[00:40] *** xiphias has quit IRC (Ping timeout: 240 seconds) [00:41] *** xiphias has joined #arpnetworks [00:47] *** xiphias has quit IRC (Ping timeout: 240 seconds) [00:48] *** xiphias has joined #arpnetworks [00:53] *** xiphias has quit IRC (Ping timeout: 240 seconds) [00:54] *** xiphias has joined #arpnetworks [00:58] *** xiphias has quit IRC (Ping timeout: 240 seconds) [01:00] *** xiphias has joined #arpnetworks [02:22] *** CaZe has quit IRC (Ping timeout: 245 seconds) [02:22] *** CaZe` has joined #arpnetworks [02:23] *** CaZe` is now known as CaZe [02:40] *** brycec has quit IRC (Ping timeout: 264 seconds) [02:40] *** lteo has quit IRC (Ping timeout: 264 seconds) [02:40] *** gizmoguy has quit IRC (Ping timeout: 245 seconds) [02:40] *** lteo has joined #arpnetworks [02:42] *** brycec has joined #arpnetworks [02:42] *** gizmoguy has joined #arpnetworks [03:04] *** xiphias has quit IRC (Ping timeout: 240 seconds) [03:05] *** xiphias has joined #arpnetworks [03:09] *** xiphias has quit IRC (Ping timeout: 240 seconds) [03:10] *** xiphias has joined #arpnetworks [03:16] *** xiphias has quit IRC (Ping timeout: 240 seconds) [03:17] *** xiphias has joined #arpnetworks [07:03] *** sabakasmotrilol has joined #arpnetworks [07:08] *** sabakasmotrilol has quit IRC (Quit: Ухожу я от вас (xchat 2.4.5 или старше)) [07:17] *** heavysixer has joined #arpnetworks [07:17] *** ChanServ sets mode: +o heavysixer [07:44] *** heavysixer has quit IRC (Quit: heavysixer) [07:49] *** heavysixer has joined #arpnetworks [07:49] *** ChanServ sets mode: +o heavysixer [08:10] *** xiphias has quit IRC (Ping timeout: 240 seconds) [08:12] *** xiphias has joined #arpnetworks [08:18] *** xiphias has quit IRC (Ping timeout: 240 seconds) [08:19] *** xiphias has joined #arpnetworks [08:23] *** xiphias has quit IRC (Ping timeout: 240 seconds) [08:24] *** xiphias has joined #arpnetworks [08:25] *** heavysixer has quit IRC (Quit: heavysixer) [08:26] *** heavysixer has joined #arpnetworks [08:26] *** ChanServ sets mode: +o heavysixer [08:28] *** xiphias has quit IRC (Ping timeout: 240 seconds) [08:29] *** xiphias has joined #arpnetworks [08:35] *** xiphias has quit IRC (Ping timeout: 240 seconds) [08:36] *** xiphias has joined #arpnetworks [08:42] *** xiphias has quit IRC (Ping timeout: 240 seconds) [08:43] *** xiphias has joined #arpnetworks [08:47] *** xiphias has quit IRC (Ping timeout: 240 seconds) [08:48] *** xiphias has joined #arpnetworks [08:55] *** xiphias has quit IRC (Ping timeout: 240 seconds) [08:56] *** xiphias has joined #arpnetworks [09:00] *** xiphias has quit IRC (Ping timeout: 240 seconds) [09:01] *** xiphias has joined #arpnetworks [09:20] *** joepie91 has quit IRC (Remote host closed the connection) [09:22] *** brycec has quit IRC (Ping timeout: 245 seconds) [09:24] *** brycec has joined #arpnetworks [10:54] *** xiphias has quit IRC (Ping timeout: 240 seconds) [10:55] *** xiphias has joined #arpnetworks [11:05] *** xiphias has quit IRC (Ping timeout: 240 seconds) [11:07] *** xiphias has joined #arpnetworks [11:13] *** xiphias has quit IRC (Ping timeout: 240 seconds) [11:14] *** xiphias has joined #arpnetworks [11:18] *** xiphias has quit IRC (Ping timeout: 240 seconds) [11:19] *** xiphias has joined #arpnetworks [11:24] *** xiphias has quit IRC (Ping timeout: 240 seconds) [11:25] *** xiphias has joined #arpnetworks [11:33] *** xiphias has quit IRC (Ping timeout: 240 seconds) [11:34] *** xiphias has joined #arpnetworks [11:35] *** lteo has quit IRC (Ping timeout: 246 seconds) [11:35] *** lteo has joined #arpnetworks [11:39] *** lteo has quit IRC (Ping timeout: 264 seconds) [11:40] *** lteo has joined #arpnetworks [11:41] *** xiphias has quit IRC (Ping timeout: 240 seconds) [11:42] *** xiphias has joined #arpnetworks [11:48] *** xiphias has quit IRC (Ping timeout: 240 seconds) [11:49] *** xiphias has joined #arpnetworks [11:51] *** heavysixer has quit IRC (Quit: heavysixer) [11:53] *** xiphias has quit IRC (Ping timeout: 240 seconds) [11:54] *** xiphias has joined #arpnetworks [12:00] *** xiphias has quit IRC (Ping timeout: 240 seconds) [12:01] *** heavysixer has joined #arpnetworks [12:01] *** ChanServ sets mode: +o heavysixer [12:02] *** xiphias has joined #arpnetworks [12:07] *** xiphias has quit IRC (Ping timeout: 240 seconds) [12:07] *** xiphias has joined #arpnetworks [12:12] *** xiphias has quit IRC (Ping timeout: 240 seconds) [12:13] *** xiphias has joined #arpnetworks [12:20] *** xiphias has quit IRC (Ping timeout: 240 seconds) [12:22] *** xiphias has joined #arpnetworks [12:26] *** xiphias has quit IRC (Ping timeout: 240 seconds) [12:27] *** xiphias has joined #arpnetworks [12:29] *** heavysixer has quit IRC (Quit: heavysixer) [12:32] *** xiphias has quit IRC (Ping timeout: 240 seconds) [12:33] *** xiphias has joined #arpnetworks [12:44] *** xiphias has quit IRC (Ping timeout: 240 seconds) [12:45] *** xiphias has joined #arpnetworks [12:50] *** xiphias has quit IRC (Ping timeout: 240 seconds) [12:51] *** xiphias has joined #arpnetworks [12:56] *** heavysixer has joined #arpnetworks [12:56] *** ChanServ sets mode: +o heavysixer [13:03] *** GluffiS_ has quit IRC (Ping timeout: 246 seconds) [13:03] *** GluffiS has joined #arpnetworks [13:07] *** heavysixer has quit IRC (Quit: heavysixer) [13:25] *** CaZe has quit IRC (Ping timeout: 260 seconds) [13:25] *** CaZe` has joined #arpnetworks [13:26] *** CaZe` is now known as CaZe [13:43] *** heavysixer has joined #arpnetworks [13:43] *** ChanServ sets mode: +o heavysixer [14:28] *** heavysixer has quit IRC (Quit: heavysixer) [14:38] *** robonerd is now known as adfadf [14:49] *** TheHiTCHO has joined #arpnetworks [14:49] *** m0unds__ has quit IRC (Remote host closed the connection) [15:22] *** TheHiTCHO has quit IRC () [15:23] *** adfadf is now known as robonerd [15:25] *** robonerd is now known as rnerd [15:32] ok I would never have expected this [15:32] ? [15:32] I moved a perl bot I am writing that connects to btc-e.com to an arpnetworks vps [15:33] from my laptop here at home [15:34] instantly it is faster and reacts so quickly it makes me $2.33 USD equivalent in btc on the spot [15:34] lol nice [15:35] thats just the test account, time to move the big account [15:36] hahaha, nice [15:38] Soon toddf's VPS will pay for itself [15:40] since 12/10 I'm only $3.31 shy of being able to pay my next months bill .. before I move it to arpnetworks, so .. it seems realistic [15:40] (my bill is $65/mo fwiw) [15:54] toddf well i helped start the #coindev "virtual currency" software / service dev community, and we host with arp [15:54] you're both welcome to join our channel, and at the right place! :) [16:00] toddf which OS? [16:01] openbsd of course [16:02] ah, fbsd here [16:02] * rnerd looks over at toddf, and gives him a nod [16:05] *** heavysixer has joined #arpnetworks [16:05] *** ChanServ sets mode: +o heavysixer [16:35] *** heavysixer has quit IRC (Quit: heavysixer) [16:43] Anyone familiar with how squid operates? [16:43] #squid seems quiet right now. I am getting no HITTs in my access log: only TCP_MISS/200 [16:43] Its set up as a reverse proxy [16:44] requests are going through, but not being cached I dont think [16:44] *** rnerd is now known as robonerd [16:45] A reverse proxy? Any reason not to use nginx/lighttpd/etc? [16:45] (varnish..) [16:46] brycec is squid truly irrelevant largely, now? [16:46] if so, an era has passed [16:46] * robonerd takes a moment of silence [16:47] I'd say it's overblown and overcomplicated for most reverse proxying [16:47] forward proxying it's still peachy [16:47] But Varnish is defintely the defacto caching reverse proxy [16:48] nice [16:51] *** heavysixer has joined #arpnetworks [16:51] *** ChanServ sets mode: +o heavysixer [17:02] brycec: can varnish be setup on say a front end web host with a separate apache server as the backend on a defferent box? [17:02] mnathani: Yes [17:02] That's what Wikipedia does, for instance [17:03] Varnish is awesome [17:03] Full disclaimer I've never used Varnish. Looked at it, but purely out of curiosity. [17:03] i wish I had the need for it though [17:03] ditty [17:03] *ditto [17:04] url to varnish? [17:04] toddf: whoa that's awesome [17:04] https://www.varnish-cache.org/ [17:04] up_the_irons :) [17:04] mnathani ty [17:06] yea varnish looks useful [17:06] pure C or? [17:07] *** heavysixer has quit IRC (Quit: heavysixer) [17:19] I just switched to varnish, but my pages are still not being cached. :-( [17:19] boo [17:20] I'd say it's time to check that the backend servers aren't sending do-not-cache headers [17:20] (tcpdump) [17:20] And of course check that varnish is configured correctly :p [17:21] Relevant https://www.varnish-cache.org/content/why-isnt-varnish-caching-requests-seem-pass-straight-through [17:21] Thanks [17:21] Apparently Cookies? [17:22] mnathani: easy test on the cookies: cURL [17:22] I swear curl is 90% of my web server debugging toolkit [17:23] curl ftw [17:23] libcurl too, if you're writing programs [17:27] how does libcurl help? [17:27] er, curl/libcurl [17:42] robonerd: rather than writing your own methods to open sockets etc, just use curl [17:43] can you get the raw binary somehow? [17:43] robonerd: you mean the raw response from the server? yeah of course. [17:44] the ppl in #osdev were speaking in the same way about dropbear last night [17:44] dropbear is an ssh server, not a client library [17:45] well yea [17:53] isn't dropbear gone? [17:53] i thought i'd heard it was KAPUT [17:54] no [19:10] i prefer trafficserver to squid or varnish [19:11] m0unds: i hope not! [19:11] i like for multiple implementations of important projects to stick around [19:11] i could have sworn it was dropbear that died [19:11] saw that it didn't, now i'm wondering what did die and i confused it with dropbear [19:12] dropbear used to have vulnerabilities [19:12] and people didn't upgrade versions [19:12] and then some people did but didn't upgrade the version number [19:12] was there a period where it wasn't maintained well? [19:12] or partially abandoned or anything like that? [19:12] really ssh servers shouldn't say the version number clearly. [19:12] no idea [19:12] i use dropbear as a client these days [19:12] it works on old hp servers (ilo2) where openssh doesn't [19:12] it connects faster over high latency links [19:12] ah [19:13] and it even connects faster over low latency links [19:13] ever tried mosh? [19:13] but it doesn't aggregate text as much [19:13] so it can be a little slower for some things, and show more constant information [19:13] like typing dmesg [19:13] in real access, latency seems better if anything [19:13] and it's pretty fast [19:13] and it's hundreds of msec faster if dropbear on both sides going international [19:14] dbclient -l root arp.meh.net.nz pwd 0.00s user 0.00s system 0% cpu 1.963 total [19:14] ssh -l root arp.meh.net.nz pwd 0.01s user 0.00s system 0% cpu 2.201 total [19:14] that's to openbsd openssh [19:14] which is 170msec away [19:14] weird ping is usually less than that [19:15] it's 1.45 seconds to a linux host at arp [19:15] with dropbear server [19:15] i still kind of wish the handshaking was quicker [19:16] oh mosh was hell slow for me [19:16] it was laggy with 10 msec ping [19:16] but one of the ends was using atom at the time [19:16] and it's very cpu hungry it seems [19:16] i thought mosh was a novel idea, but i don't like using hugely inefficient software [19:17] i typically only use it on my laptop because connectivity is so variable it saves me from having to reconnect regularly [19:17] never really noticed significant cpu utilization at either end [19:18] ahh [19:18] well atoms are pretty slow [19:18] yea [19:18] macbook is an i7 [19:18] it couldn't do samba at gigabit speeds [19:18] server i was working with was a dual xeon [19:18] ahh, bummer [19:18] xeon doesn't mean much :) [19:18] it's like saying a gas car :) [19:19] * m0unds is too lazy to go look [19:19] anyway, mosh still uses ssh to connect [19:19] mid 2012 [19:19] whatever [19:19] so recent [19:19] 192gb ram [19:19] etc [19:19] yeah [19:20] probably a e5 of sorts [19:20] e5-2690 or something maybe [19:20] wouldn't have dismissed cpu perf of a 10 year old xeon or something as being a reason for low cpu utilization... [19:20] herp derp [19:20] oh right [19:20] the old pentium4 type xeons were hell slow [19:20] err 5000 series [19:20] yep, netburst is a joke [19:20] bad arch [19:20] well about as bad as atom [19:20] good at warming up a room though [19:20] not really [19:21] *** r0ni has joined #arpnetworks [19:21] maybe a small room [19:21] 3.2GHz prescott was like a 130W cpu [19:21] tdp [19:21] or something absurd [19:21] *** r0ni has quit IRC (Client Quit) [19:22] they still don't use that much power normally [19:22] power supplies have got more efficient since then [19:22] and there's been some better idle, gating etc support [19:22] and cpus run closer to 0% load now [19:22] but in reasonable to high usage there's not much diff between old and newer cpus really [19:23] like 5000 series to 5400 series isn't huge jump [19:23] 5400 to 5500 is a bit [19:23] 5500 to 5600 is tiny i think [19:23] then a bit of a jump again from 5600 to whatever the next series was [19:23] is that e3 next? [19:23] it is isn't it [19:23] e3/35 [19:23] e3/e5 [19:24] and the affordable e5s aren't much faster than i5-2600s [19:24] err i7-2600 [19:24] and power/efficiency/performance has hardly changed since then again [19:24] apparently ivy bridge cpus are more efficient than haswell at load [19:25] i suppose amd kind of fell off the train, so intel doesn't have aynone to compete with except arm type things that are coming up [19:26] amd still does well in hpc [19:26] with video cards? [19:26] mostly because of thread count [19:26] haven't seen a lot of amd powered gpu compute [19:26] nvidia seems to own that space [19:26] you mean stuff that powerpc is winning in power/performance? [19:26] that's curious [19:26] consdiering that amd is much better than nvidia at bitcoin [19:27] i assumed that amd was better in general with opencl type things [19:27] yea, i think part of it is opencl vs cuda [19:27] cuda being more ubiquitous [19:27] which is silly [19:27] i haven't touched either [19:27] it sounded like a nice idea [19:27] i've been playing with ssd raid [19:28] adn reading up about stuff [19:28] i think cuda's been around longer, sorta like physx or whatever in teh gaming world [19:28] it seeems intel motherboards can't do more than about 1500mb/sec on onboard sata [19:28] so adoption is higher [19:28] cos they have a 2GB/sec link for pci-e [19:28] intel mobo w/the SATA controller attached at the southbridge or a separate controller like marvell or whatever it was they used? [19:28] x4 i suppose [19:28] intel onboard direct [19:29] i get about 1400mb/sec with x4 pci-e x2 [19:29] err pci-e 2 [19:29] i think slightly over 1400mb/sec [19:29] so i suppose it basically it is x4 limitations [19:29] at x8 i was getting faster speeds [19:30] but at those speeds it doesn't amtter too much [19:31] but i think i/o bus speeds for non video is likely to be the next significant improvement [19:31] video cards are iommu now [19:31] err can do iommu [19:31] it may not be that far away that there are dual cpus on small computers [19:31] with a lower power fanless cpu and a full cpu [19:32] that's something AMD's working on, actually [19:32] arm + x86 [19:32] oh, interesting [19:32] arm already do that kind of thing [19:32] http://www.amd.com/us/aboutamd/newsroom/Pages/presspage2012Oct29.aspx [19:32] first step is servers [19:33] that's damn cool. [19:33] time xterm -e pwd" isn't much faster than it was on a p120. [19:34] that + APU stuff (xbox, ps4, low-end laptops) are where they're really kinda pushing stuff a bit [19:34] there's lots of inbuilt latency etc limitations [19:34] and downclocking big cpus is probably not going to keep latency low [19:35] i haven't really considered amd seriously for years [19:35] last i knew their memory performance and single threaded performance was way down [19:35] yep [19:35] multithreaded performance is stellar [19:35] and most things don't scale to multicore [19:35] yep [19:35] which is what i wondered about back with pentium pros [19:36] not much scales on linux yet [19:36] lbzip2 is damn cool. [19:36] compiling scales well [19:36] but i think we're really close to needing things like multithreaded file copying [19:37] like if you copy 100,000 files [19:37] it shouldn't iterate through the files one by one [19:37] but on single hard-disks they don't really cope with multiple transactions at once well [19:37] but ssd's and raid arrays should be ok [19:37] having lots of threads for encoding (single encode process per core) is nice [19:37] so that's my personal thing i wnat to look at soon [19:38] for multiple files? [19:38] yep [19:38] yeah multipel files is easy [19:38] i don't do any encoding myself [19:38] i don't even do mp3 encoding from cd's [19:39] i do some audio work for friends, occasionally do video [19:39] but cpus are fast enough now that single cpu'ing that is still going to keep up with the cd [19:39] yea [19:39] i ripped heaps of music years back [19:39] 17 seconds to encode a 70 minute album from wav -> flac or wav, flac or whatever -> other format [19:39] it's pretty hilarious [19:39] heh [19:40] doing 8 tracks at once [19:40] it used to use up 50% cpu to play a mp3 [19:40] and i had to fuck around with nice lervels and shit [19:40] to make sure it didn't skip [19:40] i couldn't play mp3s on the machine i had til i was in HS [19:41] cpu was too slow, 486DX-33 [19:41] haha [19:41] haha [19:41] i went 486dlc40 to p75 i think [19:41] err [19:41] it was 486dlc40 to some amd thingy [19:41] like a fast 486 [19:41] with a weird name [19:41] then to a p120 [19:42] i went 486dx-33 4MB RAM -> 20MB RAM -> 300MHz AMD K6-2 [19:42] i overclocked the p120 to 133 mhz and removed the fan [19:42] so much headroom in those days [19:42] 4mb ram gotta suck [19:43] it was cutting edge in 1993 [19:43] double speed cdrom too [19:43] heh [19:43] shit 91 [19:43] mp3s were around in 91? [19:43] oh you said 93 [19:43] how did i read it as 91 [19:43] yeah, i had that machine til 1999 [19:43] hahaha [19:43] heh [19:43] i got amd k6-2 sometime too [19:43] with 128mb of ram [19:43] cdrom interface was via the isa soundcard [19:44] i had 24mb before that [19:44] and couldn't run X [19:44] i didn't have enough vram in the 486 to run x at higher than 640x480 [19:44] heh [19:44] did you overclock isa bus? [19:44] nah [19:44] i didn't have enough memory [19:44] i think i had s3 virge [19:44] i just used svgatext mode [19:44] my 486 was 1MB cirrus logic or something [19:45] i could get 800x600 in wfw 3.11 [19:45] and had like 100 columns or so [19:45] heh [19:45] i dunno how much memory i had [19:45] probably 2mb? [19:45] i think it was s3 trio64 [19:45] i have a pci version of one of those somewhere [19:45] s3 trio or verge maybe [19:46] lots of people had vergs [19:46] trios were cheaper [19:46] i went to radeon 9000 [19:46] which could do 1920x1440? [19:46] but on crt that flickered heaps [19:46] i usually just used 1280x1024 [19:46] yeah, probably low refresh or whatever [19:47] yeah, same [19:47] and then got annoyed with the lack of space sometimes and jumped up to 1600x1200 [19:47] yeah 1280x1024 was 85 hertz [19:47] and 1920x1440 was like 60 hertz [19:47] 1600x1200 75 hertz or something [19:47] 19" crt [19:47] but when youwent up resolution things got less crisp [19:48] but yeah everything uses up so much memory niow [19:48] PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND [19:48] 17037 root 20 0 333392 100136 80780 S 2.7 1.2 0:09.62 Xorg [19:48] i have 5 xterms open, and pidgin [19:49] but i was surprised how well linux seems to deal with 8gb of ram [19:50] compared to windopws [19:50] sometime memopry volumes went up enough that it doesn't really matter now [19:50] i've got 2GB free in win7 right now on this particular workstation [19:51] on 8gb ram? [19:51] yep [19:51] i have 5 gb free on 16gb right now [19:51] on the windows box [19:51] but i'm running chrome [19:51] probably superfetch caching stuff [19:51] and skype and putty [19:51] i disabled that stuff [19:51] why? [19:51] cos samsung told me to [19:51] lol [19:52] samsung magician software suggests you disable it [19:52] hahah [19:52] my corsair SSD has 3 years of use and still indicates 100% wear level with like 8TB of writes [19:52] i dunno i had 8gb for a while in it [19:52] cos i was curious how it'd go [19:52] (i default to thinking 16gb makse sense thse days) [19:52] i wonder if it's a write-saving measure for samsung [19:52] yea, my home ws is 16gb [19:52] this one's at my office [19:52] and i noticed the upgrade to 16 [19:53] it's cos i junmped from 6 to 16 and thought 8 may have been enough [19:53] but i'm in similar situation with my linux bvox too [19:53] i have ram lying around [19:53] i could upgrade to 16gb or 24gb on server [19:53] but it's on 8gb [19:53] and i run chrome on it too [19:53] and it deals much better on 8gb [19:53] and i dual boot the windows box with linux too [19:53] and i notice no diff between 8/16 [19:54] my original justification was for virtualisation though [19:54] and i was doing linux and windows on the same box for a bit [19:54] and upgrading from 16 to 24gb helped [19:55] samsung suggests disabling superfetch w/ssds as a memory saving measure [19:55] oh interesting [19:55] i was curious why they'd suggest it - iguess it makes sense in lower memory ssd envs like laptops or whatever [19:55] so i should reenable it [19:56] like if you've got a notebook with 4gb of ram, it might make a diff [19:56] the biggest difference i find with more ram is uncompressing large files yhou've juist copied on [19:56] but even then, i dunno [19:56] like say you have a 4gb archive [19:56] that you just copy on [19:56] it's nice if that 4gb stays in memory [19:57] but generally speaking it's still going to be slow [19:59] *** CaZe has quit IRC (Ping timeout: 246 seconds) [20:00] *** CaZe` has joined #arpnetworks [20:00] *** CaZe` is now known as CaZe [20:09] *** heavysixer has joined #arpnetworks [20:09] *** ChanServ sets mode: +o heavysixer [20:32] *** heavysixer has quit IRC (Quit: heavysixer) [20:53] *** TheHiTCHO has joined #arpnetworks [21:23] *** joepie91 has joined #arpnetworks [22:02] *** TheHiTCHO has quit IRC (Remote host closed the connection)