[01:54] *** HighJinx has quit IRC (Quit: Leaving) [01:55] *** HighJinx has joined #arpnetworks [02:01] *** ivan-kan` has quit IRC (Remote host closed the connection) [02:03] *** ivan-kanis has joined #arpnetworks [02:09] *** mig5 has joined #arpnetworks [02:58] *** mig5 has left [03:05] does arp have a looking glass? [03:34] Wat? [03:34] a mirror? :) [04:21] *** heidar has quit IRC (Read error: Connection reset by peer) [04:22] *** heidar has joined #arpnetworks [04:24] dxtr: you know, a queriable BGP server type thing so you can see how the internet appears from the prespective of different places on the internet [05:04] why would arp need one? [05:24] RandalSchwartz: I guess it's just a 'nice to help' but personally I don't see the point in every provider doing it [05:36] Yeah - how many different views do you need? :) [05:36] you can also just traceroute yourself. I do that a lot. [05:36] I'm a new customer, and would have found one useful to see how a VPS at arp would route traffic to various places before I signed up [05:36] since it varies from day to day, how does that help? [05:37] RandalSchwartz: there are useful aspects though, but as I said, I don't see the point in every provider doing it [05:37] BGP shouldn't really be that unstable [05:37] if you make purchase decisions based on a particuler view one afternoon, I think you're missing the picture. :0 [05:37] plett - arp has a lot of redundant links [05:37] so yes - it does vary, with load, with what's up vs down, etc. [05:38] RandalSchwartz: Looking Glasses seem to only be of any use if stuff is broken [05:38] and if stuff is broken, up_the_irons should be fixing it. not waiting for us to diagnose it remotely :) [05:38] RandalSchwartz: I'm talking in a general sense [05:38] I generally do too. :) [05:39] and who's to say the brokenness is up_the_irons' fault [05:39] I didn't say that [05:39] RandalSchwartz: Indeed. I have seen the lists of peers on arp's website. It's the details of how it would be routed in the normal situation of all transits and peerings being up that I would have been interested in [05:39] plett - again, depending on load. :) [05:40] there is some sort of ASN mapper out there. [05:40] plett: why not traceroute from the other end? [05:40] ok - time to go get on a plane from Porto Alegre to Rio. [05:40] * RandalSchwartz waves off [05:40] G: I did. The routing is asymetric in some cases [05:42] RandalSchwartz: BGP doesn't include the load on the links when chosing a route for a prefix, unless arp are doing something non-standard [05:46] Routes to the UK (which is mostly what I'm interested in) improved quite a lot when arp added PacketExchange/mzima/GTT/whoever_they_are_today transit [05:47] A looking glass at ARP's end would let me see where it would go if the PacketExchange path were down, etc [06:24] plett: hah it's mzima routing that's borken for me [06:24] well i mean suboptimal [06:24] i was wondering how much it's preferred by and what my transit provider is doing [06:24] it used to be fine [06:25] oh actually it was only mzima foward path to the vps [06:25] reverse path is via trit.net [06:25] i dunno who trit.net is [06:26] mzima seems to be forward path going via paix. [06:27] it's easier to change forward-path routing than reverse-path routing though [06:27] https://www.trit.net/corp/ [06:27] wow [06:27] web site coming soon [06:31] mercutio: From here in the UK, it's mzima in both directions and a nice short route. It helps that we peer with PacketExchange/mzima in London, and ARP has transit from them in LAX [06:31] plett: what's short? [06:31] like what latency? [06:32] i have a uk vps [06:32] hmm it goes via mzima 150msec [06:33] I was meaning hop count rather than latency, but I get 160ms from my ADSL at home [06:33] that's not too bad [06:33] i suppose [06:33] i'm in new zealand [06:33] Ironically, IPv6 is consistently 20ms lower - 140ms rather than 160 [06:33] my latency was around 126 msec not including dsl segment [06:33] now it's like 167 not including dsl segment [06:33] dsl is around 10 msec [06:34] plett: that happens soemtimes [06:34] with one provider that i used to use it's 133 msec frmo uk [06:35] anyway the difference between 169 msec and 126 msec is actually noticable [06:35] IPv6 being lower latency is mostly just down to "everyone using HE" [06:35] but i realise that it's probably from my transit provider having fucked routing now [06:35] hmm [06:35] he.net is good [06:35] in osome ways [06:35] not very reliable [06:35] but some of their routing is pretty good [06:36] although on some paths it can vary [06:36] like going via denver/chi or phoenix/dallas/atlanta [06:36] We only get v6 from them in London as they peer for free [06:36] i usd to think he.net was just a budget lame provider [06:36] oh yip [06:36] i've got he.net tunnel [06:36] I have no knowledge of what their v4 routes are like [06:37] to the uk they're pretty good [06:37] hmm 155msec frmo san jose [06:37] that's not that good [06:37] oh that's right [06:37] they're better from la to uk [06:37] then san jose [06:38] san jose goes via ashville/paris [06:38] los angles gos via new york/london [06:39] i dropped the la provider that used he.net [06:39] due to excessive packet loss though [06:39] don't think it was he.net's doing though [06:39] 3. lonap.he.net 0.0% 8 10.2 2.8 1.3 10.2 3.1 [06:39] 4. 10gigabitethernet4-4.core1.nyc4.he.net 0.0% 8 70.6 72.4 69.3 79.4 3.9 [06:39] 5. 10gigabitethernet8-3.core1.chi1.he.net 0.0% 8 86.7 87.2 86.6 88.9 1.0 [06:39] 6. 10gigabitethernet3-2.core1.den1.he.net 0.0% 8 119.6 111.8 110.3 119.6 3.2 [06:39] 7. 10gigabitethernet1-4.core1.lax2.he.net 0.0% 8 132.7 134.2 132.5 142.0 3.3 [06:40] but athat's what their london to los angeles routing is like [06:41] who was it before mzima? [06:42] Originally it was all PacketExchange, then they created the mzima name for transit [06:43] nah i meant that arp was using [06:43] Oh, It used two or three different transit providers to them from here [06:44] I don't know about the other direction, as I didn't have anything at arp at that point [06:45] really packet loss makes more diff than latency [06:46] like 5% packet loss is worse than 40msec more latency [06:48] Looking at the BGP tables from the London end, every transit feed we have has picked packetexchange as its best route to arp [06:49] heh [06:49] And for v6, everyone (apart from Level3 of course) uses HE to get to arp [06:49] what do your transit tables say for best route to 202.49.71.59? [06:49] (my IP in new zealand) [06:49] it mostly goes via cogent from uk i think to me [06:50] Only if the UK isp is using cogent, I would assume [06:50] No short route to 202.49.71.59. Best route from my view in the UK is via Level3 - 3356 4648 4610 17746 9559 [06:50] i dunno it may be peered [06:51] hmm [06:51] i'm guessing cogent is one of those asnumbers [06:51] oh hey it's not [06:51] No, Cogent is 174 [06:51] do you get less than 300 msec ping? [06:52] That's Level3 -> Netgate -> Odyssey -> Orcon -> Plain Communications [06:52] yip [06:52] None of those names mean anything to me, they all look .nz to me :) [06:52] heh [06:52] 295ms [06:52] netgate is big provider here [06:52] odyssey is pretty much transit of orcon [06:52] orcon is upstream [06:52] plain communications is a small isp i work at [06:53] 295 msec yeh.. [06:53] I also work for a small ISP, but in the UK :) [06:53] that's where latency gets annoying :) [06:53] ahh ok [06:54] do you do many diff things ?/ [06:54] it's weird with big companies how people like really specialise [06:54] The usual. DSL, ethernet connections, voip, etc [06:55] Most of our income is DSL [06:55] nah i meant do you [06:55] not the isp [06:55] Ahh, yeah. Anything technical [06:55] ahh yip :) [06:55] that's basically my role heh [06:55] Most small companies are the same - everyone does everything [06:55] shit we don't make shit all off dsl. [06:56] none of the providers here make much off dsl [06:57] ethernet, voip etc have more profit [06:57] Ethernet has bigger profit margins, but you sell it to far fewer people [06:57] true [06:57] but like dsl [06:58] there is so small margins [06:58] i dunno, like here, you pay a lot to one company who owns dsl equipment [06:58] that connects to the users etc [06:58] so [06:58] that basically eats into things [06:59] and also bandwidth is expensive here [06:59] that said all the isp's here have data caps [06:59] Here it's British Telecom who owns the majority of the copper to peoples houses and DSL DSLAMs etc. We buy end user DSL tails from them [07:00] yeh, [07:00] so probably similar deal. [07:00] i have a friend in the uk [07:00] in cambridge [07:00] he was using dsl [07:00] his dsl was really shit [07:00] i think it had interleaving on [07:00] but like he got poor sync speeds. [07:00] then he moved, and he got slightly better syn cspeeds. [07:00] speeds [07:00] but it seems like exchanges are often far away there? [07:01] like he was at 30 or 40 db attenuation [07:01] something really high [07:01] but like here they're doing cabinets everywehre to give people faster sync rates [07:01] so most people sync at between around 8 to 20 mbit [07:01] Cambridge is a bit of an anomaly. It has lots of high tech business and a university etc, so has lots of high tech fibre connections. But apart from that, it's a small towm with copper that has been there for 60+ years [07:01] i sync around 2100/1260 [07:02] oh [07:02] so that's why his dsl sucks? [07:02] I know other people in cambridge, and they all say the same [07:02] he said that he wouldn't get virgin cable cos they wouldn't stop pestering him when he told them that the person that had cable connected when he moved in has left. [07:02] and were impossible to deal with [07:03] but i think virgin did 50mbit cable or something [07:03] err 2100 i meant 21500 [07:03] They're rolling out fibre-to-the-cabinet here too. I have it here at home, and get about 20Mb down and 10Mb up [07:03] vdsl? [07:03] Yeah. VDSL2 [07:03] there's vdsl2 equipment here [07:03] like i'll be on a vdsl2 port [07:03] but no-one is selling it yet [07:04] Maximum sync rate is 40M down and 10 up [07:04] oh and atm in t he trails you have to have interleaving on [07:04] so it's high ping [07:04] i used to live in a different city [07:04] and had like 20 msec higher ping [07:04] but like the diff between 10 and 30 msec ping [07:04] is actually quite noticable to me [07:04] Fibre-to-the-premises is on the roadmap here too. That will give 100M downstream [07:05] heh [07:05] thing is [07:05] web browsing doesn't go anywhere near 100 mbit [07:05] It's all the streaming video which is the killer [07:05] not here :) [07:06] yet. [07:06] well i suppose there's youtube [07:06] but like [07:06] that doesn't do anywhere near 100 mbit [07:06] We're a UK ISP, and Wimbledon tennis tournament is on this week - lunchtime data usage is sky high on our business customers [07:06] heh [07:07] is there a cdn for it? [07:07] The BBC stream it all live. They use Akamai and Highwinds as the CDN for it [07:07] which you'll have plenty of capacity to? [07:07] hmm, highwinds connectivity to nz isn't great [07:08] if its' the one i think it is [07:08] Oh yes. gig-e ports to Akamai. Highwinds refused to peer with us as we're too small [07:08] that's weird? [07:08] i thought cdn's would have open peering :) [07:08] The capacity problems come on the DSL backhaul links to BT [07:08] You'd think so, wouldn't you. Maybe that's why they're losing out to Akamai ;) [07:09] oh right [07:09] that's what was happening here [07:09] then they fixed it [07:09] well kind of [07:09] you have to change the way you're doing things [07:09] to connect [07:09] akamai is really slow i reckon [07:10] like with uncached content [07:10] well here it is at least [07:10] i honestly though they'd use more smarts [07:11] Peering is very big in the UK. Over half our traffic goes via peering rather than paid transit [07:12] wow! [07:12] about 10% of traffic is peered here? [07:12] i can't find a highwinds test download [07:13] We have huge peering LANs that have hundreds of members on them in the UK. I understand it's done differently in other parts of the world, with smaller per-building or per-provider public peering LANs, and lots of direct cables for private peering [07:13] anyway [07:13] hwcdn.net goes to LA [07:13] here like [07:13] the biggest providers don't peer [07:13] but the smaller ones do [07:13] but like [07:14] there's no google or anything here [07:14] so like there's not a lot of content peering [07:14] and like bittorrent etc [07:14] goes all over the place [07:14] Well, the big content providers or end-user eyeball networks peer here. But the big transit providers fairly obviously refuse to. [07:14] but usually not to peers [07:14] much [07:14] yeh [07:14] but like BT wouldn't peer right? [07:15] and virginmedia wouldn't? [07:15] and you were saying highwinds don't with small providers [07:15] Virgin don't, no [07:15] so it's interesting you can still get 50% peering traffic [07:15] i suppose youtube [07:15] accounts for a bit? [07:15] Yeah, we peer with google and facebook, which account for lots of traffic [07:16] cool [07:16] i suppose it is harder here [07:16] i've been trying to develop a system to improve performance over high latency connections atm [07:16] With .nz being thousands of miles from anywhere else? [07:16] err over the internet i mean [07:16] cos like international web browsing sucks from here [07:17] i've been like proxying with persistent connections [07:17] using geo ip lookups [07:17] to send to close proxies [07:17] i've kind of figured out that it wroks [07:17] works [07:17] I can't see an easy way to fix that, apart from dragging .nz somewhere else, or laying lots more undersea fibre [07:17] but now i want to try doing tcp acceleration [07:17] mm, [07:18] well my proxy system does google's initcwnd thing [07:18] like with sending more pcakets initially [07:18] which helps a lot [07:18] Ahh, window size stuff [07:18] but also like lots of sites don't have window sizes that get big etc [07:18] yeah [07:18] but even being closer to a destination [07:18] if a site has a small amount of packet loss [07:18] and you're close [07:18] resends have to go less distance [07:19] so as long as you don't have packet loss to a site that's near it [07:19] if the site near it has a big of packet loss to destination it doesn't hurt nearly as much. [07:19] but like, i suppose i'm a bit of a geek, i just want to see how fast i can actually get web browsing to get :) [07:19] which means having good peering etc. [07:20] but i can't think of any cheap way to do that [07:20] in my tests i've noticed that you have to be within about 20 msec from destination to get good speeds. [07:21] but like the internet is a huge place.. you can't really be 20 msec from everywhere [07:21] like norway is over 20 msec frmo london [07:22] but how many web sites other than opera are in norway? [07:23] actually looking at a map, i think norway could be 20 msec from london if there was a direct cable [07:24] i assume there probably isn't [07:35] I've got to run [07:35] mercutio: Nice talking to you. [08:49] *** dxtr has quit IRC (Ping timeout: 250 seconds) [08:56] *** dxtr has joined #arpnetworks [08:57] *** ivan-kanis has quit IRC (Remote host closed the connection) [09:02] *** ivan-kanis has joined #arpnetworks [09:06] *** ivan-kanis has quit IRC (Remote host closed the connection) [09:08] *** ivan-kanis has joined #arpnetworks [09:11] *** ivan-kanis has quit IRC (Remote host closed the connection) [09:14] *** ivan-kanis has joined #arpnetworks [09:22] *** ivan-kanis has quit IRC (Remote host closed the connection) [09:23] *** ivan-kanis has joined #arpnetworks [09:26] *** ivan-kanis has quit IRC (Remote host closed the connection) [09:36] *** ivan-kanis has joined #arpnetworks [09:42] *** ivan-kanis has quit IRC (Remote host closed the connection) [09:43] *** ivan-kanis has joined #arpnetworks [09:48] *** ivan-kanis has quit IRC (Remote host closed the connection) [09:49] *** ivan-kanis has joined #arpnetworks [09:50] *** ivan-kanis has quit IRC (Remote host closed the connection) [10:00] *** ivan-kanis has joined #arpnetworks [10:01] *** ivan-kanis has quit IRC (Remote host closed the connection) [10:03] *** ivan-kanis has joined #arpnetworks [10:12] *** ivan-kanis has quit IRC (Remote host closed the connection) [10:12] *** ivan-kanis has joined #arpnetworks [10:17] *** ivan-kanis has quit IRC (Remote host closed the connection) [10:19] *** ivan-kanis has joined #arpnetworks [11:06] *** homosaur has joined #arpnetworks [11:32] *** homosaur_ has joined #arpnetworks [11:33] *** homosaur has quit IRC (Ping timeout: 258 seconds) [11:33] *** homosaur_ is now known as homosaur [12:43] *** ivan-kanis has quit IRC (Remote host closed the connection) [16:21] *** EhtyarWRK has joined #arpnetworks [18:07] *** jebba has joined #arpnetworks [18:21] *** heidar has quit IRC (Read error: Connection reset by peer) [18:22] *** heidar has joined #arpnetworks [20:32] *** homosaur has quit IRC (Quit: pocketful of goat cheese, ready to party)