: +47 21 54 41 27
Re: Cogent input
I'm skeptical as to where this info came from since this seems nothing more then nay-say? if people are going to make grandiose statements then they should justify them with reputable evidence. I would be extremely surprised if Cogent engineering isn't working on a IPv6 plan or doesn't have one already in place. Bret On Thu, 2009-06-11 at 10:37 -0400, Steve Bertrand wrote: Stephen Kratzer wrote: And, they have no plans to support IPv6. Ouch! I hope this is a non-starter for a lot of folks. Steve
Re: Cogent input
Stephen Kratzer wrote: Should have said And, they have no plans to deploy IPv6 in the immediate future. On Thursday 11 June 2009 10:33:25 Stephen Kratzer wrote: We've only recently started using Cogent transit, but it's been stable since its introduction 6 months ago. Turn-up was a bit rocky since we never received engineering details, and engineering was atypical in that two eBGP sessions were established, one just to advertise loopbacks, and another for the actual feed. The biggest issue we have with them is that they don't allow deaggregation. If you've been allocated a prefix of length yy, they'll accept only x.x.x.x/yy, not x.x.x.x/yy le 24. Yes, sometimes deaggregation is necessary or desirable even if only temporarily. And, they have no plans to support IPv6. Cogent's official stance on IPv6 is that we will deploy IPv6 when it becomes a commercial necessity. We have tested IPv6 and we have our plan for rolling it out, but there are no commercial drivers to spend money to upgrade a network to IPv6 for no real return on investment. Stephen Kratzer Network Engineer CTI Networks, Inc. On Thursday 11 June 2009 09:46:45 Justin Shore wrote: I'm in search of some information about Cogent, it's past, present and future. I've heard bits and pieces about Cogent's past over the years but by no means have I actively been keeping up. I'm aware of some (regular?) depeering issues. The NANOG archives have given me some additional insight into that (recurring?) problem. The reasoning behind the depeering events is a bit fuzzy though. I would be interested in people's opinion on whether or not they should be consider for upstream service based on this particular issue. Are there any reasonable mitigation measures available to Cogent downstreams if (when?) Cogent were to be depeered again? My understanding is that at least on previous depeering occasion, the depeering partner simply null-routed all prefixes being received via Cogent, creating a blackhole essentially. I also recall reading that this meant that prefixes being advertised and received by the depeering partner from other peers would still end up in the blackhole. The only solution I would see to this problem would be to shut down the BGP session with Cogent and rely on a 2nd upstream. Are there any other possible steps for mitigation in a depeering event? I also know that their bandwidth is extremely cheap. This of course creates an issue for technical folks when trying to justify other upstream options that cost significantly more but also don't have a damaging history of getting depeered. Does Cogent still have an issue with depeering? Are there any reasonable mitigation measures or should a downstream customer do any thing in particular to ready themselves for a depeering event? Does their low cost outweigh the risks? What are the specific risks? Thanks Justin In Europe they have been good and stable most of the time. In the US well, they are cogent and I have so many bad experiences with them here I cannot in all honestly recommend them. But if your looking for cheap bandwidth to complement another provider its not an unreasonable thing to do as they price point is competitive. Manolo
Re: Cogent input
We've been using Cogent for 4 months now and I have no major complaints.
Re: Cogent input
Hello, * Stephen Kratzer We've only recently started using Cogent transit, but it's been stable since its introduction 6 months ago. Turn-up was a bit rocky since we never received engineering details, and engineering was atypical in that two eBGP sessions were established, one just to advertise loopbacks, and another for the actual feed. The biggest issue we have with them is that they don't allow deaggregation. If you've been allocated a prefix of length yy, they'll accept only x.x.x.x/yy, not x.x.x.x/yy le 24. Yes, sometimes deaggregation is necessary or desirable even if only temporarily. Interesting. I requested exactly that when filling in their BGP questionnaire, and they set it up - no questions asked. Also, we have a perfectly normal single BGP session. The loopback address of the router we're connected to is found within the 38.0.0.0/8 prefix, which they announce to us over that session like any other route. And, they have no plans to support IPv6. I have been promised, in writing, that they will provide us with native IPv6 transit before the end of the year. I'm based in Europe, though. Perhaps they're more flexible and customer-friendly here than in the US? Best regards, -- Tore Anderson Redpill Linpro AS - http://www.redpill-linpro.com/ Tel: +47 21 54 41 27
Re: Cogent input
On Jun 11, 2009, at 9:33 AM, Stephen Kratzer wrote: The biggest issue we have with them is that they don't allow deaggregation. If you've been allocated a prefix of length yy, they'll accept only x.x.x.x/yy, not x.x.x.x/yy le 24. Yes, sometimes deaggregation is necessary or desirable even if only temporarily. In our peering session with Cogent, we requested several /16 le /24 prefixes configured and received no pushback or problems. We are using their 1Gbps service so I'm not sure if that buys us additional leeway with requests like this or not. We've been customers for nearly two full years and intend to continue for at least a third. -brad fleming
Re: Cogent input
Perhaps you missed my quote: Cogent's official stance on IPv6 is that we will deploy IPv6 when it becomes a commercial necessity. We have tested IPv6 and we have our plan for rolling it out, but there are no commercial drivers to spend money to upgrade a network to IPv6 for no real return on investment. This came rom a contact at Cogent (not sure of the role, probably sales rep). On Thursday 11 June 2009 10:49:13 Bret Clark wrote: I'm skeptical as to where this info came from since this seems nothing more then nay-say? if people are going to make grandiose statements then they should justify them with reputable evidence. I would be extremely surprised if Cogent engineering isn't working on a IPv6 plan or doesn't have one already in place. Bret On Thu, 2009-06-11 at 10:37 -0400, Steve Bertrand wrote: Stephen Kratzer wrote: And, they have no plans to support IPv6. Ouch! I hope this is a non-starter for a lot of folks. Steve
Re: Cogent input
Perhaps you missed my amendment: Should have said And, they have no plans to deploy IPv6 in the immediate future. :) On Thursday 11 June 2009 11:06:38 Bret Clark wrote: Far different response then whoever quoted...And, they have no plans to support IPv6. On Thu, 2009-06-11 at 11:03 -0400, Stephen Kratzer wrote: Perhaps you missed my quote: Cogent's official stance on IPv6 is that we will deploy IPv6 when it becomes a commercial necessity. We have tested IPv6 and we have our plan for rolling it out, but there are no commercial drivers to spend money to upgrade a network to IPv6 for no real return on investment. This came rom a contact at Cogent (not sure of the role, probably sales rep). On Thursday 11 June 2009 10:49:13 Bret Clark wrote: I'm skeptical as to where this info came from since this seems nothing more then nay-say? if people are going to make grandiose statements then they should justify them with reputable evidence. I would be extremely surprised if Cogent engineering isn't working on a IPv6 plan or doesn't have one already in place. Bret On Thu, 2009-06-11 at 10:37 -0400, Steve Bertrand wrote: Stephen Kratzer wrote: And, they have no plans to support IPv6. Ouch! I hope this is a non-starter for a lot of folks. Steve
Re: Cogent input
Hi! Cogent's official stance on IPv6 is that we will deploy IPv6 when it becomes a commercial necessity. We have tested IPv6 and we have our plan for rolling it out, but there are no commercial drivers to spend money to upgrade a network to IPv6 for no real return on investment. Thats strange they are running pilots with customers on v6 in Amsterdam. Not really so strange. ISPs often pilot features (v6, multicast, etc) that they have no immediate intention of deploying, so that they have experience if and when it becomes profitable/sensible to deploy them. Not from what i have been told, but hey i am not working there. We got a v6 transit offer as pilot from them so perhaps they are moving towards live service Would not be strange in this current stage... Bye, Raymond.
Re: Cogent input
On Jun 11, 2009, at 11:03 AM, Stephen Kratzer wrote: Perhaps you missed my quote: Cogent's official stance on IPv6 is that we will deploy IPv6 when it becomes a commercial necessity. We have tested IPv6 and we have our plan for rolling it out, but there are no commercial drivers to spend money to upgrade a network to IPv6 for no real return on investment. FWIW, they have said basically the same thing about multicast. Regards Marshall This came rom a contact at Cogent (not sure of the role, probably sales rep). On Thursday 11 June 2009 10:49:13 Bret Clark wrote: I'm skeptical as to where this info came from since this seems nothing more then nay-say? if people are going to make grandiose statements then they should justify them with reputable evidence. I would be extremely surprised if Cogent engineering isn't working on a IPv6 plan or doesn't have one already in place. Bret On Thu, 2009-06-11 at 10:37 -0400, Steve Bertrand wrote: Stephen Kratzer wrote: And, they have no plans to support IPv6. Ouch! I hope this is a non-starter for a lot of folks. Steve
Re: Cogent input
Here as well. We're a small content provider, and we have cogent as one of our ISPs. Though I wouldn't feel comfortable using only them, my experience has been pretty good. Their NOC is competent, and service has been reliable. seph Bret Clark bcl...@spectraaccess.com writes: I hate when these questions get asked, because as the saying goes...a person happy with a service will only tell one other person, but a person unhappy with a service with tell ten other people. So I think a lot of times you'll get skewed responses...but with that said, we've been using Cogent now for a year and no complaints at all. Had some minor downtime back in April due to a hardware failure, but Cogent responded extremely quickly, scheduled an emergency maintainance and had us running rather quickly. Face it, hardware problems happen so I can't blame Cogent on the failure. The few times I've dealt with their tech support group I found 99% of them very knowledgeable and I know that when we initially turned on the link they went the extra mile to resolve some initial problems during the weekend time frame. My 2 cents and with any provider mileage will vary, Bret On Thu, 2009-06-11 at 15:01 +0100, Andrew Mulholland wrote: At $JOB-1 we used Cogent. Lots of horror stories had been heard about them. We didn't have such problems. Had nx1Gig from them. On the few occasions where we had some slight issues, I was happy to be able to get through to some one useful on the phone quickly, and not play pass the parcel with call centre operatives. and at least in the quantities we were buying they were significantly better value than others, which was the primary reason we went with them. andrew On Thu, Jun 11, 2009 at 2:55 PM, Paul Stewartpstew...@nexicomgroup.net wrote: Our experience with them was at least one major (longer than an hour) outages PER MONTH and many of those times they were black holing our routes in their network which was the most damaging aspect. The outages were one thing but when our routes still somehow managed to get advertised in their network (even though our BGP session was down) that really created issues. I have heard from some nearby folks who still have service that it's gotten better, but we are also in the regional offering when it comes to IP Transit and have sold connections to many former Cogent customers who were fed up and left. I have found with Cogent that you will get a LOT of varying opinions on them - there are several other players (at least in our market) that are priced very similar now and have a better history behind them. The specific de-peering issues never effected us much due to enough diversity in our upstreams and a fair amount of direct/public peering... Thanks, Paul -Original Message- From: Justin Shore [mailto:jus...@justinshore.com] Sent: Thursday, June 11, 2009 9:47 AM To: NANOG Subject: Cogent input I'm in search of some information about Cogent, it's past, present and future. I've heard bits and pieces about Cogent's past over the years but by no means have I actively been keeping up. I'm aware of some (regular?) depeering issues. The NANOG archives have given me some additional insight into that (recurring?) problem. The reasoning behind the depeering events is a bit fuzzy though. I would be interested in people's opinion on whether or not they should be consider for upstream service based on this particular issue. Are there any reasonable mitigation measures available to Cogent downstreams if (when?) Cogent were to be depeered again? My understanding is that at least on previous depeering occasion, the depeering partner simply null-routed all prefixes being received via Cogent, creating a blackhole essentially. I also recall reading that this meant that prefixes being advertised and received by the depeering partner from other peers would still end up in the blackhole. The only solution I would see to this problem would be to shut down the BGP session with Cogent and rely on a 2nd upstream. Are there any other possible steps for mitigation in a depeering event? I also know that their bandwidth is extremely cheap. This of course creates an issue for technical folks when trying to justify other upstream options that cost significantly more but also don't have a damaging history of getting depeered. Does Cogent still have an issue with depeering? Are there any reasonable mitigation measures or should a downstream customer do any thing in particular to ready themselves for a depeering event? Does their low cost outweigh the risks? What are the specific risks? Thanks Justin The information transmitted is intended only for the person or entity to which it is addressed
RE: Cogent input
I'm aware of some (regular?) depeering issues. The NANOG archives have AFAIR, there has never been a black-holing, just disappearance of routes. If you are properly multihomed, this is irrelevant and you continue to eat your ice cream and chuckle while they fight it out. It's amusing, really. I also know that their bandwidth is extremely cheap. This of course creates an issue for technical folks when trying to justify other upstream options that cost significantly more but also don't have a damaging history of getting depeered. It's not as relatively cheap as it used to be. A variety of others can be had in the same ballpark (Telia, Tiscali, Seabone, etc.)
Re: Cogent input
On Jun 11, 2009, at 5:28 PM, Raymond Dijkxhoorn wrote: Not from what i have been told, but hey i am not working there. We got a v6 transit offer as pilot from them so perhaps they are moving towards live service Would not be strange in this current stage... same thing here. routing table still looks pretty empty though: dan...@jun1.bit-2a show route aspath-regex .* 174 .* table inet6.0 | match BGP | count Count: 0 lines --Daniel.
Re: Cogent input
Tore Anderson wrote: advertise loopbacks, and another for the actual feed. The biggest issue we have with them is that they don't allow deaggregation. If you've been allocated a prefix of length yy, they'll accept only x.x.x.x/yy, not x.x.x.x/yy le 24. Yes, sometimes deaggregation is necessary or desirable even if only temporarily. Interesting. I requested exactly that when filling in their BGP questionnaire, and they set it up - no questions asked. It would be a show-stopper for us if they didn't let us deaggregate. We're not really wanting their service for our existing service area. We're wanting to use it to expand to a new service area that is only connected by a much lower-speed service back to the bulk of our current network for specific services like voice. Our PI space is currently broken up to 1) let us effect some measure of load-balancing with Cox (any prefixes we advertised out Cox instead of our much larger tier-1 resulted in a wildly disproportionate amount of preference given to Cox; not sure why) and 2) let this new venture get started with a reasonably-sized allotment of IP. It will be advertised out local providers in that area and also at our main peering point with significant prepending. Visa versa for our other prefixes. We have to deaggregate a little bit to make this work (but not excessively of course). I have been promised, in writing, that they will provide us with native IPv6 transit before the end of the year. I hope at least some SPs make this commitment back in the states. I can't find any tier-1s that can provide us with native v6. Our tier-1 upstream has a best effort test program in place that uses ipv6ip tunnels. The other upstream says that they aren't making any public IPv6 plans yet. It's hard to push the migration to v6 along when native v6 providers aren't readily available. Justin
Re: Cogent input
I hope at least some SPs make this commitment back in the states. I can't find any tier-1s that can provide us with native v6. Our tier-1 upstream has a best effort test program in place that uses ipv6ip tunnels. The other upstream says that they aren't making any public IPv6 plans yet. It's hard to push the migration to v6 along when native v6 providers aren't readily available. GlobalCrossing told me today I can order native IPv6 anywhere on their network. Don't know if they count as Tier 1 on your list, though. VZB has given me tunnels for a while, hopefully they'll get their pMTU issue fixed so we can do more interesting things with it. -Paul
RE: Cogent input
NTT (2914) and GBLX (3549) both do native v6... most everyone else on the tier1 list does tunnels. :( There are some nice tier2 networks who do native v6, tiscali and he.net come to mind. -John -Original Message- From: Paul Timmins [mailto:p...@telcodata.us] Sent: Thursday, June 11, 2009 4:00 PM To: Justin Shore Cc: NANOG Subject: Re: Cogent input I hope at least some SPs make this commitment back in the states. I can't find any tier-1s that can provide us with native v6. Our tier-1 upstream has a best effort test program in place that uses ipv6ip tunnels. The other upstream says that they aren't making any public IPv6 plans yet. It's hard to push the migration to v6 along when native v6 providers aren't readily available. GlobalCrossing told me today I can order native IPv6 anywhere on their network. Don't know if they count as Tier 1 on your list, though. VZB has given me tunnels for a while, hopefully they'll get their pMTU issue fixed so we can do more interesting things with it. -Paul
Re: Cogent input
Does GBLX still have their data center in Chinatown(NYC)??? I remember about 10 years ago how amazed I was with that place... - Original Message - From: John van Oppen j...@vanoppen.com To: Paul Timmins p...@telcodata.us; Justin Shore jus...@justinshore.com Cc: NANOG na...@merit.edu Sent: Thu Jun 11 15:31:24 2009 Subject: RE: Cogent input NTT (2914) and GBLX (3549) both do native v6... most everyone else on the tier1 list does tunnels. :( There are some nice tier2 networks who do native v6, tiscali and he.net come to mind. -John -Original Message- From: Paul Timmins [mailto:p...@telcodata.us] Sent: Thursday, June 11, 2009 4:00 PM To: Justin Shore Cc: NANOG Subject: Re: Cogent input I hope at least some SPs make this commitment back in the states. I can't find any tier-1s that can provide us with native v6. Our tier-1 upstream has a best effort test program in place that uses ipv6ip tunnels. The other upstream says that they aren't making any public IPv6 plans yet. It's hard to push the migration to v6 along when native v6 providers aren't readily available. GlobalCrossing told me today I can order native IPv6 anywhere on their network. Don't know if they count as Tier 1 on your list, though. VZB has given me tunnels for a while, hopefully they'll get their pMTU issue fixed so we can do more interesting things with it. -Paul
Re: Cogent input
Good morning, * John van Oppen NTT (2914) and GBLX (3549) both do native v6... most everyone else on the tier1 list does tunnels. :( There are some nice tier2 networks who do native v6, tiscali and he.net come to mind. It's worth noting that being a v4 tier1/transit-free network doesn't necessarily mean that they're the same in the v6 world. For instance, Google appears to be a transit-free v6 network. It wouldn't surprise me if the same is true for other big v6 players like Tinet and HE. Anyway, when looking for v6 transit the following page might be useful: http://www.sixxs.net/faq/connectivity/?faq=ipv6transit Best regards, -- Tore Anderson Redpill Linpro AS - http://www.redpill-linpro.com/ Tel: +47 21 54 41 27
