[01:25] joy email validations work [01:25] tomorrow, I implment a beta invite system [01:26] well rather later today [01:26] now I sleep [01:26] up_the_irons: don't stay up to late [01:26] ballen: i'll be fine [01:26] :) [01:26] heh indeed [01:26] i got a lot of sleep yesterday [01:27] yea who needs sleep two days in a row [01:34] srsly [02:07] police helicopter shouting some stuff, heh... [02:07] my neighborhood sucks [02:07] lmao [02:57] *** ballen is now known as ballen|away [03:59] 64-bit cacti problems solved [03:59] [04:01] maybe i'll blog about it [05:12] ballen|away: I sorta understand why you're using Sinatra, but one you go looking for plugins, etc., for it it makes more sense to use Rails. [05:13] (If you used Rails you could use Clearance which handles email validations.) [05:25] morning all [05:49] *** heavysixer has joined #arpnetworks [06:05] *** heavysixer has quit IRC () [06:21] mornin'? [06:21] it aint mornin till i've slept [06:22] i have requested peering with 30 more networks. OMG I am tired [06:22] cd $bed [06:23] Word. [06:23] Peer with GOogle yet? :) [06:24] Wonder how many VMs you have to host for them to think that'd be practical. I think you also need two peering points? I looked into this a while ago. [06:38] *** heavysixer has joined #arpnetworks [07:01] mhoran: nah, google still giving me the finger ;) and yeah, they require multiple locations now [07:01] mhoran: as far as # of VMs, at the rate you guys use b/w you'd need a METRIC SH*T TON of VMs ;) [07:02] [06:18:40] @up_the_irons | cd $bed [07:02] up_the_irons: ^^ liar [07:02] Haha. [07:03] mhoran: just got off a nice call with Cable & Wireless; they didn't 'em up wanting to peer, but they gave me the time of day and were very nice. That's the diff between USA and GB I guess ;) [07:03] Hah, cool. [07:03] looks like I need in the range of 7+ Gbps of traffic before they'll reconsider [07:03] We'll get there! [07:03] * up_the_irons <-- way off from that ;) [07:04] Wish I were hosting something that were that popular, hah. [07:04] 7 Gbps. Wow. [07:04] I just looked at that number again, hah. [07:05] the guy was talking about wholesale transit rates, and costs getting across Russia and Pacific Ocean, and how as a result "content" traffic is becoming lower and lower in value wrt peering [07:05] Oh really. [07:05] Interesting. [07:05] mhoran: LOL, yeah, 7 *G*bps [07:06] other types of traffic are becoming more sought after, like voice traffic [07:06] Right. [07:06] i mean, with all the content we spill out for free from networks in the US, it is no wonder the rest of the world is becoming less likely to pick up the bill ;) [07:07] Yeah. I've always been interested in the point where an ISP is interested in peering vs. transit, and what it takes to get there. [07:08] yeah, i find that interesting as well. i know the basics, but some of these guys that are *into it* are really fascinating to listen to [07:29] cd $bed for real [07:55] *** vtoms has joined #arpnetworks [07:55] *** visinin has joined #arpnetworks [08:17] *** heavysixer has left [08:18] *** heavysixer has joined #arpnetworks [09:50] *** ballen|away has quit IRC ("Leaving...") [09:57] *** ballen has joined #arpnetworks [10:01] *** iStormy_WY5Q has joined #ARPNetworks [10:37] *** ballen is now known as ballen|away [10:38] *** heavysixer has quit IRC () [11:09] *** heavysixer has joined #arpnetworks [11:49] *** ballen|away is now known as ballen [11:50] Are there bandwidth graphs in the control panel? If so, I must be missing something. =x [11:54] yea up_the_irons has to add them, not added automagically [11:54] just send a request to support@arpnetworks.com [11:54] oh [11:54] ok [11:55] Will do. Just on the lunch break right now [11:55] right on, just woke up [11:55] staying up till 6am not a smart idea [13:23] Lol. No, ,that is never a good idea. [14:04] ballen: have you rebuild your kernel on your vps? [14:04] -d +t [14:04] like you would in any install [14:05] oh [14:05] have i [14:05] nah [14:05] no reason to [14:05] GENERIC kernel is just fine for me [14:06] I used to do custom kernels in FreeBSD all the time, but have found theres really not any noticeable benefit [14:08] I wasn't going to but I want to update the user land so I thought I might as well rebuild the kernel too [14:08] I've never rebuild a 64bit kernel though. It seems different [14:09] yea its a little diff [14:09] you're not doing like a make world are you? [14:09] freebsd-update ... [14:09] its how you do that now [14:09] I was going to do a make world. lol [14:10] I haven't used BSD in a long time. x [14:10] =x [14:10] freebsd-update is amazing [14:10] freebsd-update fetch [14:10] then freebsd-update install [14:10] cool [14:10] soooo much better than make world [14:10] about time it's simple. lol [14:10] also portsnap [14:10] for grabbing ports and updating the ports tree [14:11] portsnap fetch extract [14:11] portsnap fetch update [14:11] yea freebsd-update came out around 6.1 or 6.2 ish [14:12] and it was in ports before that [14:13] is this like all the ipv6 you can eat or something [14:13] lol [14:13] that is a crap load of ipv6 [14:14] /48 yea [14:14] its a retarded amount [14:14] must be trying to push people away from ipv4. lol [14:15] supposidly its the standard recommended subnet to push to hosts in the RFC [14:15] which is wild [14:15] and they wonder why they're out of ipv4. ha! [14:16] my thoughts excactly [14:16] I believe a /48 is more ip's then there is in ipv4 [14:16] yeah, if you read the RFCs, between /48 and /56 is the recommended allocation, with the /56 being geared toward residental (Cable/DSL/Dialup) ISPs [14:17] ballen: yeah, it is more than entire ipv4 [14:17] heh [14:17] brilliant [14:17] but one thing you need to keep in mind is, you wouldn't be assigning /128's to each host [14:17] the idea is this: [14:17] IPv6 is so large, we don't "think" of it in terms of individual IPs anymore [14:18] we thing in terms of subnets instead [14:18] yea [14:18] should just get rid of port numbers [14:18] how many "subnets" is a /48 ? Well, the standard subnet size is /64, so it is 2^(64 - 48) [14:18] and *that's* what you consider to be how "big" your block is. how many subnets do I have? instead of, how many IPs do I have? [14:19] true, but you can actually give a host a /128 so you still have a crap load [14:19] right, you *can*, but you lose a lot of the features built into IPv6 [14:19] oh well [14:20] whaaatever [14:20] but anyway [14:20] how many /48's can I give out before I run out? [14:20] 65536 [14:20] if i had that many customers, i'm retiring [14:20] heh [14:20] indeed [14:20] and ARIN reserves like the neighboring 8 or so /32's [14:20] So I could increase to a /24 [14:21] lmao [14:21] just in case [14:21] by just changning the subnet mask, no additional blocks needed [14:21] this is why the routing table for IPv6 is going to be a lot smaller (a lot less blocks) [14:21] ah [14:22] let's see, if I had a /24 [14:22] I could give out 16777216 /48's [14:22] then I'M DONE [14:22] DONE [14:22] i tell you [14:22] it's a little scary that some ISPs already have /21's of IPv6 [14:23] are they just hording ? [14:23] hoarding [14:23] nah, i think they actually have millions of customers [14:23] damn [14:23] like AT&T, Level3, etc... [14:23] ah [14:23] yea [14:24] how much does a /48 cost ya [14:24] err [14:24] /32 [14:25] how differently it is to rebuild a 64-bit kernel vs 32-bit? [14:25] Nothing so far, they let ISP's that already have IPv4 get an IPv6 block for free. This is like to "encourage" adoption, but in the end they'll charge at some point. Probably $2250 per year [14:25] just different options [14:26] any options that I have to worry about since it's a vps? [14:26] iStormy_WY5Q: oh hey, welcome to #arpnetworks :) [14:26] specific^ [14:26] just compare against dmesg [14:27] alright. I'll give it a shot. I don't want to kill the box already. -_- [14:27] teknicaL: I barely noticed the 64-bit kernel changes. Mostly the same stuff; just modify GENERIC and follow dmesg. [14:27] just remember how to recover [14:27] boot kernel.old [14:28] or whatever it is from the boot loader [14:44] *** ballen is now known as ballen|away [15:00] *** ballen|away is now known as ballen [15:06] *** vtoms has quit IRC ("Leaving.") [15:46] *** heavysixer has quit IRC () [16:24] *** baklava has joined #arpnetworks [16:37] *** ballen has quit IRC () [16:40] *** Rick has quit IRC (Read error: 104 (Connection reset by peer)) [16:47] cablehead: omg, Joel finally got back to me and said he'd peer ;) [16:47] ironically, I sent out 30 peer requests last night, but not to hiim [16:48] up_the_irons: oh wow [16:48] up_the_irons: nice :) [17:11] :) [17:20] irons: What's new? [18:10] man i've got a good server i need to colo [18:10] but i wanna split it up into VM's, what should i use ? [18:10] i should learn this kvm stuff, i've only deat with xen [18:12] i guess maybe i'd be stuck with xen again, i've used the dtc panel by gplhost.com and it's pretty cool [18:12] but i dont like debian [18:13] Xen is bad news. [18:13] Check out KVM. [18:13] I'm hating my life, thanks to Xen. [18:14] yea, i want to use somthing where i could have multiple freebsd vps's [18:14] jeev: I'll help ya test a couple. :) [18:15] i brought the server up at the office [18:15] it's running bsd right now but i assume i have to have it onl inux first [18:17] hehe Nat_UB [18:17] i actually need to add more ram first [18:17] usable memory = 4284665856 (4086 MB) [18:17] avail memory = 4112187392 (3921 MB) [18:17] i guess it's not ad [18:17] bad i dont want to use it for anything hectic [18:17] and this is for a new POP [18:17] *** mhoran has quit IRC (Read error: 60 (Operation timed out)) [18:19] *** mhoran has joined #arpnetworks [18:20] Hm. [18:20] *** up_the_irons has quit IRC (Remote closed the connection) [18:20] *** nuke` has quit IRC ("gone.") [18:20] up_the_irons: What was that about? [18:20] Oh, gone. [18:20] Yeah! [18:22] Major lag. [18:22] Couldn't SSH in for a good minute. [18:22] maybe he's trying to set up the peering [18:22] 75% packet loss! [18:22] SLA! [18:22] *** nukeAFK has joined #arpnetworks [18:24] jeev: I'd tell you more about my fun with Xen, but this isn't working. :) [18:25] *** up_the_irons has joined #arpnetworks [18:25] *** ChanServ sets mode: +o up_the_irons [18:26] ah [18:26] so give me an idea about kvm [18:26] can i run it off bsd or dose it have to be linux [18:26] As of now, it's Linux only. [18:26] i guess i should read [18:26] ahh [18:27] There's been an attempt made to port it to FreeBSD. [18:27] However, my FreeBSD VM runs pretty well in KVM (now that we've ironed out some issues with the timer.) [18:27] KVM runs rather nicely on Linux now...well supported and documented [18:27] The Xen team is far too fractured, and can't pick something to finish -- they get too interested in the latest and greatest, but never fully polish a feature, so it just doesn't work. [18:28] KVM support is integrated into the kernel (which is where the Xen folks would like to be, but they can't get their shit together.) [18:28] dunno [18:28] makes me nervous to want to use it for production [18:28] it's a decent server i could split 4 ways [18:31] *** heavysixer has joined #arpnetworks [19:06] *** jester1 has quit IRC (anthony.freenode.net irc.freenode.net) [19:06] *** heavysixer has quit IRC (anthony.freenode.net irc.freenode.net) [19:06] *** Qsource has quit IRC (anthony.freenode.net irc.freenode.net) [19:06] *** nukeAFK has quit IRC (anthony.freenode.net irc.freenode.net) [19:06] *** Thorgrimr has quit IRC (anthony.freenode.net irc.freenode.net) [19:06] *** Nat_UB has quit IRC (anthony.freenode.net irc.freenode.net) [19:06] *** cablehead has quit IRC (anthony.freenode.net irc.freenode.net) [19:06] *** baklava has quit IRC (anthony.freenode.net irc.freenode.net) [19:06] *** iStormy_WY5Q has quit IRC (anthony.freenode.net irc.freenode.net) [19:06] *** timburke has quit IRC (anthony.freenode.net irc.freenode.net) [19:07] *** heavysixer has joined #arpnetworks [19:07] *** nukeAFK has joined #arpnetworks [19:07] *** baklava has joined #arpnetworks [19:07] *** timburke has joined #arpnetworks [19:07] *** Nat_UB has joined #arpnetworks [19:07] *** jester1 has joined #arpnetworks [19:07] *** Qsource has joined #arpnetworks [19:07] *** Thorgrimr has joined #arpnetworks [19:07] *** cablehead has joined #arpnetworks [19:07] *** iStormy_WY5Q has joined #arpnetworks [19:55] up_the_irons: thanks [20:59] *** heavysixer has quit IRC () [22:00] *** ballen has joined #arpnetworks [22:12] ping [22:21] up_the_irons: you couldnt tell me when the next payment is due could you [23:16] *** toddf has quit IRC (Remote closed the connection) [23:47] obsidieth: billing is done between the 1st - 5th of every month [23:48] *** greg_dolley has joined #arpnetworks [23:48] woot beta invite system works, k going to sleep, none of this 6am shit tonight [23:48] lol, gn [23:48] night [23:49] jeev: i use KVM in production, obviously. trust me, it is quite ready. I use Ubuntu Jaunty 9.04 on the host, I recommend that, have had a real good experience [23:50] greg_dolley: whoa, you're in here [23:51] Garry != Greg? [23:53] *** visinin has quit IRC ("sleep") [23:53] ballen: no, greg is my bro [23:54] nasty up_the_irons [23:55] you use ubunti? [23:55] cancel me! [23:55] just kdiding [23:55] :) [23:55] up_the_irons, your blackberry is vibrating [23:56] no crackberry for me [23:56] you have an iphone ? [23:56] yep [23:56] damn [23:56] that's two strikes [23:57] i dunno, i guess i gotta read about kvm [23:58] if i do it, i'm doing it via slackware. [23:58] up_the_irons: well right on [23:58] actually going to sleep now, k bye [23:58] *** ballen is now known as ballen|away [23:59] ballen|away: l8r [23:59] jeev: right on [23:59] got anything i could read ? [23:59] does it require heavy mods to make it run on bsd ?