[00:29] *** ryk has quit IRC (Quit: leaving) [00:32] *** sako has joined #arpnetworks [00:32] *** Ehtyar has quit IRC (Quit: Going!) [00:46] *** sako has quit IRC (Ping timeout: 248 seconds) [00:55] *** k3asd has joined #arpnetworks [00:58] welcome k3asd [00:58] up_the_irons, thanks [00:58] how are you? [00:58] good, yourself? [00:59] just wondering, how did you come to find our channel? I plastered it on our vps page header today... [00:59] wondering if that worked [00:59] freebsd port updates, wee! [01:00] fine, thanks [01:01] I find the channel list on freenode and this interested me [01:02] I'm here a few days [01:02] ah cool [01:02] but I never opened my mouth :) [01:04] you vps prices are competitive [01:11] k3asd: yeah, i'd like to think so :) [01:12] competitive, but not stupidly cheap. I like to maintain a certain quality of service. :) [01:12] hahaha [01:12] stupidly cheap ones are awesome [01:12] maybe i should use that as a tagline [01:12] i kinda want to do an analysis of price vs reliability [01:12] up_the_irons, hahahahaah [01:13] lovevps offers some pretty stupid cheap prices and i've never had downtime, but i don't put critical stuff there [01:13] Webhostbudd_: i've been in this industry a long time; you get what u pay for, that's really what it comes down to [01:13] up_the_irons, in Italy, prices are higher and the quality is not always that you are looking for [01:13] then linode should give me some really good vps service lol [01:13] k3asd: i c [01:13] *** k3asd is now known as k3asd` [01:14] Webhostbudd_: well, Linode is pretty good from what I'v heard :) They also have a ton of features [01:14] up_the_irons: i've had too many outages with them, and im not a fan of their default images [01:14] up_the_irons, and pricing isn't grat [01:14] Webhostbudd_: o'rly, lots of outages? [01:15] Webhostbudd_: their pricing is pretty high nowadays, i would agree [01:15] up_the_irons: seems like my node always had problems, so im sure the experience varies [01:15] roger [01:15] i dunno, kvm always seemed better than xen anyway, and i can scratch my freebsd itch more easily on kvm [01:16] up_the_irons, i c ? which means? [01:16] k3asd: i see? [01:16] k3asd`: "i see" [01:16] ah, it's ok [01:16] Webhostbudd_: kvm rocks, hands down. xen had its day and was great ,but technology has moved on [01:16] my english is funny :) [01:17] k3asd`: you are in italy? [01:17] up_the_irons: that's all i deploy anymore, i just wish i had a decent internet connection [01:17] Webhostbudd_: decent internet where? [01:18] well, im talking a hosting setup, namely symmetric upload/download speeds [01:18] but of course that would be considered a "business feature" if even available, and pricey as heck [01:18] up_the_irons, live and work in Cagliari ( Sardinia ) [01:18] Webhostbudd_: ah right [01:19] k3asd`: cool [01:19] I think we are a bit far away [01:19] k3asd`: what time is it over there? :) [01:19] it's nineteen past ten am [01:19] good morning :) [01:19] from you? [01:19] ahahahha thank you :D [01:19] :) [01:20] such a nice temperature outside now... this only happens like 2 months out of the year in SoCal [01:22] internet has come down in cost over the years [01:22] yea, but speeds haven't increased enough [01:22] up_the_irons, here also reaches 40 degrees [01:22] is not beautiful :/ [01:22] webhost: how so? [01:22] up_the_irons, we are you from? [01:23] 10 years ago i was on 128kbit [01:23] mercutio: i guess i am one of those people who wishes it was always faster [01:23] webhost: so do i :) [01:23] mercutio: pfft i was there 5 years ago [01:23] now im only on 5mbit [01:23] 700kbit up [01:23] but from where things are now things are complicated. [01:23] ok [01:23] that's kind of slow [01:23] how much deos that cost? [01:23] $50/mo [01:23] ok [01:23] so you could get two of them [01:23] and bond them [01:24] I'm not sure you could actually do that [01:24] mm [01:24] with this provider [01:24] is that kind of internet speed common there? [01:24] k3asd`: Los Angeles [01:24] I have one possible choice for ISP, but not really [01:24] born and raised [01:24] so most people are much faster? [01:24] because i happened to be looking at mpd [01:24] mercutio: depends on where they live, some of them have access to verizon fios [01:24] which allows multilink ppp [01:25] up_the_irons, cool :) [01:25] mm [01:25] :) [01:25] that's a lot faster isn't it? [01:25] up_the_irons, is my dream go to Los Angeles [01:25] i want to setup bonding sometime heh [01:25] mercutio: it could be a lot faster, yea [01:26] mercutio: but you could pay for a really cheap, slow plan [01:26] oh? [01:26] can you resell dsl connections there? [01:26] no idea [01:27] not going to lie, this refund through google might have been the most painful refund process ever [01:27] hmm 5 megabit is adsl 1 ? [01:27] mercutio: not using asdl? [01:27] cable? [01:27] yup [01:27] docsis 1? :) [01:28] i have no idea but i wouldn't be all that surprised [01:29] k3asd`: you're not missing much ;) [01:29] you can prob tell by your modem [01:29] it may be docsis 2 [01:29] it won't be docsis 3 [01:29] mercutio: im sure, but i don't have it on hand [01:29] but the upgrade paths are pretty smooth [01:30] it seems strange to have 5 megabit cable [01:30] i actually left that for my apartment now, and im advertised to get 30/30 Mbit speed [01:30] but im behind a nat router [01:30] joy :) [01:30] i hate nat [01:30] and with all of the other tenants the speeds become pretty abysmal [01:31] someone may be uploading too much [01:31] looking into comcast because they are really the only option here, although not great [01:31] even if you have symmetric net bittorrent or such can rape the connection [01:31] anything can really [01:31] yeh [01:31] i know plenty of people here who torrent [01:31] sfq can help [01:31] but bittorrent will go faster than other net [01:32] i have no idea what they are doing as far as QOS [01:32] but from what i can tell it might be doing more harm than help [01:32] well does it lag ot get packet lossy? [01:32] oh yea [01:32] youtube never buffers [01:32] voip dies frequently [01:32] it's probably red or wred [01:32] that may be connection limits [01:32] exhausting nat table [01:32] yea [01:33] mm [01:33] it's honestly a bad time [01:33] have you tried a gre tunnel? [01:33] nope [01:33] gre is connectionless / stateless [01:33] which might mean you can fix state beetween [01:33] but generally speaking [01:33] yea, im not going to mess around much [01:34] if you get >5% packet loss [01:34] you're going to have problems [01:34] it's not that bad [01:34] like ssh connections stuttering, web page loads halting [01:34] ok cool :) [01:34] it pisses me off, and i would rather have public ip's / ipv6, so comcast will hopefully do the trick [01:34] yeh [01:35] i already have a docsis3 ipv6 modem [01:35] *** sorressean has quit IRC (Ping timeout: 260 seconds) [01:35] docsis 3 isn't actually too bad [01:35] it's not as "fast" as fibre [01:35] *** sorressean has joined #arpnetworks [01:35] i just have to pay these bastards to flip the switch to turn the cable service on [01:35] but like [01:35] transit is usually what makes speed difference [01:35] i see [01:35] at those kinds of speeds [01:35] so it doesn't relaly matter [01:35] and existing cable can be used... [01:36] like docsis 3 can do bonding and do hhundreds of megabits [01:36] but that's basically multiple access profiles [01:36] nice [01:36] but i think per connnection it's like 60megabit+ anyway [01:36] the problem is [01:36] existing docsis 2 users slow docsis 3 users down [01:36] so every docsis 2 user that moves to docsis 3 [01:36] frees up bandwidth and speeds up the net [01:37] consequence of having cable, shared transmission lines [01:37] soemtimes they hide these as free speed upgrade but you need a new modem. [01:37] yeah but it has to be shared somewhere anyway [01:37] you know when people kept saying that adsl used to use atm [01:37] and that was a shared pipe [01:37] mhmm [01:37] that could easily lead to congestion [01:38] because atm it'd be common to split the shared pipe [01:38] the old problem with cable was upload ocngestion [01:38] up_the_irons, so so :) [01:38] but that's been improved a lot these days [01:38] up_the_irons, arpnetworks are you society? [01:38] like when bittorrent etc first came out, they'd hardly provisioned enough upload bandwidth [01:38] it still has a ways to go unfortunately [01:38] k3asd`: it's my business, yeah [01:38] real? [01:38] i just appreciate that some companies are paving the way for things to improve [01:39] i dunno i've never used any amazingly-fast-net [01:39] mercutio: i really haven't either [01:39] there's just less slow and more slow [01:39] hahaha [01:39] i mean i've used by laptop on gigabit net [01:39] it still felt slower than my desktop on adsl. [01:39] unless you live in kansas city [01:40] sonic are doing gigabit i think? [01:40] that was at a data centre [01:40] dunno [01:40] but like seriously, it's way faster than dialup used to be [01:40] i won't disagree with thta [01:41] patches took ages [01:41] and dealing with a single phone line was pretty bad [01:41] patches? [01:42] i had two phone lines [01:42] games / operating systems [01:42] oh ok [01:42] i wish we had two phone lines [01:42] i didn't find patches that bad [01:42] well [01:42] os/2 was special [01:42] they made it so difficult to update [01:42] it got a little better [01:42] but i didn't upgrade as much back then [01:42] i still have my os/2 warp box and disks [01:43] up_the_irons: did you ever try tshell? [01:43] mercutio: no [01:44] there were a few kind of cool things like that around [01:44] there was another one that just simplied the wps [01:44] but like tshell made os/2 fast on 8mb ram [01:44] that was really the issue with os/2 [01:44] it wast oo memory hungry for computers that were common [01:45] also ppp was way more complicated than windows [01:45] lol serenity systems still releases bew versions of os/2 [01:45] wth [01:45] new versions* [01:45] webhost: tbh os/2 2.1 through 4 weren't that diff [01:46] i think i went back to 2.1 in the end [01:46] 3 was generlaly considered the safe easy way to go [01:46] 4 was like you had to update it with fixpacks to have minimal stability [01:46] and it was kind of just adding lots of crap [01:46] interesting [01:46] but 2.1 didn't come with tcp/ip [01:47] so you had to use that extra thingy [01:47] well, ecomstation 2.1 supports pretty recent firefox stuff [01:47] which is interesting [01:47] firefox? eww :) [01:47] i just can't believe someone develops it anymore [01:47] and claims they are profitable [01:47] who buys it [01:48] you must be stuck in a real mess if you still need os/2 machines around [01:48] i dunno [01:48] os/2 wasn't that bad [01:48] you should have seen the documentation [01:48] it's like OMG SO MANY BOOKS [01:48] but it has good references for every function call etc [01:48] quite different from windows [01:49] it was like an encyclopedia collection though [01:49] hahaha [01:50] *** Webhostbudd_ has quit IRC (Quit: Leaving) [02:15] *** heavysixer_ has joined #arpnetworks [02:15] *** ChanServ sets mode: +o heavysixer_ [02:16] *** himuraken has quit IRC (*.net *.split) [02:16] *** heavysixer has quit IRC (*.net *.split) [02:16] *** heavysixer_ is now known as heavysixer [02:24] up_the_irons, congratulations. I do a similar job [02:25] but in this moment I trying to find work in the UK [04:12] *** himuraken has joined #arpnetworks [04:48] freebsd is kind of scarily dated in places [04:48] like pf is YEARS behind [05:08] That's scary? [05:09] There was talk of them branching PF away from OpenBSD anyway. [05:09] Not sure what the decision was. [05:09] .. or if there even was one [05:12] branching pf away? [05:12] Yes. [05:15] what do you mean? [05:16] I mean not tracking the OpenBSD source, rather maintaining their own codebase based on the older version of PF. [05:16] pf is pretty awesome, compared to iptables. [05:17] what! [05:17] Obviously, yeah. [05:17] haha [05:17] they should sync up [05:17] iptables is the most painful thing ever [05:17] mercutio: Why? [05:17] freebsd has already broken lots of things! [05:17] what's with this /etc/rc.d etc? [05:17] What's wrong with /etc/rc.d ? [05:17] there's no reason to hold back if you're making radical cahnges anyway [05:17] well it's not /etc/rc.local [05:17] You just contradicted yourself I think. [05:17] what i mean is that they're drastically changing things [05:18] in incompatible ways [05:18] Incompatible with what? [05:18] that are more significant than the minor syntax change in pf [05:18] I see no problem with drastic change where drastic change is due. [05:18] with standard /etc/rclocal [05:18] err /etc/rc.local [05:18] yeh [05:18] well it's ok to change/break things sometimes [05:18] so people have to update their configs [05:18] so they should just sync with current openbsd [05:18] It's life with software. [05:18] Why? [05:19] I don't understand that. [05:19] because there are performance enhancements [05:19] and it's more flexible [05:19] and it means i don't have to screw around with the syntax when copying rules to freebsd [05:19] from openbsd.. [05:19] I dunno, I guess it's a difference of opinion. [05:20] I don't feel each project has to be the union of all BSD projects. [05:20] what i actually find most annoying about freebsd is the lack of default config files. [05:20] OpenBSD is far better than anything else as a network device, so let it do its job. [05:20] FreeBSD is far better for serving, so ... [05:21] I would actually prefer less overlap. [05:21] mm [05:21] I hate when shit from one project is hacked into the other just for the sake of doing it. [05:21] well openbsd doesn't have netgraph [05:21] It creates a mess. [05:21] and so mpd doesn't work on openbsd [05:22] Yeah, there are exceptions, they're open source projects and one is far more popular than the other. [05:22] There are differences of opinions too, so there's natural disorganization among the projects. Their efforts are definitely not aligned to some greater goal as well as they should be. [05:23] hmm [05:23] That's probably due to the problems FreeBSD folk have with Theo [05:23] Who knows... [05:23] yeh things are a little fragmented [05:23] l2tp support on unix in general is a bit behind [05:24] But, in a perfect world, the two projects would work together to be more organized. Tools necessary for network devices would be pushed into OpenBSD land and tools required for serving in FreeBSD. There would naturally be overlap, which could be optimized too. [05:24] No such world exists, though. So, it is what it is. [05:24] openbsd needs to do smp for networking sometime [05:25] Yeah, really. [05:25] I fully agree. [05:25] Also, ospfd shouldn't go completely insane when a new neighbor arrives. [05:25] haha [05:26] They also need to give you more than an on/off switch for routing daemons. It'd be nice to prioritize startups and whatnot. [05:27] I'd like my IGP converged before BGP starts advertising routes out my transit peers that don't rely on IGP. [05:28] is that openospfd? [05:28] or another? [05:28] I use OpenOSPFd, yes. [05:28] openbgpd seems slow on freebsd for some reaosn [05:28] ie takes longer than openbsd to load full rout able [05:28] route table [05:29] openospfd port looked older than openbgpd [05:29] i'm using using bgp as IGP [05:29] have been since used to use quagga years back [05:29] I just use OSPF to validate nexthops. [05:29] and quagga's ospf was too buggy [05:29] I basically carry nothing in OSPF. [05:29] ahh ok [05:29] Everything should be BGP, it actually scales. [05:29] haha [05:29] i was kind of idlely curious if i could load balance outgoing routes in freebsd [05:30] have you ever tried bird? [05:31] Nah, never. I've had good luck with OpenBSD and I hate fiddling with things until they work. I just want it to work out of the box. [05:31] ahh yip [05:31] i just tried it the other day on linux [05:31] not advertising any routes [05:31] it's more confusing than openbgpd to me [05:31] haha [05:32] but it seems preferable to quagga [05:32] Static routing seems preferable to quagga [05:32] hahaha [05:32] it's gross [05:32] That's gotta be the must annoying piece of software on Earth, aside from Microsoft Word. [05:32] well zebra was bad too [05:33] Truth is, I'd ditch all this shit in two seconds if my company would cough up the money for some Juniper hardware. [05:33] heh [05:33] I might use it on our private networks, MIS, corporate, etc... But I'd love Juniper in the backbone. [05:34] mm both transit links that i use openbgpd with are on juniper gear [05:34] Wouldn't waste the money on doing Juniper routers for offices and stuff. [05:34] Free or Open could handle that type of networking easily. [05:34] they seem pretty common here [05:34] both providers are using mx480s [05:35] i don't think mx480s are cheap [05:35] No, I'm sure they're not. [05:35] Not much (that's worth buying) in the MX line is cheap [05:36] http://www.ebay.com/itm/New-Juniper-MX480-PREMIUM-DC-Base-system-with-redundant-RE-2000-SCB-DC-Power-/251131667989?pt=US_Enterprise_Routers&hash=item3a789d2615 [05:36] hmm $35k usd [05:37] Doesn't Juniper refuse to support hardware that was sold like that? [05:37] Err hardware that changed hands, I guess [05:37] not sure [05:38] but it gives an idea on price [05:38] it's not like they're likely to list any prices online [05:38] Yeah. [05:38] You could always look on CDW or something, and divide in half. [05:38] haha [05:39] it's basically the same ast hat calculation [05:40] maybe slightly more that way [05:40] never heard of cdw [05:40] They're big in the US. [05:40] Expensive, but reliable type of thing. [05:41] ahh [05:41] i'm in new zealand [05:41] I see, yeah, I know nothing about what's out there. [05:41] hmm [05:41] Except hobbits and the like. [05:41] :) [05:41] heh [05:42] i think for isp type people juniper are very common here [05:42] for corporate type people cisco are very common [05:42] Same here and probably everywhere. [05:42] heh [05:42] brocade are more common there than here i think [05:43] and extreme networks etc [05:43] Yeah, in the past few years I've heard lots of chatter about them, but I've never come across one in any of my work experience. [05:43] hmm [05:43] Generally Cisco, Juniper or BSD. [05:43] Much less BSD than the others. [05:43] an open source router would be kind of cool [05:43] with asic forwarding [05:43] Only a matter of time. [05:43] but the software is so far behind in those regards. [05:44] yeah when i shifted to openbgpd i didn't know anyone running openbsd at all [05:44] i inherited zebra/quagga network. [05:45] and for some reason it'd fuck up really badly sometimes :/ [05:45] i thought vyatta was somewhat an open source router with asic forwarding (granted, probably not a lot of routes, but something) [05:45] vyatta is pretty new [05:45] A friend of mine did quite a bit of work making FreeBSD route fast for use in OCCAID, so I heard of him doing crazy shit with BSD for routing, but otherwise, I knew of nobody doing it. [05:45] they do software for base machines i know [05:45] but they probably have a hardware offering. no idea about asic forwarding [05:46] easymac: is your friend named dima? sounds like a guy i know... [05:46] blahdy [05:46]