[01:12] *** LT has joined #arpnetworks [01:19] *** dj_goku has quit IRC (Read error: Connection reset by peer) [01:25] *** dj_goku has joined #arpnetworks [04:31] *** LT has quit IRC (Quit: Leaving) [06:15] *** toeshred has quit IRC (Quit: WeeChat 1.0) [06:20] *** toeshred has joined #arpnetworks [07:06] *** SpaceDum1 has joined #arpnetworks [07:07] *** tellnes_ has joined #arpnetworks [07:11] *** SpaceDump has quit IRC (*.net *.split) [07:11] *** tellnes has quit IRC (*.net *.split) [07:11] *** tellnes_ is now known as tellnes [07:47] *** gizmoguy_ has joined #arpnetworks [07:49] *** gizmoguy has quit IRC (Ping timeout: 272 seconds) [07:53] *** awyeah has quit IRC (Ping timeout: 272 seconds) [07:56] *** awyeah has joined #arpnetworks [08:03] *** tellnes_ has joined #arpnetworks [08:05] *** SpaceDump has joined #arpnetworks [08:06] *** tellnes has quit IRC (Ping timeout: 260 seconds) [08:06] *** SpaceDum1 has quit IRC (Ping timeout: 260 seconds) [08:06] *** tellnes_ is now known as tellnes [08:33] *** tooth has quit IRC (Ping timeout: 250 seconds) [08:33] *** CaZe has quit IRC (Ping timeout: 250 seconds) [08:33] *** brycec has quit IRC (Ping timeout: 250 seconds) [08:33] *** Seju has quit IRC (Ping timeout: 250 seconds) [08:33] *** qbit has quit IRC (Ping timeout: 250 seconds) [08:33] *** Seji has joined #arpnetworks [08:33] *** qbit_ has joined #arpnetworks [08:33] *** tooth_ has joined #arpnetworks [08:33] *** CaZe_ has joined #arpnetworks [08:33] *** tooth_ is now known as tooth [08:33] *** CaZe_ is now known as CaZe [08:33] *** davantalus has quit IRC (Ping timeout: 250 seconds) [08:33] *** [FBI] has quit IRC (Ping timeout: 250 seconds) [08:33] *** [FBI] starts logging #arpnetworks at Fri Sep 19 08:33:22 2014 [08:33] *** [FBI] has joined #arpnetworks [08:33] *** davantalus has joined #arpnetworks [08:35] *** brycec has joined #arpnetworks [08:38] *** qbit_ is now known as qbit [08:45] *** josephb has joined #arpnetworks [08:46] *** mhoran_ has joined #arpnetworks [08:46] *** ChanServ sets mode: +o mhoran_ [08:46] *** mhoran has quit IRC (Ping timeout: 250 seconds) [08:46] *** josephb_ has quit IRC (Ping timeout: 250 seconds) [09:04] *** mhoran_ is now known as mhoran [09:30] *** toddf has quit IRC (Remote host closed the connection) [09:31] *** toddf has joined #arpnetworks [09:31] *** ChanServ sets mode: +o toddf [10:22] *** pyvpx has quit IRC (Remote host closed the connection) [11:55] *** SpeedBus has quit IRC (Ping timeout: 260 seconds) [13:08] *** jdeuce has joined #arpnetworks [13:08] how can I destroy / terminate a VM? [13:10] Email support@ [14:27] *** SpeedBus has joined #arpnetworks [16:28] http://www.martinvigo.com/a-look-into-lastpass/ [16:39] I don't think English is the native language... And I'm very confused what this is supposed to mean "2 factor authentication should be the way to go but LastPass has this option. " [16:39] "It should be this way, BUT it is this way." [16:46] http://steamcommunity.com/id/staticsafe/screenshot/545256983451439785 [16:55] ? [17:26] yeah, it's not the dudes' native language [17:27] haha, nice [17:27] staticsafe: the farm is real: https://www.planetside2.com/players/#!/5428059527817879329/killboard [17:27] man [17:28] whenever this happens, i redeploy [17:28] i'm a terrible person, but they're bad for not killing me [17:28] i hate the mission stuff, it always glitches out for me [17:30] and i especially hate that it displaces the cap point status for the area you're actually in, in favor of showing you where it thinks you should be [18:07] *** law has joined #arpnetworks [18:07] hey all, BGP n00b here, quick question: [18:07] Everybody, meet law [18:07] law, meet everybody [18:07] let's say you're a hosting provider. you own IP 1.2.3.4, but you want it hosted in multiple, geolocation-independant datacenters, with automatic failover [18:08] so, you advertise routes for that IP, both with a weight of '5' (for example) [18:08] people closer to datacenter A (from a routing standpoint) will naturally filter into datacenter A, and people closer to datacenter B (still from a routing standpoint) will filter into datacenter B, both hitting that IP [18:08] what happens if someone is routing-wise equidistant to both A and B? [18:10] (there's much lurking, especially on a Friday night) [18:12] hehe fair enough [18:12] hello law, nice to meet you [18:12] up_the_irons: \o/ [18:13] (I brought law here with his BGP question because #arpnetworks has the most BGP nerds I've ever seen) [18:14] nice to meet you too [18:14] law: there is no hard rule for what happens in that case. the bgp route selection algorithm running on the particular device sending out the packets will make a decision. Lookup Cisco's BGP Best Path Selection Algorithm and you'll get an idea of what other factors come into play. [18:15] fair enough. I was asked this in a job interview question the other day, and I had *no idea* :-) [18:15] brycec: lol really? i didn't even know we had lots of bgp nerds... now i know where to ask my own bgp questions ;) [18:15] lol up_the_irons ;) [18:16] I was asked to design a theoretical web service that was redundant across datacenters, so my immediate thought was 'BGP!' [18:16] "lots" for BGP isn't exactly a "big number" :P [18:16] I don't have a ton of BGP experience, but I know enough to talk about it at a theoretical level [18:16] that's true [18:16] then they popped that one on me and my answer was "I'm pretty sure it would more or less evenly split the traffic, but I don't know for sure" [18:19] the intent being that if datacenter A got hit by a meteor, that path's weight would automatically drop to 0 since it was no longer active, and everybody would start routing to datacenter B [18:20] * brycec loves theoretical meteorites [18:21] meteorite, backhoe error, bus-meets-telco-pole condition, etc [18:21] is that a valid/common usage of BGP, or was I pretty much just talking out of my backside on that one? [18:21] Very common [18:22] eeeexcellent :-) [18:22] For instance, 8.8.8.8 and 8.8.4.4 are both anycast [18:22] (that's what it's called, btw - anycast) [18:22] not BGP? [18:22] or BGP-as-used-for-redundancy-failover? [18:22] Anycast is when you broadcast the same /24 from multiple locations and let the routers decide which path to take [18:22] It's a specific application of BGP [18:23] @wiki anycast [18:23] Anycast :: Anycast is a network addressing and routing methodology in which datagrams from a single sender are routed to the topologically nearest node in a group of potential receivers, though it may be sent to several nodes, all identified by the same destination address. Addressing methodologies The Internet Protocol and other network addressing systems recognize four main... http://en.wikipedia.org/wiki/Anycast [18:23] But I think you can have anycast without BGP... [18:23] True, technically [18:23] the "one-to-many" relationship in anycast is theoretically possible with any routing protocol [18:24] TIL. thanks! [18:24] "anycast is usually implemented by using Border Gateway Protocol" keyword "usually" :p [18:26] law: it wouldn't even split traffic in your example above. at least, not on a particular router. The router wouldn't say, "I have two equal cost paths to the same destination, let's load balance the traffic". One route, and one route only, is selected and then installed into the FIB (forwarding information base). I suppose specific configurations could be made to load balance the traffic, but i [18:26] haven't seen this done personally in a long time [18:26] *evenly [18:27] Thanks for the clarification [18:27] no problem! [18:27] And to be clear (for law), "the router" is referring to the mystical customer that is equidistant between routes [18:27] yeah [18:28] Their routing equipment would pick a route and stick with it. [18:28] yeah [18:28] yeah