Back to Home / #linode / 2012 / 03 / Prev Day | Next Day
#linode IRC Logs for 2012-03-27

---Logopened Tue Mar 27 00:00:02 2012
---Daychanged Tue Mar 27 2012
00:00<Katana>don't you know about pi
00:00<jhanjon>There seem to be stack scripts that include mercury
00:02<linbot>XReaper: Point (0.31493567, 0.64317442) lies within the unit circle. Hits: 110514 of 140288 (π ≈ 3.151060675182482 - 0.009468021592689). http://π
00:02<linbot>XReaper: Point (0.01132795, 0.18678250) lies within the unit circle. Hits: 110515 of 140289 (π ≈ 3.151066726543065 - 0.009474072953271). http://π
00:07-!-peter [] has joined #linode
00:07-!-peter is now known as Guest7861
00:11-!-mcinerney [] has quit [Ping timeout: 480 seconds]
00:16-!-A-KO [as@2001:470:1f07:115d:e4bd:2ce3:41f2:c9c5] has quit [Remote host closed the connection]
00:21-!-joshdotsmith [] has joined #linode
00:24<Smark>I'm preparing for a project and am trying to work out how the backend is supposed to look. Everything is running on Linode but won't be using the Linode LB services... Which of the following three "plans" make the most sense as far as performance?
00:25<Smark>meaning they all make good sense, or are you being sarcastic saying it's a really hard question
00:26-!-ang [] has joined #linode
00:26<@akerl>Smark: All of those appear to be valid ways to do it, and without testing your exact stack there's no way for someone to tell you the "best" way
00:27<@akerl>I'd recommend NodeBalancers as being awesome, but I am biased
00:27<XReaper>don't know whether getting rid of apache and using php-fpm would increase performance...
00:28<Smark>The NodeBalancers are something we looked at but in the event that something happens we'd like to be as platform independant as possible.
00:28<@akerl>If I were doing it, I'd probably do #1, because it lets all servers at one level be identical
00:29-!-Kunda [] has quit [Quit: Kunda]
00:29<XReaper>true load-balancing
00:29<pharaun>just do nodebalancer, problem solved
00:29<@akerl>I'm willing to bet that the load of the static content will be tiny compared to the dynamic, but obviously if your static is hd videos or similar, things are different
00:30-!-Kunda [] has joined #linode
00:30<Smark>the static is likely to be a lot of images (such as avatars and the like)
00:30-!-sivy_ [] has joined #linode
00:31-!-ang [] has quit [Remote host closed the connection]
00:31<@akerl>Smark: I'd say your best bet is to build a config and benchmark it. If it works, cool. If it's not what you want, try another one
00:31<pharaun>honestly benchmarking it will almost always be the answer
00:32<pharaun>for custom setup/etc to pick things, etc, we are not all knowing... as much as some people might disagree :)
00:32<XReaper>don't ask me for advice, i run arch :P
00:32<pharaun>even on urmom?
00:33<Smark>haha thanks for your help everyone. This is my first time building something like this. Was wondering if there was a consensus or something that people had found out was GENERALLY better.
00:33<XReaper>1 looks good...
00:33-!-ang [] has joined #linode
00:33<XReaper>could even add more servers in later
00:33<Smark>well you would always add more servers later in all the configs. IE 20 app, 5 static or something
00:34<XReaper>yeah true
00:34<XReaper>just need to sync the static i'd guess
00:34<pharaun>that's common webscale arch
00:35<pharaun>load balancer, cachers, webapps, db
00:35<XReaper>oh like use the static as a CDN
00:35<pharaun>yeah that too
00:35<pharaun>often you see varnish or squid as the cachers
00:35<pharaun>but that depends on the kind of context you are generating
00:35<pharaun>for ex look at wikia
00:36<Smark>there a good way for any node to be able to serve any static file without duplicating all the data on each server?
00:36<Smark>such as special nodes and using NFS?
00:36<XReaper>only if it knows where to look
00:36<pharaun>XReaper: ?
00:36*XReaper makes no sense
00:36-!-sivy [] has quit [Ping timeout: 480 seconds]
00:36<pharaun>Smark: nfs or you could just rsync the static content in your deployment plan
00:36<pharaun>may i suggest puppet?
00:37<Smark>well the content changes rapidly, it's user uploaded content, not just a bunch of images we upload
00:37<Smark>much makes me think 2/3 might be the best
00:38<XReaper>another option is a CDN service... but lets not go there
00:39-!-joshdotsmith [] has quit [Quit: joshdotsmith]
00:39<pharaun>:D: :D:
00:39<linbot>The Romans used Arch. Worked out GREAT for their civilization.
00:39<pharaun>they don't exist
00:40<XReaper>wonder who made that
00:40<XReaper>people keep using it on me
00:40<XReaper>!to pharaun arch
00:40<linbot>pharaun: The Romans used Arch. Worked out GREAT for their civilization.
00:40<pharaun>i never installed arch
00:40<pharaun>not even once
00:42<XReaper>my linode my desktop my laptop my printserver
00:42<XReaper>all arch
00:43<linbot>The Romans used Arch. Worked out GREAT for their civilization.
00:43<XReaper>you HAD to do it :P
00:43<pharaun>debian, debian, debian, fedora, fedora, gentoo
00:43<XReaper>What's up with the rolling release model?
00:43<pharaun>oh openbsd, and openwrt too
00:43<XReaper>dd-wrt on my router
00:43-!-Kunda [] has quit [Quit: Kunda]
00:45*EugeneKay sticks with his Scientific Linux... FOR SCIENCE!
00:45-!-monodemono [] has quit [Quit: *POOF!*]
00:46<pharaun>then you won't mind me doing science on you with some supercharged cow prods?
00:46<EugeneKay>That's $50
00:46<Kyhwana>per prod
00:46-!-azaghal [~azaghal@] has quit [Remote host closed the connection]
00:46-!-azaghal [~azaghal@] has joined #linode
00:46*pharaun submits the bill to Kyhwana
00:47*Kyhwana zaps pharaun for each zap he's billed for
00:47<linbot>Scientific Linux: The official distribution of the Large Hadron Collider. It has yet to suck the planet into a black hole, so it must be working right!
00:47*Kyhwana submits that bill to EugeneKay
00:47-!-nmudgal [~nmudgal@] has joined #linode
00:49*XReaper loves this place
00:50*XReaper makes an infinite loop using the api to make more linodes
00:50-!-Integer [] has quit []
00:51<pharaun>there's an api limit
00:51<pharaun>5 iirc
00:54-!-rurufufuss [] has quit [Ping timeout: 480 seconds]
00:59<XReaper>ruin my fun :P
01:04-!-rurufufuss [] has joined #linode
01:04<pharaun>that's my job
01:17-!-jgornick [] has quit [Quit: jgornick]
01:17<Smark>question? How do you scale out with changing static content (such as new images)? Everything I come up with sounds hacked together. For instance, if I need more space, instead of going from 512 -> 1024, i want to go from 512 -> 512x2. How would one organize the data?
01:18<Kyh>Smark: use a load balancer to LB between the 2x512?
01:18<Smark>I can't see any way to do it cleanly without required a restructure every time you want to add a server
01:18<Smark>but then don't you need to duplicate 16gb between the two?
01:19<Smark>I'm looking to get 32gb of usable space
01:19<Smark>or 48, or 64, however high I need to go
01:19-!-jhanjon [] has left #linode []
01:19<@akerl>Smark: You don't store all content on all servers
01:20<Smark>exactly, so whats the best way to determine where you put the data? best i can gather is to md5 it and use the first digit(s) or just pick a random number and store the location in a DB record
01:20<Peng>Smark: One simple solution is "dump it all on Amazon S3".
01:21<Smark>Peng: I'm not 100% familiar with how that works, do you then serve a S3 URL, or do you mount it via NFS?
01:21<Peng>Smark: The former.
01:21<Peng>Smark: S3 doesn't support NFS. People have invented hacks to make it mountable, but I wouldn't rely on 'em.
01:22<Peng>Smark: Plus, performance is worse and it costs you more money when you have to stream the data through your servers.
01:22<@akerl>Smark: If I were doing it, and legitimately needed to split things up, I would likely start by dividing categories of static files onto different servers
01:23<@akerl>So if I'm storing avatars and some other uploaded media, they are on separate servers
01:24<Smark>what happens when you get too much of a certain category? upgrade the server? It's there a more "cloud"-y way?
01:24<@akerl>It depends
01:25-!-seanh-ansca [] has quit [Quit: Leaving.]
01:25<pharaun>eh i would just do a couple deep md5 sums or something
01:25<pharaun>and split those directories over, it should generally spread the load out
01:26<pharaun>squid does something like that for its on disk cache
01:26<Smark>OPTIMALLY if I have 20GB of data and 4 machines, I'd like 5GB on each. Seems like the best way, but in order to add a server you have to adjust your "system" quite a bit, atleast to have it spread evenly.
01:26<@akerl>Smark: Potentially relevant:
01:27<Peng>Smark: You could run some sort of distributed s.
01:28<Smark>so for instance, you have each file assigned a long number, 0/1/2 go on 1, 3/4/5 go on 2, 6/7/8 on 3, 9//0 on 4
01:28<Smark>i'll take a look akerl
01:28<Smark>whats an example of a distributed fs?
01:28<@akerl>If you want to get really fancy, you could hook that in with DNS for scaling
01:29-!-mdcollins [] has joined #linode
01:29<Peng>Smark: GlusterFS? Ceph (still experimental)?
01:29<Smark>the problem with the numbering method is that if you need to scale to five servers you have to move files around. Not extactly plug and play
01:29<Peng>Smark: Um...there are a few more.
01:29<Peng>Smark: I mean, there are tons, but there are only a few more I think are interesting. :P
01:29<Peng>Smark: MogileFS
01:30<Peng>I'm forgetting one.
01:30<Peng>Smark: Tahoe
01:30-!-EriksLV [~EriksLV@] has joined #linode
01:30<Peng>Smark: Again, or S3.
01:30<@akerl>the first digit in your unique ID is the subdomain the request to, so to start with {0..9} all point to the same IP. When you scale up, you clone half the content to server B and point {5..9} to that server
01:30<Peng>Smark: S3 storage is way cheaper than Linode, too.
01:31<Peng>"clone half the content" -- an easy and fun task.
01:31<@akerl>Peng: Depending on how you structure it, it doesn't have to suck
01:31<bob2>ha ha mongo
01:32<Smark>ah that makes a lot of sense akerl
01:32-!-JDLSpeedy [] has quit [Ping timeout: 480 seconds]
01:32<Peng>Open Stack Swift, of course.
01:33<Peng>But I'm not sure if it runs on anything less than a rackful of servers.
01:33<Smark>I may end up going with S3 but I'd definitely like to look more into a self-hosted options
01:33<bd_>Peng: s3's a hell of a lot more durable than raid10 too
01:34<Smark>out of curiousity are there DNS tricks to masking the fact that you use S3 (atleast on a URL level)?
01:34<Peng>Smark: Yes.
01:34<Peng>Smark: Unless you want SSL.
01:34<Smark>not important at this point
01:34<bd_>you can CNAME from <> to <>
01:34<Peng>Smark: But lots of people use S3. It's nothing to be embarrassed about.
01:34<bd_>you can CNAME from <> to <>
01:34<pharaun>i see all the time
01:34<pharaun>so i don't see the issue
01:35<Smark>no not saying that at all Peng, i guess from an organization standpoint or something
01:36<Peng>Smark: Ah, sure. (That sentence could come off as sarcastic. It was not.)
01:36<Smark>no need for loadbalancing or anything at all either since thats all handled by amazon. I just point it to my webspace (bucket as it is called) on S3.
01:36<pharaun>pretty much
01:36<Smark>(as far as static content)
01:37<Peng>Smark: Incidentally, one thing about distributed FSes is that they like to store several copies of the data for reliability -- the more servers you have, the more likely one of 'em will be down, and you don't want 1/$servers of your data to be inaccessible.
01:37<pharaun>Smark: most "webscale" folks i know just do static -> s3/cdn, dynamic locally
01:37<Peng>Smark: This increases your cost like 3 times.
01:37<Peng>Smark: If you use S3, it handles all of this for you, and is still relatively cheap.
01:38-!-danblack [] has joined #linode
01:39<Smark>I expect I'll go with S3/CDN for static content (atleast the large storage requirements). Out of curiousity how do CDNs/S3 keep track of stuff then? Stick it on a random server and store the filename/location/etc in a database?
01:40<Peng>Smark: I believe Amazon's official position is that S3 is implemented with magic.
01:40<pharaun>as far as you care, its magic
01:40<pharaun>you stick files on it, done
01:40<bd_>Peng: apparently some more details were released recently!
01:40<pharaun>and they do magic
01:40<Peng>bd_: Yeah, I read that.
01:40<Peng>Smark: See bd_'s link for some technical details about how S3 works.
01:40<bd_>none of that matters at all unless you're hitting a few hundred or more tps of course
01:41<Smark>oh no, was pure curiousity.
01:41<Peng>Smark: OpenStack Swift is an FOSS competitor to S3. It's probably not as well-optimized yet, but its documentation -- and source code -- would probably be worth reading if you're interested in the field.
01:41<Smark>Thanks Peng! I'll look into it.
01:42<Peng>I wonder what Wikimedia runs for their static file storage.
01:43<pharaun>probably cdn
01:43<Smark>They release ALL their configuration, so you can definitely find out
01:43<Peng> 398 IN CNAME
01:43<Peng>pharaun: Not so much
01:43<pharaun>including their password? :3
01:43<pharaun>Peng: mmh figures :p
01:43<Smark>hmm so S3 basically does the "number method" above but has a much more dynamic method os splitting
01:44-!-JDLSpeedy [] has joined #linode
01:45<Smark>some of it is there. They have a special page which has it all
01:46<Peng>Smark: Also, if you're *really* interested in S3, Amazon is hirining.
01:47<Smark>all there
01:47<bd_>Peng: every time there is any sort of internal presentation, it always ends with "... and we're hiring! <internal jobs site link>"
01:47<bd_>EVERY TIME
01:47<pharaun>without fail
01:47<pharaun>ive been to some pycons and they're like
01:48<pharaun>at the end there's a large list of people who rattle off their place of work and then say, and we're hiring!
01:49<bd_>pharaun: yeah but the thing is, even in the internal employee-only presentations, they try to poach - I mean, invite people from other teams to apply.
01:49<pharaun>pretty much
01:49<Smark>ok, so when serving static files between two mirrors, how do you keep them mirrored effectively? rsync? what about the times between when rsync runs? How do you know which data store is more up to date?
01:50<bd_>Smark: if you're using rsync synchronization, you'll want to designate a master
01:50<bd_>or use unison I guess, but it's much simpler if you have a master
01:50<Smark>then when do the slave(s) get hit?
01:51<bd_>You sync from master to slave. Then put DNS round-robin or a LB in front of the slaves (and the master too if you have a small fleet I guess)
01:51<Smark>how do you prevent a file from being requested on a slave which doesn't have it yet?
01:51<bd_>oh, this is dynamically added data, then?
01:52<Smark>yes, sorry
01:52<Peng>Reading my last link, Apparently Wikimedia just sticks images on really big Sun servers with 48 hard drives, Solaris and ZFS.
01:52<Peng>If the link is still current.
01:52<bd_>probably better to have a bunch of caches in front of your origin server(s) then, and hit the origin server(s) on a cache miss
01:52<bd_>particularly if you're only adding/removing files, never overwriting
01:53<Peng>Ah. Other link:
01:54<Smark>even then how do you know which origin server to hit? Or do you have one origin server and several caches?
01:54<Peng>bd_: Yeah, Wikimedia seems to use that strategy. Lots of Squid servers.
01:55<Peng>They looove the Squid.
01:55<bd_>Smark: if you have multiple origin servers then it gets complicated :)
01:56<pharaun>Peng: same with wikia too, except its varnish
01:56<Peng>Hmm, another page says that server is their *former* image server.
01:56<Smark>i can live with one origin and multiple cache servers, if it gets too big then you just move to S3. But multiple cache servers with one origin would be the correct way, yes?
01:56<bd_>that's one way I guess
01:56<bd_>but then what happens when your origin dies?
01:57<pharaun>you cry?
01:57<bd_>pretty much
01:57<pharaun>suppose you could nodebalance the origin
01:57<Smark>till rsync to a backup/slave and eat the dataloss since the last sync?
01:57<bd_>pharaun: yeah but keeping the origins in sync is the hard part
01:57<Peng>Ah, it moved from ms1 to ms7.
01:58<pharaun>bd_: of course
01:58<Peng>which is another, slightly bigger Sun
01:58<bd_>Smark: you could also just go and see if S3 is cheap enough for your needs - they take care all of this hard stuff for you ;)
01:59<Peng>"We have zfs replication of our image data set up to ms8; it is running every 15 minuntes."
02:01<Smark>Alright everyone, I've taken notes. Thanks for your continued help! Goodnight.
02:01<Peng>ms1, ms7 and ms8 being names of servers, of course
02:04<pharaun>shame they didn't name it urmom1 urmom7 urmom8
02:05<Peng>Hmm.... universal reliable media object manager
02:05<pharaun>see how it rolls of the tongue
02:06<Peng> <- Dear god look at the regex
02:07<rnowak>I lol'd
02:09<Peng>So...what does run, I wonder.
02:09<bd_>looks like akamai
02:09<Peng>I've seen articles about their architecture, but I'm not going to dig one up right now.
02:09<Peng>bd_: Oh, you're right.
02:09-!-psil0 [] has quit [Quit: Leaving]
02:11-!-TimTim [] has quit [Read error: Operation timed out]
02:11<Peng>Speaking of long regexes, I wrote a Python one to parse RFC 3986 URI-references.
02:12<rnowak>what do I get if I break it?
02:12<Peng>A warm, fuzzy feeling?
02:13<Peng>The IPv6 address part of the regex alone is 1.5 KB :)
02:15<Peng>rnowak: What did you mean by "break it"?
02:16-!-igufi [] has quit [Quit: leaving]
02:16<pharaun>Peng: he means, take it out back and do unimaginable things to it so that its mind is broken
02:17<Peng>pharaun: If anything that would probably make it *more* sane.
02:17<pharaun>one does not simply make that regex sane
02:18<pharaun>tho its not that bad compared to the good ol' email one
02:18-!-igufi [] has joined #linode
02:18<pharaun>v4 aint bad its the v6 that is a real joy
02:18<Peng>What, in email or URIs?
02:20<rnowak>ha ha the perl email regex, it is cute
02:20<pharaun>yup ^
02:20<danblack>so i saw the other day even email can have an '@' in the local part. its not that simple either if you follow rfc5831
02:20<rnowak> Peng
02:20<pharaun>This regular expression will only validate addresses that have had any comments stripped and replaced with whitespace (this is done by the module).
02:21<pharaun>does that regex even handle unicode email addresses?
02:22<pharaun>or punny code domain names too? :p
02:22<danblack>even the perl regex probably doesn't exclude \127 characters the same way as \000-\031
02:22<rnowak>son, you'll keep unicode out of my emails
02:22<hawk>unicode all of the things
02:22<danblack>punny code was i think compatible with rfc5831
02:23<rnowak>fancypants unicode, and hipster ipv6
02:23<rnowak>pharaun: I don't even
02:23<pharaun>dont ask
02:30-!-message144 [] has quit [Quit: gone]
02:35<Peng>Stupid Firefox dark image background
02:46-!-jarr0dsz [] has joined #linode
02:52-!-mpkossen [~mpkossen@] has joined #linode
02:57-!-EriksLV [~EriksLV@] has quit [Remote host closed the connection]
03:01-!-hipsterslapfight [] has joined #linode
03:03-!-Zr40 [~zr40@] has joined #linode
03:22-!-EriksLV [~EriksLV@] has joined #linode
03:32-!-hipsterslapfight [] has quit []
03:34-!-vodka [] has quit [Quit: Computer has gone to sleep.]
03:44-!-Kunda [] has joined #linode
03:44-!-amoe [] has joined #linode
03:46-!-amoe_ [] has quit [Ping timeout: 480 seconds]
03:48-!-igufi [] has quit [Quit: leaving]
03:54-!-mpkossen [~mpkossen@] has quit [Quit: Leaving.]
03:54-!-mpkossen [~mpkossen@] has joined #linode
03:55-!-MissionCritical is now known as Guest7882
03:56-!-MissionCritical [] has joined #linode
04:02-!-Netsplit <-> quits: numk, HalJordan, Peng, xach_, Torenn, Frools, jspiros, RichGuk, jonsowman, Kabaka, (+214 more, use /NETSPLIT to show all of them)
04:02-!-Netsplit <-> quits: SirFunk, dr_jkl, Dianoga, unforgiven512, encode, brainproxy_, @caker, virtualroot, jkwood, @akerl, (+33 more, use /NETSPLIT to show all of them)
04:04<dwfreed>amitz: you killed it O.o
04:04-!-teacup [] has joined #linode
04:04-!-Netsplit over, joins: amoe, mpkossen, Kunda, EriksLV, Zr40, jarr0dsz, JDLSpeedy, mdcollins, ang, Guest7861
04:04-!-Netsplit over, joins: VS_ChanLog, fayimora, darkbeholder, gflanders, squircle, Linear, derek, Craighton, Boohemian, joshiee (+4 more)
04:04-!-Netsplit over, joins: brambles, SNy, CornishPasty, Chowzzf_, x4, aaronyy, pjkh, BarkerJr, niftylettuce, wgas (+15 more)
04:04-!-Netsplit over, joins: Bry8Star-{GB, CaptObviousman, tomaw, rmayorga, ivan\, jeremyb, milk, internat, jed, straterra (+108 more)
04:04-!-Netsplit over, joins: Myon, mac-mini, jotamjr, zibri, Vadtec, JStoker, jetlagmk2, jtsage, rnowak, marius (+46 more)
04:06-!-hipsterslapfight [] has joined #linode
04:06-!-mode/#linode [+v Perihelion] by ChanServ
04:06-!-mode/#linode [+v array] by ChanServ
04:06<joshiee>Perihelion what did you do?!
04:06<Peng>She forgot to feed the hamsters running the generator.
04:06-!-joshiee [lol@2001:470:b:235:1337:1337:1337:1337] has quit [Remote host closed the connection]
04:06-!-joshiee [lol@2001:470:b:235:1337:1337:1337:1337] has joined #linode
04:06-!-mode/#linode [+v heckman] by ChanServ
04:06-!-Netsplit over, joins: tharkun, dr_jkl
04:06-!-Netsplit over, joins: akerl, dopey
04:06-!-Netsplit over, joins: @tparker, @brianok, bd_, brainproxy_, virtualroot, VladGh, Aexoden, Raynes, PrgmrBill, sneakums (+3 more)
04:06-!-ServerMode/#linode [+ovoo akerl akerl tparker brianok] by
04:06-!-Netsplit over, joins: d1b, Musfuut, pixl, Hobbsee, Dianoga, jpg
04:06-!-Netsplit over, joins: caker, heidi, danieldg, crux_, sirpengi, SamT, SirFunk, phliver
04:06-!-Netsplit over, joins: @tasaro
04:06-!-ServerMode/#linode [+ovov caker caker tasaro tasaro] by
04:06-!-Netsplit over, joins: jkwood, ido, CrackMonkey
04:06-!-Netsplit over, joins: @Praefectus, rmyers, linbot, syntaxman, lberk, pygmalio1, encode, ajmitch
04:06-!-mode/#linode [-o tparker] by ChanServ
04:06-!-Praefectus is "zeus" on @+#linode #moocows +#Corsair #linode-staff #linode-notifications
04:06-!-mode/#linode [-o brianok] by ChanServ
04:09-!-teacup [] has quit [Remote host closed the connection]
04:10<joshiee>i just noticed the features page still has outdated images
04:14<Zr40>it also has a fancy but slightly annoying image animation
04:15<Zr40>when you click any of the images
04:16-!-psx [] has joined #linode
04:16-!-psx [] has quit []
04:16-!-psx [] has joined #linode
04:16<Peng>Ah, that.
04:17<joshiee>pretty fast for me
04:17<marius>I rmember that manager :o
04:18<Zr40>sure it's not slow, but... it first resizes vertically, then resizes horizontally and fades background, and only then fades in the image
04:18-!-EriksLV [~EriksLV@] has quit [Remote host closed the connection]
04:18<Zr40>it could do all that animation at the same time :)
04:20-!-teacup [] has joined #linode
04:22-!-monodemono [] has joined #linode
04:24-!-teacup [] has left #linode []
04:28-!-vodka [] has joined #linode
04:33-!-thews [] has quit [Quit: thews]
04:48-!-mdcollins [] has quit [Ping timeout: 480 seconds]
04:50-!-straterra [~straterra@2001:470:8a81::4] has quit [Read error: Connection reset by peer]
04:51-!-EriksLV [~EriksLV@] has joined #linode
04:53-!-Bartzy [] has joined #linode
04:56-!-straterra [] has joined #linode
05:01-!-Bartzy [] has quit [Ping timeout: 480 seconds]
05:05-!-michael_mbp [~michael_m@] has joined #linode
05:07-!-flashingpumpkin [] has joined #linode
05:13<Peng>Necroing an earlier topic, the article about Tumblr I saw recently doesn't mention their static file setup.
05:14<sirpengi>does tumblr serve static files?
05:14<sirpengi>I think they put all that stuff on s3
05:14<sirpengi>at least, the pictures that I post on my tumblr all resolve to s3 urls
05:15<Peng>sirpengi: Oh? I thought it was Akamai.
05:16<CornishPasty>S3, Peng
05:16<sirpengi>I dunno, the one on my own tumblr go to s3
05:16<sirpengi>well, it seems they're defined with a tumblr endpoint, but get redirected
05:16*Peng doesn't use Tumblr.
05:17<sirpengi>in any case, that answers the question about their static file setup
05:18<Peng>Ah. The 500px 'thumbnails' on the, uh, stream pages are Akamai.
05:18<Peng>The full-size images are S3. Apparently
05:19<Peng> = Akamai
05:19<Peng>It could be fronting S3, of course.
05:21<CornishPasty>Seems they use edge cast for the stream pages, Peng ?
05:21-!-dehed [] has joined #linode
05:22<dehed>Hi, is there anyone I can ask questions about linode programs?
05:23<linbot>If you have a question, feel free to just ask it -- someone's always willing to help. If you don't get a response right away, be patient!
05:23<praetorian>not sure what you mean, but ^
05:24<Peng>CornishPasty: Perhaps it depends on the popularity of the images.
05:24<dehed>OK, I mainly was wondering if there is support transferring the existing website to linode, and also what is the best option for me if I have one big main website, and a lot of other smaller websites to host
05:24<CornishPasty>That's true, seems like they use edgecast for smaller images, and S3 for the larger ones?
05:24<CornishPasty>dehed: git clone and restore db from backups?
05:25<Peng>CornishPasty: I see: Small image on Akamai -
05:25<Peng>CornishPasty: Large image links to
05:25<dehed>CornishPasty: I'm not sure I understand what you mean... I'm quite a noob...
05:25<Peng>CornishPasty: Which ultimately redirects to S3
05:26<Peng>CornishPasty: (With a really long URL)
05:26<CornishPasty>Hmm yeah
05:26<@akerl>dehed: Linodes are self-managed, so you'd be responsible for setting up your server and such
05:26<Kyh>dehed: linode's are basically selfmanaged linux boxes
05:26<Peng>CornishPasty: It's entirely possible they mix CDNs. I notice your image is SSL, too.
05:26<CornishPasty>Yeah, probably to save on costs depending on which is best for that particular image?
05:27<CornishPasty>i.e. akamai may have low transit costs, but high storage costs, whereas others may have low storage but higher transit..
05:27<dehed>one for the main website and one for all the others?)
05:28<dehed>Oh, my message got cut off
05:28<Santo>erhm nvm
05:28<praetorian>also akamai might be better in certain parts of the world vs others
05:28<praetorian>eg akamai is in .au but nothing else is
05:28<CornishPasty>Hm, true
05:28-!-igufi [] has joined #linode
05:29<CornishPasty>dehed: it depends on the traffic of the main site, surely?
05:30-!-TimTim [] has joined #linode
05:30-!-TimTim [] has quit []
05:33-!-dehed [] has quit [Remote host closed the connection]
05:37<@akerl>praetorian: I still want an Arch flag :)
05:40<chesty>praetorian: if I buy you an arch lanyard, will you hang yourself from it?
05:40<praetorian>chesty: usually not
05:40<praetorian>i usually hang small children
05:41<chesty>akerl: praetorian says mean things to me all the time, I'm just paying one back
05:41<@akerl>chesty: I have no idea what you're talking about. everyone here is so nice
05:42<praetorian>akerl: obviously you dont include SpaceHobo in those figures
05:42<chesty>SpaceHobo rocks
05:42<chesty>GET UP
05:42<praetorian>chesty send me your address. im going to post you an arch linux lanyard when it arrives
05:43<chesty>like a sex machine
05:43<praetorian>SpaceHobo: Don't stop, never give up. hold your head high and reach the top
05:43<dwfreed>chesty: I prefer
05:45<chesty>dwfreed: yeah, I noticed you used that range
05:47<chesty>SpaceHobo: I saw a crystal meth head, sores all over his body, trying to buy a burner at the local supermarket without ID. was that you?
05:48-!-Bartzy [~bar@] has joined #linode
05:48-!-danblack [] has quit [Ping timeout: 480 seconds]
05:49<chesty>SpaceHobo: nah, it was a $2 SIM card, but burner sounded better
05:50<chesty>yes, in au and usa, I assumed every country
05:50<chesty>the drug dealers must love it
05:51<Peng>You need ID to buy a SIM in US? Good to know.
05:51<chesty>praetorian: my info comes from the wire
05:51<chesty>Peng: ^
05:51<Peng>chesty: Ha
05:56<chesty>actually, I'm not sure if ID was required now in the wire, they used to drive all over the state buying burners every 2 weeks though
05:56<linbot>New news from forums: Your IP Addresses Allocation Stinks! in Feature Request/Bug Report <>
05:56<Kyh>lol what
05:58-!-danblack [] has joined #linode
06:00<amitz>id for sim card? heh, here in ind....
06:00<amitz>SpaceHobo: nah, i'm referring to a long running joke i'm pulling on an ex-#linode-r.
06:01<Peng>"can't get none" :X
06:01<amitz>ask chesty, he provoked me to it.
06:03<chesty>there's this arse hat that asked amitz what country he was from, then answered himself with, oh i remember, india. so we just went with it and amitz now lives in india
06:04<amitz>and i never lie, not a single time, while keep misleading him as if i'm from india :-p
06:04<Peng>Next year amitz moves to Indiana?
06:06<amitz>Peng: you should see the log. at one time, i explicitly said i'm not indian, but with misleading context hence he missed the hint :))
06:06<amitz>it was entertaining.
06:08<amitz>it's a bit boring now, but it's probably a temporary thing.
06:12-!-Josh-_ [lol@2001:470:b:235:1337:1337:1337:1337] has joined #linode
06:15-!-joshiee [lol@2001:470:b:235:1337:1337:1337:1337] has quit [Ping timeout: 480 seconds]
06:16-!-wkl [~wkl@] has quit [Quit: wkl]
06:18-!-Bartzy [~bar@] has quit [Remote host closed the connection]
06:20-!-Josh-_ [lol@2001:470:b:235:1337:1337:1337:1337] has quit [Remote host closed the connection]
06:21-!-Santo [] has quit [Quit: Leaving]
06:21-!-joshiee [] has joined #linode
06:31<igufi>What would be a good ssh client for windows if I would like to use something else than putty?
06:32<joshiee>i use both
06:32<@akerl>Virtualbox :)
06:32<Kyh>igufi: kitty
06:33<igufi>Kyh: is there any difference to putty?
06:33<joshiee>is it sad that virtualbox runs better than openvz?
06:33<Kyh>tharkun: It's a fork and has some improvements
06:34<igufi>Zoc looks impressive but with a 80USD price tag :-/
06:34-!-fayimora_ [~fayimora@] has joined #linode
06:35-!-fayimora_ [~fayimora@] has quit [Remote host closed the connection]
06:35-!-fayimora [~fayimora@] has quit [Read error: Connection reset by peer]
06:35-!-sebnash [~sebnash@] has joined #linode
06:37-!-fayimora [~fayimora@] has joined #linode
06:41<CornishPasty>Woah, $80 for something that's built-in to OS X?
06:42<@akerl>What's built into osx?
06:42<joshiee>it has a few lazy features
06:42<Kyh>ssh client
06:42<joshiee>that the standard terminal does not
06:42<Kyh>such as?
06:42<joshiee>reading the features page might tell you
06:42<igufi>another thing that would be nice to have on the client is support for OpenPGP card for authentication
06:43<joshiee>i only got zoc from my job so yeah
06:43<Kyh>"openpgp card"?
06:43<Zr40>most of those features seem kind of gimmicky
06:43<CornishPasty>I wonder whether it uses PuTTY on Windows?
06:44<joshiee>useful things like that
06:44<joshiee>and, the tabs
06:44<Peng>I hope you can't stick "../../../../../../etc/passwd" in that URL...
06:45<Kyh>Lets find out!
06:45<CornishPasty>try it peng
06:45<Zr40>joshiee: OSX terminal has had tabs for as long as I can remember
06:45<CornishPasty>Zr40: indeed
06:45<praetorian>seemingly not peng
06:45<joshiee>i'm not using osx
06:45<CornishPasty>Does it support zsh, joshiee?
06:45<CornishPasty>(on the local machine)
06:46<joshiee>it should
06:46<praetorian>whats so cool about that image
06:46<CornishPasty>I don't know praetorian, but it costs $80US!
06:46<joshiee>thumbnailed tabs
06:46<Zr40>praetorian: probably the same as what's so cool about firefox's tabbed tabs tab thingy
06:46<praetorian>i must admit, the ssh client i use at work, is SecureCRT
06:47<CornishPasty>I use opens at work
06:47<joshiee>and it has its own scripting language/tool
06:47<praetorian>when i spend my days ssh'd into 10's of systems .. putty got annoying
06:47<praetorian>i must admit, its not great for workhouse stuff
06:47<praetorian>but zoc looks like crap
06:48<CornishPasty>Get a real OS then, praetorian :P
06:48<praetorian>CornishPasty: if only i could use my mac.
06:48<praetorian>iterm2 ftv.
06:48<CornishPasty>Hmm, I can use AppleScript with What's this about a custom scripting language, joshiee?
06:49<praetorian>tell "finder" to "stfu"
06:49<igufi>Well, secureCRT seems to support smartcard authentication but it's even more expensive at 99USD :)
06:49<Zr40>if you require smartcard auth, $99 isn't expensive any longer
06:50<igufi>it's free with putty via patch but yeah .. as I'm trying to find an alternative to putty
06:50<praetorian>find me something that does putty decently in tabs
06:50<praetorian>i could not find anything decent
06:50<igufi>there is something called superputty that is "putty with tabs"
06:51<Zr40>praetorian: any unix-based system with a GUI
06:51<praetorian>Zr40: thank you . back to sleep.
06:51<praetorian>igufi: he has added so many features his git repo is empty.
06:52<igufi>try this
06:52<igufi>I pulled the wrong link from my browser's history..
06:52<praetorian>looks ok
06:52<praetorian>but it steals alot of chrome
06:53<linbot> |
06:53-!-Ghost [~Ghost@] has joined #linode
06:53<praetorian>wonder how customisable that is
06:53<praetorian>thanks hawk :p
06:53<hawk>praetorian: np
06:53<linbot>setup is not a verb. Please see
06:53<linbot>Maybe not a verb. But you can do it here:
06:53<@mikegrb>mmm bacon
06:53<linbot>Bacon is what makes food good!
06:53<@mikegrb>mmm bacon
06:53<Zr40>yay bacon
06:54<Kyh>yay login is a verg!
06:54<linbot>login is not a verb -
06:54<Zr40>!botsnack :)
06:54<Kyh>bah! slow linbot
06:54<linbot>Thanks, Zr40! Om nom nom
06:54<retro|blah>How do I lodged on?
06:54<linbot>Backup is not a verb. Please see
06:54<Zr40>retro|blah: use /lodgeon
06:54<joshiee>!rm -rf
06:54-!-fayimora_ [~fayimora@] has joined #linode
06:55<linbot>derpy deeds, done derp cheap
06:55<praetorian>igufi: superputty looks like it is putty connection manager
06:57<Peng>"MySQL it difficult to scale out, so having an ability to use some scalable database like mongodb would be a good option." :D
06:58<CornishPasty>Peng: wat?
06:58<dwfreed>um, tumblr uses MySQL, scales just fine for them
06:58<Peng>CornishPasty: It's a quote.
06:58<Peng>I saw it on a website. :)
06:59<joshiee>nah lets use some good ole windows sql
06:59<joshiee>hahaha not..
07:01-!-fayimora [~fayimora@] has quit [Ping timeout: 480 seconds]
07:01-!-fayimora_ is now known as fayimora
07:03<Zr40>woah, nice. irssi survived a sleep -> wake -> wifi reconnect :)
07:04<CornishPasty>nice Zr40
07:09-!-Ricki [~Ricki@] has joined #linode
07:11<Zr40>depends on the implementation though
07:11<Zr40>the TCP implementation could just decide to reset the connection because the link was gone
07:15<Peng>If I wrote a TCP implementation, it would kill -9 everything with an open connection when the link goes down, just to be safe :)
07:22<dwfreed>vodka: you don't want to send a signal that'll produce a core dump, because then the programmer will be able to determine why the signal was sent
07:22<vodka>you mean he won't be able to ~? :)
07:22<dwfreed>and you don't want the program to be able to ignore it, so -9 is the only option
07:23<vodka>I have spotted some lovely botnet processes in the past that gleefully ignore -9
07:24<dwfreed>They can't, unless they're in D-state
07:24<dwfreed>"The signals SIGKILL and SIGSTOP cannot be caught, blocked, or ignored."
07:24<Peng>Send syscall 0f05 to be extra safe.
07:25<dwfreed>Peng: do I want to know which one that is?
07:25<Peng>dwfreed: It's the "crash KVM guests" one.
07:25<Peng>Someone linked it here earlier :)
07:26<Peng>s/0f05/opcode 0f05/
07:30<dwfreed>I think I saw it earlier, just never read the bug report
07:33-!-fayimora [~fayimora@] has quit [Quit: fayimora]
07:37-!-michael_mbp [~michael_m@] has quit [Ping timeout: 480 seconds]
07:41-!-Linear [] has quit [Quit: Linear]
07:45-!-mpkossen1 [~mpkossen@] has joined #linode
07:45-!-mpkossen [~mpkossen@] has quit [Read error: Connection reset by peer]
07:49-!-fayimora [] has joined #linode
07:52-!-skcin7 [] has joined #linode
08:00-!-Consdale [] has joined #linode
08:01-!-danblack [] has quit [Quit: solitude begins]
08:04-!-monodemono [] has quit [Quit: *POOF!*]
08:04-!-igufi is now known as Guest7908
08:04-!-igufi [] has joined #linode
08:04-!-Zr40 [~zr40@] has quit [Quit: leaving]
08:05-!-A-KO [as@2001:470:1f07:115d:c569:bdf2:fdd7:5e7c] has joined #linode
08:14-!-pascal [] has joined #linode
08:14-!-jarr0dsz [] has quit [Read error: Connection reset by peer]
08:16-!-fayimora [] has quit [Quit: fayimora]
08:17-!-igufi [] has quit [Remote host closed the connection]
08:18-!-Guest7908 [] has quit [Remote host closed the connection]
08:20-!-michael_mbp [~michael_m@] has joined #linode
08:23-!-Ricki [~Ricki@] has quit [Ping timeout: 480 seconds]
08:26-!-Ricki [~Ricki@] has joined #linode
08:31-!-laser [] has joined #linode
08:31-!-laser is now known as laser`
08:31-!-laser is "(unknown)" on (unknown)
08:35-!-Xuefer [] has joined #linode
08:36-!-swayz [~swayz@] has joined #linode
08:36-!-Xuefer [] has quit []
08:42-!-nmudgal [~nmudgal@] has quit [Quit: Sleeping :-) ]
08:50-!-fayimora [] has joined #linode
08:51-!-Kabaka [] has quit [Quit: BRB]
08:54-!-Kabaka [] has joined #linode
08:57-!-Ghost_ [~Ghost@] has joined #linode
08:57-!-Ghost [~Ghost@] has quit [Read error: Connection reset by peer]
09:12-!-niemeyer [] has joined #linode
09:13-!-BarkerJr [BarkerJr@2002:1880:41b:1:c879:6573:7dac:a0c6] has quit [Read error: Connection reset by peer]
09:13-!-BarkerJr [BarkerJr@2002:1880:41b:1:c879:6573:7dac:a0c6] has joined #linode
09:13-!-nmudgal [~nmudgal@] has joined #linode
09:15-!-thews [] has joined #linode
09:18-!-vraa [] has quit [Ping timeout: 480 seconds]
09:26-!-laser` [] has quit [Remote host closed the connection]
09:32-!-vynsynt [~Adium@] has joined #linode
09:34-!-BarkariII [BarkerJr@2002:1880:41b:1:c879:6573:7dac:a0c6] has joined #linode
09:34-!-BarkerJr [BarkerJr@2002:1880:41b:1:c879:6573:7dac:a0c6] has quit [Read error: Connection reset by peer]
09:50-!-igufi [] has joined #linode
09:50-!-Dedalo [~Turbo@] has joined #linode
09:51-!-AviMarcus [~avi@] has joined #linode
09:52<igufi>I have a tmux session running and I detached it and closed the ssh session. Now I'm back and when I try to do tmux attach I get "no sessions" but I can see the tmux running via ps -ux .. what am I doing wrong?
09:52<EugeneKay>Using tmux.
09:52*EugeneKay hides under rock
09:53<igufi>I'm happy with screen too but I like the panels on tmux but if this keeps happening I don't think I'll risk it..
09:54<igufi>I can also see /tmp/tmux-1000
09:55-!-karstensrage [] has quit [Quit: Leaving]
09:55<mwalling>igufi: you could always kill tmux... :/
09:55<@akerl>tmux attach-session ?
09:55<@akerl>Are you logged in as the same user?
09:56<igufi>akerl: yes, same user as shown via ps -ux
09:56<igufi>and the owner of /tmp/tmux-1000
09:57<@akerl>what does "tmux list-sessions" output?
09:58<igufi>"failed to connect to server: Connection refused"
10:03<XReaper>install arch! all problems solved!
10:03<linbot>The Romans used Arch. Worked out GREAT for their civilization.
10:03<XReaper>one day
10:03<EugeneKay>I still don't know what that means.
10:03<XReaper>somebody will do what i tell them to do
10:04<XReaper>the romans died out
10:04<XReaper>the rolling release model of arch it not good for a server
10:04<EugeneKay>"not good" is putting it lightly
10:04-!-michael_mbp [~michael_m@] has quit [Quit: michael_mbp]
10:06<@mikegrb>mmm cake
10:06<XReaper>EugeneKay: have some cake on me
10:06<XReaper>and a slice of bacom
10:06<@mikegrb>mmm bacon
10:07<XReaper>maybe some lol
10:07<EugeneKay>Right after I get my rnoawk shaved.
10:07<XReaper>is the lulz guy ever on here
10:08<XReaper>or is he just in here to be annoying
10:09<EugeneKay>I've never seen 'im
10:10<XReaper>he's a netop
10:10<XReaper>i know that much
10:11<Peng>He works at Linode.
10:11<XReaper>And is a netop?
10:11<XReaper>I see what is going on here
10:12<BP{k}>he's a LinOp!
10:12<XReaper>linode has control over the network they have their channel in
10:12<XReaper>how quaint
10:13<Peng>XReaper: As a netop I don't believe he is officially representing Linode.
10:13<CornishPasty>Isn't that because they provide an irc server, XReaper?
10:13<XReaper>CornishPasty: i see
10:13<XReaper>He's representing linodes server
10:14<XReaper>lindoe needs to be less awesome
10:14<linbot>IRC supports complete sentences. Less <CR> more content, please.
10:14<squircle><3 lindoe
10:14<CornishPasty>Man, the OFTC site sucks
10:14<XReaper>nvm... EsperNet has has one of their servers on a linode on Dallas (my DC <3)
10:14<Peng>CornishPasty & XReaper: On OFTC, server owners are not granted special status.
10:14<XReaper>Peng: i know this
10:15<XReaper>ve read the crappy site :(
10:15<CornishPasty>Peng: wat?
10:15<XReaper>CornishPasty: you supply a box
10:15<XReaper>they use it
10:15<XReaper>end of stort
10:15<XReaper>EFNet is worse
10:15<XReaper>no services
10:16<CornishPasty>XReaper: that's a bit silly
10:16<XReaper>CornishPasty: whats a bit silly?
10:17<CornishPasty>Well, if someone who provided a server got angry, they could cause damage by elevating themselves...
10:17<XReaper>CornishPasty: and be de-linked
10:17<XReaper>Offering a server does not gain any benefits whatsoever
10:17<Peng>XReaper: Warm fuzzies
10:17<XReaper>Bout it
10:17<CornishPasty>XReaper: not before they could cause issues...
10:18<linbot>I'm leaving Linode for the cloud!
10:18<XReaper>Espernet got all their servers null-routed once
10:18<XReaper>!to XReaper bomb
10:18<XReaper>SpaceHobo: get it right
10:18<linbot>XReaper: IRC supports complete sentences. Less <CR> more content, please.
10:18<Nivex>ah, I did not know about !to. Will have to keep that one in mind.
10:19<XReaper>i like the <enter> my favourite key :D
10:19<Peng>XReaper: MINE IS CAPS LOCK
10:19-!-wkl [~wkl@] has joined #linode
10:19<XReaper>SpaceHobo: tell pm's me
10:19<CornishPasty>Peng: mInE tOoOoO!
10:19-!-Bdragon [] has joined #linode
10:19<XReaper>SpaceHobo: Uhm...
10:19<amitz>mine are the letter E N T R P I S
10:20<CornishPasty>I JuST haVE to PRESs iT alL thE TIme!
10:20<XReaper> L I N O D E should apply for a gTLD
10:20<CornishPasty>lol Peng
10:20<XReaper>OH GOD
10:21<XReaper>porn.linode would be interesting...
10:22<Peng>Is the entire channel drunk this morning?
10:22<@heckman>Well, I'm glad I jumped on IRC from the airport to see that
10:22<@heckman>My day is complete...
10:23<XReaper>Peng: its 1:23AM
10:23<XReaper>heckman: hi :D
10:23<Peng>It's morning Linode Time, and that's what matters!
10:23<@heckman>Ugh, I was up before the buttcrack of dawn today
10:23*XReaper wonders what linode time is
10:24<Peng>XReaper: America/New_York
10:25-!-fayimora [] has quit [Quit: fayimora]
10:25-!-igufi is now known as Guest7922
10:25-!-igufi [] has joined #linode
10:26<igufi>Hmmh, I think I did something very stupid as I tried to upgrade tmux. Ubuntu 10.04 has a fairly old version of tmux available and it has some issues with putty so I wanted a newer version.
10:27-!-stafamus [] has joined #linode
10:27<igufi>I wgetted the tar package, untarred it and did sudo ./configure and make install and all went well
10:27<@heckman>igufi: bug fixes are usually ported back to the version on there.
10:27<igufi>but apparently that did not update the package
10:27<Peng>heckman: Really? It's in universe
10:28<@heckman>Oh, I didn't realize it was in universe
10:28<igufi>now I tried to remove it via apt-get --purge remove and of it went but when I ran "tmux" it happily loaded again..
10:28<@heckman>That's because you install it from source...
10:28<XReaper>don't randomly install from source
10:28<@heckman> /
10:28<XReaper>it breakes thingks
10:29<linbot>I'm leaving Linode for the cloud!
10:29<igufi>I'm new. but I realize that's no exuce..
10:29<igufi>excuse even
10:29<@heckman>igufi: You could go back in to the folder you extracted to see if make uninstall works
10:29<igufi>mind you this linode is just for playing around so no harm done as such..
10:29<igufi>heckman: yeah, I'll try that next
10:30<igufi>But for future reference, things would have been ok if I just apt-get --purge remove'd first and then installed from the source?
10:32-!-igufi [] has quit [Quit: leaving]
10:34-!-Bryen [~bryen@] has quit [Read error: No route to host]
10:34-!-igufi [] has joined #linode
10:34<igufi>well, make uninstall just removed a couple of folders and no error messages
10:35-!-alx- [~alx-@] has joined #linode
10:37<@heckman>Hopefully that worked
10:40-!-Bryen [~bryen@] has joined #linode
10:41<igufi>Well, I'll know when I'll try to re-attach the session again.. :)
10:44-!-pascal [] has quit [Quit: Leaving...]
10:45-!-gflanders [] has quit [Quit: gflanders]
10:46-!-BlendedByChris [] has quit [Quit: Leaving.]
10:54-!-mpkossen1 [~mpkossen@] has quit [Quit: Leaving.]
10:54-!-Da-ada-2 [] has quit [Quit: Leaving]
10:57-!-laser` [] has joined #linode
11:05-!-Edgeman [] has quit [Remote host closed the connection]
11:06-!-Dedalo [~Turbo@] has quit [Quit: Linkinus -]
11:07-!-BlendedByChris [~Adium@] has joined #linode
11:07-!-Deathvalley122 [] has quit [Quit: ZNC -]
11:11-!-DephNet[Paul] [] has joined #linode
11:13-!-Austin__ [~austin@] has joined #linode
11:15-!-bbeausej [] has joined #linode
11:19-!-nmudgal [~nmudgal@] has quit [Quit: Sleeping :-) ]
11:20-!-nmudgal [~nmudgal@] has joined #linode
11:22-!-nmudgal [~nmudgal@] has quit []
11:23-!-nmudgal [~nmudgal@] has joined #linode
11:24-!-swayz [~swayz@] has quit [Remote host closed the connection]
11:24-!-dueyfinster [] has joined #linode
11:32-!-Edgeman [] has joined #linode
11:34-!-nmudgal [~nmudgal@] has quit [Quit: Sleeping :-) ]
11:36-!-jarr0dsz [] has joined #linode
11:37-!-Dataforce is now known as Guest7928
11:37-!-Dataforce` [] has joined #linode
11:37-!-Dataforce` is now known as dataforce
11:37-!-Guest7928 [] has quit [Read error: Connection reset by peer]
11:37-!-Dataforce` is "(unknown)" on (unknown)
11:38-!-rurufufuss [] has quit [Remote host closed the connection]
11:39-!-skcin7 [] has quit [Remote host closed the connection]
11:47-!-Zr40 [] has joined #linode
11:49-!-Deathvalley122 [] has joined #linode
11:52-!-jgornick [] has joined #linode
11:53-!-egronei [] has joined #linode
12:01-!-vodka [] has quit [Ping timeout: 480 seconds]
12:08-!-stephenplatz [] has joined #linode
12:09-!-Deathvalley122 [] has quit [Quit: ZNC -]
12:10-!-alester [] has joined #linode
12:11<linbot>New news from forums: How-to run the stock Arch Linux kernel (3.x) on Linode in Linux Tips, Tricks, Tutorials <>
12:12<XReaper>I should go on there and say
12:12<XReaper>i run linode kernels on my arch install
12:13<Katana>i run arch kernels in my popcorn machine
12:13-!-Deathvalley122 [] has joined #linode
12:13<linbot>The Romans used Arch. Worked out GREAT for their civilization.
12:13<XReaper>everyone disses arch
12:13-!-Deathvalley122 [] has quit []
12:14-!-karstensrage [] has joined #linode
12:14<Katana>occupy the arch
12:15<karstensrage>were there any potential issues that might have affected time synchronization in the newark datacenter around 14:00 eastern?
12:15<XReaper>topic for #archlinux on freenode "Welcome to Arch Linux World Domination, Inc."
12:15<Tea>You be dissin' Arch I be kissin' it
12:15<XReaper>no idea...
12:15<XReaper>i don't think linode uses ntp
12:16<XReaper>bah forget what i wrote
12:16-!-laser` [] has quit [Remote host closed the connection]
12:16<squircle>karstensrage: time synchronization of what? hosts? your linodes?
12:16<XReaper>i'm tired
12:16<karstensrage>i have ntp running on my vps but my understanding is that my vps is just a slice of some real hardware that also has a clock
12:16<karstensrage>squircle: yes, my vps hosts running on linode
12:17<karstensrage>what is the difference between host and linodes
12:17<squircle>karstensrage: well, I was making the host/VM distinction
12:17<Peng>karstensrage: Unless you're running the incredibly old 2.6.18 kernel, your node's clock is independent from the host's.
12:17<squircle>karstensrage: the host has a clock, but your linode doesn't use it. it uses a xen-provided clock that is unique to each instance
12:17<squircle>so what Peng said
12:17-!-hipsterslapfight [] has quit []
12:17<karstensrage>ubuntu 10.04 i think
12:17<karstensrage>not sure what kernel
12:18<Peng>karstensrage: uname -a
12:19*karstensrage fires up lish
12:19<Peng>karstensrage: uname -r is more concise, actually.
12:19<Peng>karstensrage: Not normal ssh?
12:19<karstensrage>no ssh access
12:20<vegardx>that's a first!
12:20<purrdeta>some companies and epublic places block ssh
12:21<XReaper>lish ajax is awesome
12:21<XReaper>cept my logging program has been set to wall everything to hvc0
12:23-!-nmudgal [~nmudgal@] has joined #linode
12:24<Katana>i wall everything to /dev/null
12:25<Katana>it's a great way to log
12:28<karstensrage>i purposely shut off ssh, no password access only public key and i dont have the public key here
12:31-!-seekinglatestkernel [] has joined #linode
12:32<XReaper>11.04 dies in a few days/weeks
12:34<martin->isn't it 1.5 years for regular releases?
12:34<squircle>October 2012
12:34<squircle>is EOL for 11.04
12:34<martin->10.10 however, ends in april :/
12:34<squircle>it may be days/weeks in Australia for XReaper, but it's a couple of months everywhere else
12:34<karstensrage>well the time is fine right now but i saw an anomaly yesterday at around 14:00
12:34<martin->means I have to reboot/rebuild my linode I guess
12:35<karstensrage>im just wondering if there was anything going on that might explain the anomaly
12:35<squircle>karstensrage: can you describe the anomaly?
12:35<karstensrage>time being off by a few seconds
12:35<XReaper>oh? i didn't know it was 1.5 yrs
12:36<XReaper>theres a box i have access to running 8.04.3
12:36<XReaper>THAT dies in april
12:36<squircle>XReaper: april 2013
12:36<squircle>so a year from now
12:36<squircle>karstensrage: well clock drift happens sometimes... it depends on how often you update ntp
12:36<XReaper>my dates are screwed
12:36<squircle>XReaper: 10.10 dies next month
12:36<Zr40>squircle: ntpd
12:36<squircle>yeah, yeah.
12:36<XReaper>i forgot i was in 2012
12:37<XReaper>I run arch...
12:37<Peng>squircle: "update ntp"? huh?
12:37<XReaper>and don't !arch me
12:37<XReaper>had it enough times
12:38<Zr40>!to XReaper arch
12:38<linbot>XReaper: The Romans used Arch. Worked out GREAT for their civilization.
12:38<squircle>karstensrage: it depends on how often your system's clock is updated by ntpd
12:38<squircle>Peng: I sincerely apologize for my lack of coherence.
12:38<Peng>squircle: ntpd updates the clock constantly.
12:38<karstensrage>im checking that out
12:38<Peng>squircle: Well, once per second, I think.
12:38<XReaper>squircle: i use ntpd on all my boxen
12:38<squircle>well you learn something new every day
12:39<Zr40>Peng: once per what it thinks is a second
12:39<Peng>Zr40: D:
12:39<XReaper>squircle: ntpd also uses pings and shit to account for error
12:39<squircle>that much I knew
12:39<squircle>at least I know what year it is, XReaper :P
12:40-!-Ricki [~Ricki@] has quit [Ping timeout: 480 seconds]
12:40<XReaper>No idea how i thought 5 = 4
12:41<Peng>XReaper: It doesn't actually ping. It just measures the RTT of its UDP packets.
12:41<Zr40>Peng: ping doesn't actually ping. It just measures the RTT of its ICMP packets.
12:41<Peng>XReaper: Which is a pretty inaccurate measure, because routes are frequently asymmetric.
12:42<Zr40>a well-configured ntpd uses multiple ntp servers
12:42<squircle>so using linsides' server exclusively is a Bad Idea™?
12:42<XReaper>and can cross-compare
12:43*squircle changes his ntpd.conf
12:43<Peng>Use at least 4.
12:43<Peng>Actually, use precisely 4.
12:43<XReaper>squircle: linode hosts don't care about the xen-instances
12:43<Peng>Maybe 5, if you really want to.
12:43<squircle>XReaper: you're right, they don't.
12:43<XReaper>my domain has 10 ns records
12:44<Peng>XReaper: . has more!
12:44<Zr40>squircle: it's not a Bad Idea™ if you don't depend on the clock being very accurate
12:44<XReaper>ie... don't mind if in a day the clock is a week off
12:45-!-aaronpk [] has quit [Quit: WeeChat 0.3.6]
12:45<Zr40>that won't even happen with a ntpd configured to use only a single server
12:45<Zr40>the difference would be at most the RTT
12:46<Peng>Zr40: Unless the server is off.
12:46<Zr40>Peng: the difference with the server.
12:46<Peng>Zr40: Well, yes. But normally you care about Earth-time, not some-NTP-server-time.
12:46<Zr40>Peng: but that's exactly what you're doing when you're using NTP
12:47<Peng>Zr40: But you use several servers in order to probably get a good approximation of UTC, even if one or two of them suck.
12:48<hawk>If you have the ntp server in the local network the rtt will be very low and thus all of this much less of an issue, no?
12:48<Zr40>Peng: that's irrelevant. I can use several servers which all use Zr40 time and still not get a good approximation of UTC
12:48<Peng>hawk: Depends on what "all of this" we're talking about.
12:48<Peng>hawk: I, for one, don't know.
12:49<Zr40>hawk: it's only a real issue if you depend on the clock being synchronized with other systems
12:49<Zr40>if you just want to be accurate within, say, a few milliseconds, one server is quite okay (though there's really no disadvantage to using more)
12:51<hawk>Peng: Well, not exactly sure what could be included in "all of this", so that was probably poor wording, but I meant the error introduced from ntp itself and in particular with regard to only using for instance one of these Linsides servers.
12:51<Peng>Zr40: As long as that server never goes down. And is always within a few milliseconds of what you want to be accurate relative to.
12:51<Zr40>Peng: if it goes down, the clock doesn't suddenly start drifting wildly (especially if you keep your own ntpd running)
12:51<Peng>Zr40: True.
12:52-!-seanh-ansca [] has joined #linode
12:52<Peng>Using 1 NTP server means you get no time service if 1 of your NTP servers goes down, and it will be difficult/impossible for NTP to tell if any of your time servers have gone crazy.
12:52<purrdeta>My clock does. If it cant reach a server it automagically adds or subtracts rand(0,10294858) seconds. Just for fun!
12:52<Peng>Using 3, 4, 5 or 7 guards against those issues to an increasing degree.
12:52<Peng>If you like single points of failure, be my guest.
12:55<hawk>Anyway, the error introduced is something like half the difference between the two trips, right?
12:55<Zr40>I don't know if the Linode Arch image came with ntpd or that I installed it myself, but its config defaults to three servers
12:55<purrdeta>oh arch :P
12:55<Zr40>hawk: at most half
12:55<SnoFox>Speaking of ntp,
12:55<SnoFox>Why does my laptop now think it's an hour behind since the last reboot?
12:55<Zr40>hawk: plus the error on the server
12:56<Zr40>SnoFox: daylight savings?
12:56<Zr40>SnoFox: plus dual boot?
12:56<SnoFox>I sense it has something to do daylight savings, but ... Shouldn't NTP deal with that?
12:56<hawk>Zr40: The error on the server?! Crazy talk! The server is magic!
12:56<SnoFox>No, no dual boot.
12:56<Zr40>hawk: so, magic + half the round trip :)
12:56<Zr40>SnoFox: are you running Windows?
12:56<SnoFox>Zr40: No
12:57<Zr40>SnoFox: using the correct time zone?
12:57*SnoFox went derp face
12:57<SnoFox>Apparently ntpd != ntp-client
12:57<SnoFox>I thought "ntp was running", but ntpd was running.
12:57<Zr40>ntp-client = one-time time sync
12:58<Zr40>ntpd = continuous sync with optional server
12:58<SnoFox>ntp-client fixed it although ntpd was already running
12:58<Zr40>I believe ntp-client doesn't actually do anything if ntpd is running
12:59<SnoFox> - it did something :)
12:59<purrdeta>for ntp, should I be using like the entire pool or just like northamerica
12:59-!-seekinglatestkernel [] has quit [Remote host closed the connection]
12:59<Zr40>that's not ntp-client, that's the ntp-client init script
12:59-!-mac [] has joined #linode
12:59<Peng>SnoFox: NTP typically just deals in UTC. Time zones are a different setting.
12:59<Peng>purrdeta: The main does geolocation anyway.
13:00<Zr40>it does
13:00<Zr40>two of three A records for are relatively local to me
13:00-!-mdcollins [] has joined #linode
13:00<SnoFox>Peng: my TZ environment var is correct. /etc/localtime shouldn't have changed, and strings shows me it looks like "PST" (the correct TZ)
13:00<Peng>SnoFox: If you're in the US, PDT is correct now.
13:01<Zr40>SnoFox: is /etc/localtime a symlink? if so, what does it point to?
13:01<Peng>It does geolocation to your country. If your country only has a few NTP servers, you should stick the the regional pool or neighboring countries in too.
13:01<SnoFox>You mean it doesn't take daylight savings into account? :\
13:01<Zr40>SnoFox: usually you should specify a timezone location instead of a timezone name
13:01<SnoFox>It is not a symlink, Zr40.
13:01<Zr40>like in my case, Europe/Amsterdam instead of CEST
13:01<SnoFox>Zr40: I did.
13:01-!-mac [] has quit [Remote host closed the connection]
13:02<Zr40>SnoFox: what is the output of 'date'?
13:02<Peng>SnoFox: Oh, sorry, I missed the part where you said 'strings'.
13:03<SnoFox> - Zr40, Peng, the output of `date` and a few other things to confirm my TZ settings were correct. The time is correct since I ran the ntp-client initscript
13:03-!-sebnash [~sebnash@] has quit [Quit: Linkinus -]
13:04<hawk>SnoFox: Ah, so it does say PDT
13:04<Zr40>that output seems to be correct
13:04<SnoFox>Eh. Perhaps it was a one time issue with DST. I'll bug people again if it skews off by an hour again.
13:04<SnoFox>I need to go get ready for class, though.
13:05<SnoFox>Thanks for hearing me out. :p
13:05<Zr40>hmm. one last possibility
13:05<Zr40>just a wild guess
13:05*SnoFox is still here
13:06<Zr40>your system stores the time as local time instead of UTC (usually for dual-boot compatibility with Windows). Crossing over into DST updated the system clock, but for some reason ntpd did not expect that
13:06<Zr40>again, just a wild guess and is probably incorrect :P
13:06-!-dataforce [] has quit [Remote host closed the connection]
13:08<Zr40>or... you had that shell prompt open for one hour before you typed sudo /etc/init.d/ntp-client start
13:09-!-Dataforce [] has joined #linode
13:09<mbreslin>does !mtr do ipv6
13:09<linbot>mtr combines the functionality of traceroute and ping into one easy to use tool, and the output can be useful for determining where the source of a problem is. It can be downloaded from or for Windows. MTR summaries can be retrieved in-channel using the command !mtr-CITY where CITY is fremont, atlanta, newark, dallas or london.
13:09<Zr40>!mtr-london ::1
13:10<squircle>mbreslin: I believe so
13:10<linbot>Zr40: [mtr] ::1: 1 hops, no loss, last hop average RTT was 0.1ms (urmom)
13:10<SnoFox>I'd be able to confirm your first idea if I remembered which config file tells the machine to save the time to the hwclock on shutdown, but I can't find that config file. :p
13:10<squircle>!mtr-newark 2607:f2c0:a000:138:129a:ddff:fe50:5421
13:10<Zr40>apparently it does
13:10<linbot>squircle: [mtr] 2607:f2c0:a000:138:129a:ddff:fe50:5421: 7 hops, no loss, last hop average RTT was 20.0ms
13:10<linbot>mbreslin: That URL appears to have no HTML title.
13:10<linbot>Zr40: [mtr] 6 hops, no loss, last hop average RTT was 1.7ms
13:11<Zr40>yes, !mtr-fremont is broken
13:11<mbreslin>unreachable for me for the last 10 minutes
13:11<mbreslin>back now
13:11<squircle>the site for mtr-fremont just says "GO AWAY"
13:11<Zr40>!help mtr-fremont
13:11<linbot>Zr40: (mtrfremont <an alias, 1 argument>) -- Alias for "web title$1".
13:11<Zr40>nice domain
13:12<Peng>squircle: Oh, I think it's an access restriction so only linbot can use it.
13:12<squircle>linbot <3
13:12<linbot><3 squircle
13:12<Zr40>Peng: I think it's a default vhost
13:12<SnoFox>Zr40: the hwclock is indeed in localtime, but not for Windows compatibility :p
13:13<SnoFox>Okay, I'm out for a while.
13:13*SnoFox gets less stinky and more fed
13:14-!-flashingpumpkin [] has quit [Quit: Leaving]
13:14<linbot>XReaper: That URL appears to have no HTML title.
13:15<linbot>XReaper: [mtr] 14 hops, no loss, last hop average RTT was 67.8ms (urmom)
13:15<Peng>Zr40: "Peng: nope, that response only happens when there is an invalid IP hitting the script and the linbot IP is o"
13:15<linbot>Zr40: timed out
13:16<Peng>Zr40: err, ok
13:16<Zr40>mtr timed out or the http request did?
13:16<linbot>XReaper: [mtr] 1 hops, no loss, last hop average RTT was 1.2ms (urmom)
13:16<Peng>Zr40: The HTTP request timed out.
13:16<XReaper>1 hop :D
13:16<linbot>Zr40: Unknown host.
13:16-!-Dataforce` [] has joined #linode
13:16-!-Dataforce is now known as Guest7934
13:16-!-Dataforce` is now known as dataforce
13:17<Peng>SelfishMan: !mtr-fremont seems to be having issues?
13:17<linbot>Zr40: [mtr] 9 hops, 20.0%/24.5ms
13:17-!-Guest7934 [] has quit [Read error: Operation timed out]
13:18<linbot>XReaper: [mtr] 1 hops, no loss, last hop average RTT was 0.7ms (urmom)
13:18<Peng>Protip: The !mtr commands can be run in pm windows ;)
13:18*CaptObviousman adds more hops, converts linbot to an IPA
13:18<XReaper>Protip: I know
13:19<Zr40>s/The !mtr//
13:19<CaptObviousman>it's getting to warm for stouts, sadly
13:20-!-Protip [] has joined #linode
13:20<Peng>But it's never too warm for stdout.
13:24-!-ziad [] has joined #linode
13:27-!-bbeausej1 [] has joined #linode
13:32-!-ziad [] has quit [Remote host closed the connection]
13:33-!-bbeausej [] has quit [Ping timeout: 480 seconds]
13:33-!-Oloryn [] has joined #linode
13:35-!-bbeausej1 [] has quit [Quit: Leaving.]
13:35-!-Deathvalley122 [] has joined #linode
13:35-!-Protip [] has quit [Read error: Operation timed out]
13:40-!-akiva [~akiva@] has joined #linode
13:41-!-akiva [~akiva@] has quit []
13:43-!-pwnguin [~jldugger@2600:3c00::f03c:91ff:fe96:f9b1] has left #linode []
13:52<squircle>"64 bit's speed is double over 32bit so yeah its a lot better."
13:52-!-Tom39Away [] has quit [Ping timeout: 480 seconds]
13:52*squircle sighs
13:53-!-vodka [] has joined #linode
13:54<mwalling>who said that
13:54<mwalling>i'll go cluebat them
13:54<squircle>some troll in an opensuse irc channel
13:56<dcraig>yeah there's 4 billion times as many more bits, so it's really more than a billion faster
13:58-!-Tom39Away [] has joined #linode
14:09-!-Kunda [] has quit [Quit: Kunda]
14:10-!-hipsterslapfight [] has joined #linode
14:12-!-Sputnik7 [] has joined #linode
14:12<hawk>dcraig: what? I don't even...
14:13<Zr40>4 billion times as many addressable bits
14:14<SelfishMan>Peng: maybe it is just you
14:14<KyleXY>God damnit iOS devices not supporting ovpn
14:15<mwalling>KyleXY: inorite?
14:15<mwalling>they did just add support for juniper though
14:15<Zr40>ah, openvpn. The proprietary open source VPN.
14:16-!-sasha [] has joined #linode
14:16<Zr40>that mtr is broken
14:16<linbot>Peng: [mtr] 1 hops, no loss, last hop average RTT was 0.9ms (urmom)
14:16<Zr40>or it isn't.
14:16<sasha>Hey guys, once I sign up for a certain service plan, is it difficult to upgrade or downgrade?
14:16<Peng>sasha: No.
14:16<Zr40>not difficult at all
14:16<linbot>Linodes can be resized to a different plan size via the Resize tab in the Linode Manager. Doing so will shut down your Linode and copy your disk images to their new host(this will take a few minutes). Your IP addresses and data will be unaffected, but you will need to resize your disk images.
14:17<sasha>cool, so it's pretty quick and easy?
14:17<Peng>sasha: No. It's *very* quick and easy.
14:17<sasha>awesome, thanks
14:17<Zr40>if you don't need the full disk capacity, you can create a disk image for only the size you will need. Smaller disk image = faster copying to the new host
14:17-!-sasha [] has left #linode []
14:17<Zr40>or rather s/faster/shorter/
14:17<mwalling>Zr40: for resizes it's almost a wash
14:18<mwalling>it's been timed before
14:18<KyleXY>mwalling: Trying to install GuizmOVPN, while behind a firewall
14:18<KyleXY>mwalling: Freaking irritating
14:18<mwalling>for transfers, absolutely
14:18<Zr40>mwalling: is the estimated duration provided by the manager accurate?
14:18*KyleXY grabs squid and configures it on his computer so he can tunnel it all
14:18<mwalling>Zr40: donno
14:18<mwalling>KyleXY: i ended up writing my own version of getcloak
14:18<Peng>SelfishMan: Uh-huh
14:18<Zr40>mwalling: because it's estimating 2 minutes for my 2.7 GB of disk
14:19<Zr40>if it's linear, that would mean around 15 minutes for 20 GB of disk
14:19<mwalling>and how long does the resize take?
14:19<Zr40>I haven't done one before
14:19<mwalling>yeah, longer then 13 minutes to go from 20 to 2.7
14:19<mwalling>hence, wash.
14:19<Zr40>you're talking about resizing the disk image?
14:20<Zr40>I'm talking about upgrading to another linode size
14:20-!-goose [] has joined #linode
14:21-!-Kunda [] has joined #linode
14:22<Zr40>ah, I see what you mean
14:22<Zr40>indeed, don't go shrinking your images just in preparation of upgrading
14:24<hawk>Yeah, for a transfer between hosts in the same DC it's probably not worthwhile
14:25-!-vynsynt [~Adium@] has left #linode []
14:26<Zr40>in my specific case, 2.7 GB is more than enough. If I need more RAM, it will take less time to move than if I had created a 20 GB image
14:32-!-fayimora [~fayimora@] has joined #linode
14:33<mwalling>sheesh, my /home is 10 gb
14:33<mwalling>no, 14
14:33<Zr40>Katana: yeah, SEO. Search engines don't actually care, do they?
14:33<mwalling>/dev/xvdb 19G 14G 5.0G 74% /home
14:33<Peng>mwalling: Porn or IRC logs?
14:33<mwalling>Peng: urmom
14:34<mwalling>!to Peng rimshot
14:34<Zr40>mwalling: impossible, there's still 5.0G free
14:34<Katana>Zr40: they focus on the content, not some damn url
14:34<mwalling>Zr40: LZMA compresses everything, even peng's mom
14:34<Zr40>Katana: right.
14:35<Peng>mwalling: Have you seen her jeans? She can't get any more compressed.
14:35<mwalling>... dude, i'm goign to tell your mom you talk about her like that
14:35<Peng>Yeah, you would, too.
14:36<Zr40>mwalling: you know that for any compression algorithm, some inputs exist that must result in a larger output. Obviously, you did not attempt to compress.
14:37<Zr40>s/compression/lossless compression/
14:39-!-stephenplatz [] has quit [Read error: Connection reset by peer]
14:39-!-EriksLV [~EriksLV@] has quit [Remote host closed the connection]
14:41-!-stephenplatz [] has joined #linode
14:48-!-hipsters_ [] has joined #linode
14:51-!-hipsterslapfight [] has quit [Ping timeout: 480 seconds]
14:51-!-hipsters_ is now known as hipsterslapfight
14:53-!-fayimora [~fayimora@] has quit [Remote host closed the connection]
14:53-!-fayimora [~fayimora@] has joined #linode
14:54-!-fayimora [~fayimora@] has quit [Remote host closed the connection]
15:00-!-jarr0dsz [] has quit [Quit: Leaving...]
15:06-!-EriksLV [~EriksLV@] has joined #linode
15:08-!-stephenplatz [] has quit [Read error: Operation timed out]
15:10-!-linville [] has quit [Quit: Leaving]
15:16<Peng>"Rackspace For Suckers"?
15:16<hawk>That would be an example
15:16<Katana>Windows Cloud -> Windows For Suckers
15:16<squircle>Linode For Suckers Asia Pacific?
15:16<Peng>"Linode For Suckers Asia-Pacific!"
15:17<Katana>also there's no mention of cloud in the /topic :U
15:18-!-message144 [] has joined #linode
15:19<linbot>I'm leaving Linode for the cloud!
15:19<linbot>New news from forums: Linode disk image to VirtualBox? in General Discussion <>
15:23<mbreslin>i'm leaving linode for the for suckers
15:30-!-nmudgal [~nmudgal@] has quit [Quit: Sleeping :-) ]
15:31-!-Xiy [] has joined #linode
15:31-!-Xiy [] has quit [Remote host closed the connection]
15:31-!-alx- [~alx-@] has quit [Quit: alx-]
15:43-!-wkl [~wkl@] has quit [Quit: wkl]
15:44-!-wkl [~wkl@] has joined #linode
15:45-!-wkl [~wkl@] has quit []
15:46-!-DreamPhysix [] has joined #linode
15:47<DreamPhysix>Which kernel should I be using? Right now I'm using Latest 3.0 (3.0.18-x86_64-linode24) but more recent versions are available in the list
15:47<MotoHoss>Linode is the only cloud where the sun shines all the time.
15:47<squircle>latest 3.0
15:47<EugeneKay>If there's a more recent version than what you have running you cna reboot to get it. Or don't, because meh. :-p
15:47<Peng>DreamPhysix: Stick to Latest. If the more modern ones worked right, they'd be Latest.
15:48<DreamPhysix>Alright, thanks
15:48<Zr40>DreamPhysix: any kernels available that are newer than Latest are either experimental or have known issues
15:48<DreamPhysix>ah ok
15:49<DreamPhysix>if i run apt-get upgrade on a box and it says its downloading a new kernel (not a linode box) does that mean on reboot itll automatically use the latest? the version seemed to match what i was already using though
15:50<Zr40>it depends on the distro, but most likely yes. You can check using 'uname -r' before and after rebooting
15:50<DreamPhysix>debian 64-bit stable release with kernel 2.6.32-5-amd64 which i think is the latest
15:51<mwalling>that's not a uname -r output
15:51<mwalling>but you're smarter than us
15:51<DreamPhysix>2.6.32-5-amd64 is a uname -r output
15:52<DreamPhysix>but i think debian stable is still on 2.6.x kernel
15:53-!-Beirdo_ is now known as Beirdo
15:57-!-DreamPhysix [] has quit [Quit: DreamPhysix]
15:58-!-bbeausej [] has joined #linode
16:02<MotoHoss>DreamPhysix, debian squeeze(stable) is 2.6.32-5-amd64... but it's actualy the 2.6.32-41 kernel.... that's as of like yesterday...
16:04-!-avorntur [] has joined #linode
16:05-!-Dedalo [~Turbo@] has joined #linode
16:10-!-avorntur [] has quit [Quit: leaving]
16:11-!-avorntur [] has joined #linode
16:12<hawk>MotoHoss: 2.6.32-41squeeze2 even
16:14<MotoHoss>yes hawk u r correct... ;-) only reason I had that much was I did a security update on one of my boxes last nite :p
16:19<Katana>anyone have experience with either the Radeon HD 6850 or GeForce GTX 460?
16:20<@heckman>I would go NVidia personally
16:20<Katana>Buddy's in the market atm.
16:21<mbreslin>460 is a very popular card
16:21<@heckman>Katana: why not go for a 5XX series?
16:21<JoeK>the 460? no
16:21<JoeK>the 560 or 560 Ti is
16:22<Katana>I imagine he's budget limited
16:22*Katana asks
16:22<mbreslin>the 460 was very popular during the 4 series' run
16:22<@heckman>ASUS 560 Ti can go a long way
16:23<JoeK>i have the 560 Ti counterpart
16:23<JoeK>the 6870 :)
16:23<mbreslin>i have 2x 5970
16:23<mbreslin>upgrading in oct
16:23<JoeK>wanna fight?
16:23<mbreslin>no, you win
16:23<mbreslin>i won't be going multi-gpu again
16:23<mbreslin>scaling's never worth it
16:24<Katana>but, but
16:24<Zr40>I like my mobile 6770M just fine. The regular 6850 is probably much better :)
16:24<Katana>if you scale it's automatically better!
16:24<JoeK>the xbox 720 will have the 6670 in it
16:24<Zr40>!cloud gpu!
16:24<linbot>I'm leaving Linode for the cloud!
16:24<JoeK>so look on the bright side
16:24<EugeneKay>Remind me which of the new 6xx cards is actually a new chip, and is at the $200ish price point?
16:24<mbreslin>well the 6xxx series actually scales
16:24<JoeK>define new chip?
16:25<Katana>heckman: confirmed, budget.
16:25<Kyhwana>Bah, just get the GTX 680, kicks the crap out of everything
16:25<JoeK>my 6870 costed $210
16:25<mbreslin>the 5xxx series when you add a second gpu you get like 20% max
16:25<EugeneKay>Not a rebadged Fermi or wtf the 5xx ones were
16:25<Katana>heckman: £653.52 doesn't work for him
16:25<@heckman>Katana: what is his budget and what is he trying to do?
16:26<JoeK>ah i read that as 6xxx
16:26<Katana>I pity the UK and the taxes they have.
16:26<JoeK>not 6xx
16:26<@heckman>Katana: he can get a 560 for like 150 GBP
16:27<Katana>he does a lot of source engine modding anyways
16:27<Katana>right now i'm hosting a git repo for the portal 2 mod he's working on
16:27<Katana>(and damn, that's a big repo)
16:27<mbreslin>680 or gtfo!
16:28<Katana>gitlab hates me when i open up his repo XD
16:28<mbreslin>if it's not a my little pony total conversion he's wasting his time imo
16:29<JoeK>host it on a 680
16:29<JoeK>take the mbreslin approach
16:29<Katana>heckman: ~£500 budget, and "Source modding, light 3D modelling, video editing (inc. FRAPS capture) and general gaming."
16:29<JoeK>fraps would stress the cpu before the gpu
16:30<JoeK>and would need an ssd unless you like lag
16:30<Kyhwana>Katana: what CPU does he have?
16:30<@heckman>Fraps is pretty terrible, Dxtory is better
16:30<@heckman>Also, GTX 560 or even 570 should fit in that range
16:30*Katana looks up dxtory
16:30*Katana sees kanji
16:31<@heckman>Better performance / framerates
16:31<KyleXY>w 101
16:31<mbreslin>tell your friend not to worry about his budget
16:31<mbreslin>at 5pm pdt today i'm going to win the lottery
16:31<mbreslin>i'll spring for his build
16:32<mbreslin>i've never played so i've got to have beginners luck
16:32<Kyhwana>mbreslin: can you pay for my linode for the next year too?
16:32<KyleXY>mwalling: Now I can stab cydia, heh
16:32<mbreslin>i figure i'm all set
16:32<Katana>i5-2500k apparently, clocked at 3.3ghz
16:32<Katana>mbreslin: i'd like a new system myself. this laptop's pretty restrictive for gaming ;-;
16:32<mbreslin>Kyhwana: "if i win the lottery i will pay for Kyhwana's linode(s) for 1 year --mbreslin"
16:32<@heckman>Okay, I need a nap
16:32<@heckman>o/ even
16:33<Kyhwana>Katana: hmm, should go faster than that? although that might not be the turboed freq.. Mine sits at 4.3ghz quite nicely
16:33<Katana>that's what he said it is
16:33<Katana>so probably what it's marketed as
16:34<mbreslin>363 million
16:34<mbreslin>260 million lump sum
16:34<mbreslin>-25% federal tax
16:34<mbreslin>195 million
16:34<Katana>would be nice.
16:35<Katana>invest it somewhere with 10% interest, get 100k a year to live off of.
16:35<Kyhwana>I'd take the lump sum, spend a few mil, invest the rest and live off the returns.
16:35<Katana>95mil for play money
16:35<Kyhwana>Well, only spend an amount that I could maintain said assests with the return
16:36<mbreslin>195 million lump sum or 8 million ish a year for 26 years which is 208 million
16:36<Katana>only 26 years? meh
16:36<mbreslin>i think i'd give up the extra 13 mil and go lump sum
16:36<Zr40>I'd say the interest over 195 million for 26 years is more than 13 million
16:36<Kyhwana>Zr40: yep
16:37<mbreslin>nevermind cds or anything like that just doing normal savings account is ~2mil/year
16:37<Katana>only insured up to...what, 250k?
16:37<Zr40>1% per year? that's... quite low
16:37<mbreslin>well obviously you go wherever bill gates goes
16:37<Katana>some have ceilings for interest anyways
16:38<Katana>i would rather *not* go to hell
16:38<mbreslin>Zr40: in the states 1% is normal for savings account
16:38<Katana>right now anyways
16:38<mbreslin>i mean whichever bank
16:38-!-DreamPhysix [] has joined #linode
16:38<Katana>used to be higher, but meh. crappy people not paying their bills.
16:38<DreamPhysix>Is it stupid to store private SSH keys on Dropbox?
16:38<Zr40>mbreslin: I see. Here in Europe, my plain old savings account's at 2.5% currently
16:38<purrdeta>Zr40: dang
16:39<mbreslin>Zr40: must be nice
16:39<Katana>my checking account, 12 transactions a month and you get 4% interest that month >.>
16:39<purrdeta>I've considered moving to Europe. Unsure why they would want me though :P
16:39<Kyhwana>DreamPhysix: do you trust dropbox?
16:39<Kyhwana>hmm, my savings account here gets ~3.5% atm, if I don't take money out of it
16:39<Katana>> buy soda every day for class with debit card > meet qualifier each month
16:39<mbreslin>fdic isn't even per account which is annoying
16:40<mbreslin>it's per bank
16:40<DreamPhysix>I do trust Dropbox
16:40<Zr40>Katana: what, interest on a checking account?
16:40<Katana>Zr40: for mine, yea
16:40<mbreslin>250k in a checking account and the rest burried in my backyard
16:41<Zr40>DreamPhysix: I would put a passphrase on remotely stored keys
16:41<Katana>also obligatory lol dropbox.
16:41<mbreslin>am i paranoid for having a passphrase on all my keys?
16:41<mbreslin>i thought that's just what you do
16:41-!-Abel [] has joined #linode
16:42<Zr40>mbreslin: heck, I use a passphrase on my keys, yet I never have to enter it :)
16:42<Zr40>hi Abel
16:42<Abel>I was wondering if linode has any banners?
16:42<Katana>Zr40: i do too, my yubikey is nice for handling that :D
16:42<Zr40>Katana: you do get a prompt before you activate your yubikey?
16:42<mbreslin>Zr40: i have pageant running at work and home
16:42<hawk>Abel: banners?
16:43-!-bbeausej [] has quit [Ping timeout: 480 seconds]
16:43<Katana>Zr40: depending how i do it yar
16:43<Abel>thanks Kyhwana
16:43<Peng>Kyhwana: !pr fwiw
16:44<karstensrage>how do you kill the screen session associated with lish?
16:44<Kyhwana>Peng: pr?
16:44<Peng>Kyhwana: It gives that link.
16:44<Zr40>karstensrage: what do you want to do?
16:44<Abel>type exit?
16:45<karstensrage>kill it so it has to recreate a new one
16:45<Peng>karstensrage: The *screen*? Why?
16:45<Peng>karstensrage: Or do you just want to kill the getty?
16:45<karstensrage>Ctrl-A, K doesnt seem to work
16:45<hawk>karstensrage: What did you do to it?
16:45<karstensrage>nothing, just ssh'ed into the lish console
16:45<Peng>karstensrage: Why do you want to kill screen?
16:46<karstensrage>because it keeps coming back up with the same screen and all the previous screen scroll
16:46<Katana>type "clear"
16:46<Abel>type logout
16:46<Zr40>karstensrage: lish shows the console, just like you would if you attach a physical screen
16:47-!-DreamPhysix [] has quit [Quit: DreamPhysix]
16:48<hawk>karstensrage: Ok, sounds like it's working as intended then... But you want to somehow clear it? Don't think it does that.
16:49<Katana>....okay, I kinda like working with gitlab much more than with github.
16:49<Katana>Cleaner interface >.>
16:50<karstensrage>so logging into lish, it then sets up a screen to your vps
16:50<Zr40>Katana: I don't have experience with gitlab, but github seems quite clean to me
16:50<karstensrage>you can detach from that screen.. but you cant shutdown the screen
16:51<Zr40>karstensrage: it shows whatever the physical screen would show if it were a physical machine
16:51<karstensrage>Zr40: understood but its running 'screen' to do that
16:51<karstensrage>im asking if you can shut that 'screen' down
16:52<squircle>karstensrage: lish is using screen to attach to your linode's console; the only way to get rid of screen is to plug in a physical display, or use ssh
16:52<karstensrage>afaik if you detach from the 'screen' process, the 'screen' process is still running
16:52<Kyhwana>karstensrage: then you wouldn't have access to your node via LISH?
16:53<karstensrage>Kyhwana: well in terms of what actually happens when i ssh to the lish console it could right then launch screen attached to my vps
16:53<Zr40>karstensrage: Lish using screen is how it is implemented. It's not running on your own Linode but on the Lish server, the only interaction being with your Linode's console
16:53<karstensrage>launch 'screen' and THEN attach it to my vps
16:54<karstensrage>hmm ok i think i see
16:54<karstensrage>so the 'screen' belongs to the LISH server
16:54<Zr40>karstensrage: when you detach from the console, you have access to Lish commands
16:54<karstensrage>oh ok, thanks
16:55<karstensrage>ah ok, i understand now
17:04-!-jmulder [] has joined #linode
17:06-!-MrGeneral [] has quit [Read error: Operation timed out]
17:06-!-MrGeneral [] has joined #linode
17:08-!-Abel [] has quit [Remote host closed the connection]
17:10-!-supo47 [] has joined #linode
17:11-!-descender [] has quit [Remote host closed the connection]
17:15-!-descender [] has joined #linode
17:23-!-Abel [] has joined #linode
17:33-!-eyepulp [] has joined #linode
17:36-!-supo47 [] has left #linode []
17:40-!-atula [] has joined #linode
17:45<Abel>how many of you have a blog?
17:45<Zr40>at all, or an active one?
17:46<Abel>active ones
17:46<Abel>I'm starting one, maybe we can exchange links for our blogroll
17:46<Zr40>now that's a term I haven't heard in years
17:47<Abel>check it out:
17:47<Abel>Zr40 which is your blog?
17:47<Zr40>I don't have an active blog
17:48<Abel>how do you sleep at night?
17:48<Abel>I kid
17:48<Zr40>warm and cozy in my bed
17:48<Abel>all these people and no one has one
17:48<Zr40>a bed?
17:49<Abel>a blog
17:49<Zr40>that could be interesting... a bed blog
17:52<Abel>that's a nice one
17:52<Abel>how come you have comments disabled?
17:52<staticsafe>eh, i don't care for comments :P
17:52<staticsafe>there is a post on it
17:52<staticsafe>or two
17:52-!-dubenstein [] has joined #linode
17:54-!-dubenstein [] has quit []
17:55-!-dubenstein [] has joined #linode
17:55<Abel>I see
17:57<Abel>watch out with those google ads
17:57<Abel>google disabled my account for no reason...I didn't even do anything.
17:57<Abel>took me a year to get $200 and then they decided to disable it :(
17:58-!-Abel [] has quit [Remote host closed the connection]
17:58-!-Abel [] has joined #linode
17:59<Abel>asonine, you don't write often on your blog
17:59<staticsafe>i don't indeed
17:59<staticsafe>i have major writer's block >.>
18:00<Abel>I'm going to try to write at least one every day
18:00<staticsafe>good luck sir
18:01<Abel>how much traffic you get on your blog?
18:02<staticsafe>average of 50 pgviews per day according to Wordpress stats
18:03<Abel>you should be getting more than that
18:04-!-monodemono [] has joined #linode
18:04-!-vraa [] has joined #linode
18:04<linbot>New news from forums: rutorrent in General Discussion <>
18:05<staticsafe>Abel: i would if I posted more (thats my theory anyways)
18:05<Abel>probably lol
18:05<retro|blah>rutorrent :/
18:06<Abel>you want to trade links staticsafe?
18:06<staticsafe>Abel: sure, whats yours?
18:06-!-hipsterslapfight [] has quit [Ping timeout: 480 seconds]
18:07<Abel>what title do you want?
18:07-!-Dedalo [~Turbo@] has quit [Quit: Linkinus -]
18:08-!-jmulder [] has quit [Quit: jmulder]
18:08-!-hipsterslapfight [] has joined #linode
18:09<staticsafe>Abel: hmm just put Asinine Tech
18:09<KyleXY>staticsafe: My god, that guy that want's the services wants to do transactions via IRC
18:09<KyleXY>staticsafe: lord help us all
18:10<staticsafe>KyleXY: O_O
18:10*staticsafe facedesks
18:10<Kyhwana>Where is your god now!?
18:10<KyleXY>17:06 <@Lunaqus> we want to do secure financial transactions with bitcoin
18:10<KyleXY>17:06 <@Lunaqus> and irc
18:10<KyleXY>17:07 <@Lunaqus> well secure transactions via IRC
18:10<KyleXY>I don't even need to list what can go wrong
18:11<Abel>then you better send him all the cash you can!
18:12<staticsafe>Abel: added the link
18:12-!-jgornick [] has quit [Quit: jgornick]
18:12<Abel>same here :)
18:12-!-csmrmn [] has joined #linode
18:13<Katana>KyleXY: on a scale of 1 to Boned
18:13<Katana>KyleXY: you're deep down in the rabbit hole now
18:13-!-Marzuk [] has quit []
18:13<KyleXY>Katana: Ain't me :)
18:13-!-Gshock [] has joined #linode
18:13<Katana>then run while you still can
18:14-!-Boohemian_ [] has joined #linode
18:14<Abel>staticsafe: your irc link doesn't work on your blog
18:14<staticsafe>the port number changed
18:14*staticsafe fixes
18:15<staticsafe>Abel: thanks, fixed
18:16-!-Boohemian [] has quit [Read error: Operation timed out]
18:17-!-fayimora [] has joined #linode
18:17-!-hfb [] has joined #linode
18:20<linbot>New news from forums: nginx mime type problem in Web Servers and Web App Development <>
18:21<staticsafe>wow that if block
18:21<Peng>It's awesome, isn't it?
18:22<praetorian>looks like something chesty wrote
18:22<KyleXY>holy crapers
18:22<Abel>what r you guys wowing about?
18:22<KyleXY>I can only wonder why that's not working
18:22<KyleXY>Abel: that post
18:23<Abel>lol wtf
18:24<Peng>mwalling & jed: Plotting my demise?
18:25<Peng>Will it involve long regular expressions and/or cats?
18:26-!-thegodlikehobo_ [] has quit [Remote host closed the connection]
18:27<Zr40>but will it involve long regular cat expressions?
18:27-!-Austin__ [~austin@] has quit [Quit: Leaving.]
18:28<Peng>No such thing as a regular cat.
18:28<Katana>regular expression HELL
18:28<Zr40>Peng: regular (cat expression), not (regular cat) expression
18:29<Katana>regular longcat expression
18:31<mbreslin>problem+cat+regex = 3 problems
18:32<mbreslin>and if the cat isn't spayed or neutered you're going to have even more problems
18:32<mbreslin>that's what they say on the price is right anyway
18:34<EugeneKay>We should send mikegrb to get spayed and/or neutered
18:34<staticsafe>poor mikegrb
18:34<Katana>we don't want the lulz to breed
18:37<@mikegrb>mmm bacon
18:37<Kyhwana>oh, forgot the !
18:38-!-joshdotsmith [] has joined #linode
18:42<linbot>New news from forums: Php 5.3 to 5.2 in Web Servers and Web App Development <>
18:44-!-EriksLV [~EriksLV@] has quit [Remote host closed the connection]
18:44<Kyhwana>new thing for linodes to do
18:45<Katana>other than mine bitcoins?
18:45<Katana>also IS THAT OCTOCAT
18:45<staticsafe>heh i've already finished the game
18:45<staticsafe> - enjoy
18:46<Katana>first thing i see
18:46<Katana>broken images
18:47<Katana>lol broken tiles
18:48<staticsafe>Katana: hmm im not seeing any
18:48<Katana>chrome or FF?
18:49<Katana>judas priest, even in FF it's broken
18:49<staticsafe>latest FF nightly
18:49<bob2>i don't think he was a priest
18:49<Kyhwana>works in chrome
18:50<Katana>Works in neither for me. Aurora nor Chrome dev.
18:50<bob2>tiles messed up in chrome nightly
18:50<Katana>can't get a reliable connection to the server it seems
18:51<bob2>let's blame webgl
18:51<joshiee>Perihelion ilu
18:53*praetorian and rnowak join forces to hunt down joshiee
18:55<Raccoon>6:51 PM <bob2> let's blame webgl -> wat? webgl != html5 canvas :P
18:56<bob2>it's all newfangled magic
18:56*Raccoon yells witchcraft
18:56<Kyhwana>Who's running that?
18:57<Raccoon>but it's using node.js
18:57<Raccoon>so all fun is dead
18:57<mbreslin>the web sockets are not socketing all the things
18:57<bob2>it's like twisted python but in a more annoying language and also it hangs every few days also it ships the entire v8 source tree
18:57<Raccoon>the site works in IE but not opera
18:57<Raccoon>that made me chuckle
18:58<Raccoon>apparently internet explorer supports more standards then opera now
18:59-!-Bullrush [] has quit [Ping timeout: 480 seconds]
18:59<staticsafe>heckman: got another abuse report for ya :)
19:01*praetorian abuses staticsafe
19:01<praetorian>there, write another
19:01<bob2>be nice
19:01-!-DephNet[Paul] [] has quit [Quit: Leaving]
19:01<@Perihelion>joshiee: o/
19:02<praetorian>Perihelion: \o/
19:03*Katana files staticsafe as an abuse report
19:03<@Perihelion>praetorian: o/
19:04<Katana> hah
19:04<Katana>that's probably why it's so broken for me
19:05<Katana>that's a second node connected to browserquest that i can see
19:05-!-ivan\ [~ivan@] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)]
19:06-!-Bullrush [] has joined #linode
19:08<Katana>found out why it's broken
19:09<Katana>it's using cross-origin requests
19:10<Katana>so much "KABOOM"
19:10<praetorian>Perihelion: i need breakfast
19:10<@Perihelion>So get some
19:10<praetorian>yeah i will in a moment
19:11<Abel>what time is it over there for breakfast?
19:14<@akerl>It's always breakfast time
19:14<@mikegrb>mmm bacon
19:14<EugeneKay>Mmmm bacon cake
19:16-!-undrt [] has joined #linode
19:17-!-hipsterslapfight [] has quit []
19:17-!-Oloryn [] has left #linode []
19:17-!-luisb_cba [] has joined #linode
19:20-!-ivan\ [] has joined #linode
19:20-!-bbeausej [] has joined #linode
19:22<praetorian>its business time
19:22<praetorian>ITS BUSINESS
19:22<praetorian>ITS BUSINESS TIMEEEEE
19:22<praetorian>party time is over.. .baby..
19:22<praetorian>Abel: its 10am.
19:23-!-Bullrush [] has quit [Ping timeout: 480 seconds]
19:23<Kyhwana>Lies, it's 12:13pm
19:24<squircle>lies, it's 7:24pm
19:24<@akerl>lies, it's caketime
19:25<squircle>i can agree with that.
19:25-!-Bullrush [] has joined #linode
19:26*MotoHoss had spaghetti w meat sauce for breakfast. course that was 11 hrs ago.
19:26-!-BlendedByChris1 [~Adium@] has joined #linode
19:26-!-BlendedByChris1 [~Adium@] has quit []
19:29-!-BlendedByChris [~Adium@] has quit [Ping timeout: 480 seconds]
19:30-!-undrt [] has quit [Quit: Colloquy for iPad -]
19:31-!-luisb_cba [] has quit [Remote host closed the connection]
19:31-!-fayimora [] has quit [Quit: fayimora]
19:31-!-fayimora [] has joined #linode
19:32-!-AviMarcus [~avi@] has quit [Quit: Ex-Chat]
19:33-!-ZeeO [] has joined #linode
19:38-!-mbarnett [] has joined #linode
19:38-!-mbarnett [] has quit []
19:44<auraka>need new linode hotness
19:44<SirSquidness>I am the hotness.
19:44<bob2>auraka, switch to xen, it's great!
19:45-!-teacup [] has joined #linode
19:46<auraka>bob2: there are 1400 512 linodes available...why are you slacking....go buy
19:47<bob2>I moved to the cloud
19:47<linbot>I'm leaving Linode for the cloud!
19:48-!-Bullrush [] has quit [Ping timeout: 480 seconds]
19:49<linbot>I'm leaving Linode for the cloud!
19:49-!-teacup [] has left #linode []
19:50<auraka>I just ordered some new cloud nodes.....pretty awesome
19:50-!-treec [] has joined #linode
19:51-!-Bullrush [] has joined #linode
19:54*praetorian tunes in the rdio
19:54<praetorian>Perihelion: i had breakfast.
19:54<praetorian>and sif people didnt get business time >_>
19:54<praetorian>Kyhwana especially
19:57-!-karstensrage [] has quit [Quit: Leaving.]
19:58-!-userme [] has joined #linode
19:58-!-userme0 [] has joined #linode
20:03*joshiee hugs Perihelion. :3
20:04-!-Bullrush [] has quit [Ping timeout: 480 seconds]
20:05-!-BlendedByChris [] has joined #linode
20:06-!-userme [] has quit [Ping timeout: 480 seconds]
20:06-!-joshdotsmith [] has quit [Quit: joshdotsmith]
20:06-!-danblack [~danblack@] has joined #linode
20:06-!-Bullrush [] has joined #linode
20:08-!-Josh-_ [~lol@] has joined #linode
20:08-!-Josh-_ [~lol@] has quit [Remote host closed the connection]
20:08-!-Josh-_ [lol@2001:470:b:235:1337:1337:1337:1337] has joined #linode
20:09-!-eyepulp [] has quit [Quit: eyepulp]
20:13-!-Austin__ [] has joined #linode
20:13-!-joshiee [] has quit [Ping timeout: 480 seconds]
20:23-!-userme0 [] has quit [Quit: userme0]
20:23-!-userme [] has joined #linode
20:31-!-joshdotsmith [] has joined #linode
20:32-!-Bass10 [] has joined #linode
20:34-!-Austin__ [] has quit [Quit: Leaving.]
20:39-!-userme [] has quit [Ping timeout: 480 seconds]
20:42-!-hfb [] has quit [Quit: Leaving]
20:44-!-nisstyre [] has quit [Ping timeout: 480 seconds]
20:49-!-Austin__ [] has joined #linode
20:53<Smark>what term would you use for static content (such as images) that changes (such as new files, old file deleted, but the content of each file doesnt change)?
20:53<bob2>static content
20:59-!-nisstyre [] has joined #linode
21:00<Katana>temporary files
21:00<EugeneKay>dynamic resources
21:00<EugeneKay>mutable content?
21:01<EugeneKay>^ That's the one
21:01<@mikegrb>mmm bacon
21:01<Katana>not-bacon files
21:01-!-jgornick [] has joined #linode
21:03*gdr wiggles and giggles
21:04-!-message144 [] has quit [Quit: gone]
21:05-!-Abel [] has quit [Remote host closed the connection]
21:07-!-jgornick [] has quit [Quit: jgornick]
21:11-!-danblack [~danblack@] has quit [Quit: solitude begins]
21:11-!-phliver [] has quit [Ping timeout: 480 seconds]
21:15-!-Adam1 [] has joined #linode
21:16-!-Adam1 [] has quit []
21:18-!-karstensrage [] has joined #linode
21:20-!-LinodeJavaUser [] has joined #linode
21:22-!-Craighton [] has quit [Quit: Linkinus -]
21:22<LinodeJavaUser>hello - when I putty into my linode server, I'm logging in as "console". Is that a linode setting or just a newb issue?
21:22<bob2>you mean your username is 'console'?
21:23<LinodeJavaUser>no, into console (sorry). If I fire up two putty sessions I see the same thing.
21:23<bob2>I'm lost
21:23<Kyhwana>er, are you sshing into LISH?
21:23<bob2>of course putty gives you a console
21:24<Kyhwana>That's why, because LISH is the equivilant of the "console", as if you were standing infront of the "box"
21:24<LinodeJavaUser>the console view (what would appear on the monitor connected to the computer).
21:24<LinodeJavaUser>ahh, ok I'm dumb.
21:24<Kyhwana>Sure you don't want to be sshing into the actual node via sshd?
21:25<gdr>LISH is used for things like.. if networking is disabled on your machine, you're still able to get into it with LISH
21:25<LinodeJavaUser>wait, not the website lish tool. I'm actually using putty.
21:25<purrdeta>you can SSH to lish though
21:25<Kyhwana>yes, you can ssh staright into lish
21:25<linbot>LISH allows you to perform certain actions without having to log in to the Linode Manager. LISH's primary function is to allow you to access your Linode's console, even if networking is disabled.
21:25<gdr>yes, you can use putty/anything else with the lish user/address
21:25*SnoFox LISHes into SSH
21:26<LinodeJavaUser>ah, ok. that makes sense. thanks!
21:27<gdr>np <3
21:27-!-userme [] has joined #linode
21:28<linbot>mbreslin: IP:; rDNS:; ASN adv net:; ASN: AS8708; ASN owner: RCS & RDS S.A.; Abuse contact(s):; City: Brasov; State: Brasov; Country: Romania;
21:28<Katana>mm mm romania
21:28*Katana checks sfs
21:29<mbreslin>that was the spammer from the other day
21:29<Katana><+Yukari-chan> Katana: Result from SFS API: looked up [], no results.
21:29<Katana>nothing from stopforumspam
21:29<Katana>not a site spammer i guess
21:29<MotoHoss>good food in romania.
21:29<mbreslin>what's stopforumspam?
21:29<mbreslin>is that new?
21:29<staticsafe>anti-spam tool for forums
21:29<Katana>wee little site
21:29<KyleXY>mbreslin: not really new
21:29-!-Brandon [] has joined #linode
21:29<KyleXY>mbreslin: useful though
21:29<Katana>staticsafe: not tool, database
21:30<mbreslin>sorry i was baiting you ;p
21:30<Katana>we don't provide the tools
21:30<Katana>mbreslin: we've been around since the dawn of mankind
21:30<Katana>i watched you all struggle with inventing C
21:30<mbreslin>remember that 24 hour period when you didn't mention sfs
21:30<mbreslin>good times
21:31<Katana>it actually was ongoing for a week until just now
21:31-!-Brandon [] has quit [Remote host closed the connection]
21:31<Katana>i'm blaming you for it
21:31<mbreslin>i'm too lazy to lastlog and mock you more ;p
21:31<Katana>lazy lazy
21:31<Katana>it's okay so am i
21:32<gdr>is it lazy, or a tired lazy
21:32<mbreslin>03/11(12:32) < Katana> we (the sfs team) found a site with anonymous users able to report registered users to stopforumspam through that site.
21:32<gdr>sluggish feeling, don't want to do anything
21:32<mbreslin>03/27(18:28) * Katana checks sfs
21:32*mbreslin shocked
21:32<gdr>been feeling that a lot lately, super tired, wake up tired
21:33<mbreslin>Katana: i stand corrected, well done!
21:33<gdr>however, if I had a bed maybe I would not be so tired
21:33-!-Bullrush [] has quit [Ping timeout: 480 seconds]
21:35*Katana bows to mbreslin
21:35<Katana>so that was what, two weeks maybe?
21:35-!-seanh-ansca [] has quit [Quit: Leaving.]
21:35<Katana>mind's half melted, can't count
21:35<Kyhwana>hmm, written over 2PB to a bunch of intel 320 160GBs now. (that's 2PB per drive)
21:35<mbreslin>16 days in between sfs' and 4 days in between stopforumspams
21:35<mbreslin>still probably some kind of record
21:35<mbreslin>i keed.
21:36<mbreslin>1.5 hours until i win the lottery
21:36*mbreslin nervous
21:36<linbot>You just lost.
21:36<Kyhwana>Setup some SSD linodes
21:36<Kyhwana>The intel 520's are pretty fast
21:37<mbreslin>apparently first step is to get a lawyer to setup a trust and petition the lottery to let me claim anonymously
21:38<Katana>that's actually the first mistake
21:38<Katana>paying a lawyer.
21:38<Katana>once they get their teeth in
21:38<Katana>black hole money pit
21:41<mbreslin>if they find a way to turn a 1500$ trust creation and suck up my 194 million
21:41<mbreslin>they win
21:41<mbreslin>i will just shrug and say well played
21:42<mbreslin>as for linode i will finally catch them all
21:42<mbreslin>not that i have the slighest need for a tokyo linode but what the hell right?
21:42-!-joshdotsmith [] has quit [Quit: joshdotsmith]
21:42<Katana>get prenuptuals
21:43<mbreslin>i'm married there isn't any way in which i wouldn't lose half the winnings
21:43<mbreslin>married already+
21:44<mbreslin>someone tried to go the anonymous route last year because her and her husband weren't even living together and were on the verge of divorce
21:44<mbreslin>he got half
21:46-!-LinodeJavaUser [] has quit [Ping timeout: 480 seconds]
21:46<mbreslin>it's sad that the most(only?) competant entity in our government is the irs
21:49-!-Linear [] has joined #linode
21:50-!-Linear [] has quit []
21:50-!-Linear [] has joined #linode
21:50<amitz>you don't kknow if they secretly work on commission basis.
21:51<mbreslin>hadn't thought of that
21:54<Katana>oh so you're screwed
21:55*Katana kicks Handbrake into gear, tries a re-encode for his psp
21:56-!-userme8 [] has joined #linode
21:56-!-BlendedByChris [] has quit [Quit: Leaving.]
22:00-!-userme [] has quit [Ping timeout: 480 seconds]
22:00-!-wkl [~wkl@] has joined #linode
22:01-!-Bullrush [] has joined #linode
22:01-!-Brandon [] has joined #linode
22:02<Brandon>Hello, Does anyone know if a vPS CPanel license will work on a linode server?
22:02-!-wkl [~wkl@] has quit [Read error: Connection reset by peer]
22:03-!-wkl [~wkl@] has joined #linode
22:03<@heckman>Brandon: of course
22:03<Brandon>Awsome! Ordering now :)
22:06-!-Nemykal [] has joined #linode
22:08-!-userme8 [] has quit [Ping timeout: 480 seconds]
22:09-!-monodemono [] has quit [Quit: My battery probably died...]
22:09<Nemykal>hey, how do I set up private networking between two linodes? one in tokyo, one in fremont. I've given them their respective ip's e.g. but they can't ping each other?
22:09<retro|blah>You can't
22:09<@akerl>The private network is private. It doesn't leave the datacenter
22:09<Nemykal>ah okay, so its datacentre only
22:09<Nemykal>fair enough
22:10<Nemykal>misunderstood, thought it was across all linode datacenters
22:10<Kyhwana>uh, yeah
22:10<JoeK>linode should so totally do that though
22:10<JoeK>set up a VPN between the D/C's :P
22:10-!-pgentoo [] has joined #linode
22:10<Kyhwana>Ohoh, can they set it up so that I can access my linode via a private IP from home! and i'll get a 2nd one in a DC, I want ot access that via private IP too!
22:11<pgentoo>hey guys, just curious how cpu in a linode scales with ram. For example, if i double my ram do i end up with double cpu?
22:11<@heckman>All Linodes have access to four VCPUS
22:11<@heckman>That doesn't change based on plan size
22:11-!-rurufufuss [] has joined #linode
22:11<linbot>CPU time is rarely an issue - you get access 4 cores of a Xeon 5xxx CPU, usually a L5420. The number of Linodes per host is limited, and all Linodes on a host are the same type. In the worst-case scenario, you're splitting CPU time evenly with your fellow Linoders, but are still able to use the full potential of the host if others are idle(which they usually are).
22:12<pgentoo>i know xen has cpu credit though. does that change linearly based on ram out of the total available on the box?
22:12<Kyhwana>The bigger the plan, the less people you're sharing that box with
22:12<Nemykal>okay, onto another topic. how do I remove this linode i just bought from tokyo, and make a new one in fremont?
22:12<retro|blah>lol squircle
22:12<linbot>Linodes can be migrated between datacenters by simply opening a ticket. Your disk images will be copied intact, but you will be issued a new IP address. Once a migration is configured, it makes a button on the dashboard available for you to start the migration yourself. Migration typically takes about 1-3 minutes per GB.
22:13<pgentoo>but, is credit scheduling set to evenly space out the cpu under load?
22:13<squircle>pgentoo: I would assume so, but from what I've seen, most hosts are usually close to idle.
22:13<pgentoo>i understand that anyone is allowed to burst though...
22:14<pgentoo>ok, just curious. I know i ran into issues with one vm spiking and causing load issues in others (on my own physical xen hosts) which was resolved by setting proper weights, just curious if linode did the same
22:15<squircle>pgentoo: well if you have multiple linodes, they're all on different hosts
22:15<pgentoo>i do like that the linode dashboard shows the host load as well as the vm load. :)
22:15<@heckman>We make sure scheduling is fair. I honestly can't think of a case where I've seen CPU contention to be an issue.
22:15<pgentoo>heckman, thanks. that is what i was looking for. :)
22:15<squircle>why does nobody set IPv6 PTR records >:(
22:16<pgentoo>just have some heavier traffic coming soon for our sites (currently across 6 linodes and some physical boxes) and thought about it so figured i'd ask. :)
22:16<@heckman>squircle: I do for my stuff at home.
22:16-!-skcin7 [] has joined #linode
22:16<Kyhwana>My ISP won't delegate my v6 prexies at home :(
22:16<@heckman>I don't for my Linodes as I haven't sold myself on an addressing convention
22:16<squircle>heckman: so do I, but I have this 21-hop MTR, and the only PTRs are my computer, my router and the final destination
22:18<pgentoo>heckman, when copying vm clones between datacenters, do you compress and then move, or just move as is?
22:19<@heckman>byte for byte copy
22:21<pgentoo>you try scp'ing with compression or something? Seems like it could maybe be faster?
22:21-!-userme [] has joined #linode
22:21<squircle>just compress it with md5sum
22:22<squircle>makes it really tiny
22:22*squircle runs away, laughing evilly
22:23<retro|blah>Have fun
22:23-!-monodemono [] has joined #linode
22:24<squircle>i'm trying to find that yahoo answers post asking how you can uncompress md5. (I used md5sum to compress but I can't figure out how to get the original file back!)
22:24-!-SelfishMan [] has quit [Quit: meh]
22:25-!-Brandon [] has quit [Remote host closed the connection]
22:26-!-skcin7 [] has quit [Read error: Connection reset by peer]
22:27<pgentoo>linode have any plans on expanding to south america or Australia in the near future?
22:27<Kyhwana>pgentoo: they don't pre announce plans like that
22:27<pgentoo>ok, anyone have any recommendation for linux vm providers in those regions?
22:28<Kyhwana>pgentoo: you have users that are latency sensitive in those areas?
22:28-!-Turl [] has quit [Ping timeout: 480 seconds]
22:29<pgentoo>we already serve up from tokyo, uk, east and west coast in the US> but have large number of users from mobile devices in both those areas too
22:30<pgentoo>not urgent, but something we're looking into...
22:31<Kyhwana>I think rimuhosting has stuff in aussie and NZ
22:31<pgentoo>yeah, i've heard of them
22:31<Kyhwana>But be aware that transit/network stuff in aussie/NZ is higher than the US/UK
22:31<ajmitch>too damn high
22:34<pgentoo>yeah, i understand that. we use very little bandwidth though, so i think it would be ok.
22:34-!-tyger86 [] has joined #linode
22:35*ajmitch was pleasantly surprised at the cost of an auckland-based VPS with rimuhosting when pricing it up last week
22:35<tyger86>hello can any one help me with the mess of an ssh set up I made
22:35<pgentoo>ajmitch, did you sign up or just price it out?
22:35-!-Austin__ [] has quit [Quit: Leaving.]
22:36-!-goose [] has quit [Quit: Staying alive can kill you, it's taken years off of my life.]
22:36<ajmitch>pgentoo: just priced it
22:36<Kyhwana>!to tyger86 ask
22:36<linbot>tyger86: If you have a question, feel free to just ask it -- someone's always willing to help. If you don't get a response right away, be patient!
22:36<ajmitch>a little OT in here, I guess :)
22:36-!-SelfishMan [] has joined #linode
22:36<pgentoo>hehe, fair enough. thanks.
22:37-!-pgentoo [] has quit []
22:37<tyger86>So I added my development rsa key to the know_hosts on my vps and no I get Received disconnect from 2: Too many authentication failures for deployer
22:37<tyger86>I upped the ConnectionAttempts to 100 and restarted the ssh server but it didn't help
22:37<ajmitch>I hope you mean authorized_keys, not known_hosts?
22:38<Kyhwana>tyger86: you want to add it to .ssh/authorized_keys, not known_hosts
22:38<tyger86>Sorry yeah I did
22:38<Kyhwana>(or use ssh-copy-id, which does it automagically for you)
22:38<Kyhwana>tyger86: you copied the public key over?
22:38<retro|blah>ConnectionAttempts? Is that from fail2ban or denyhosts or something?
22:39<tyger86>yes with this command cat ~/.ssh/ | ssh $1@$2 'cat >> ~/.ssh/authorized_keys'
22:39<retro|blah>Oh, nm
22:39-!-skcin7 [] has joined #linode
22:40<Kyhwana>tyger86: can you still login using your username/password?
22:41<tyger86>yes with my phone but my lap top tries to use the key and gets booted
22:42<Kyhwana>so tell it not to use the key?
22:44<tyger86>Ok but I do want to get the key working at some point what was that utility that did it automatically called
22:45<tyger86>cool I will give that a shot thank you
22:45<Kyhwana>are you sure you did it as the right user?
22:45-!-Kuukunen [] has quit [Ping timeout: 480 seconds]
22:46-!-userme [] has quit [Ping timeout: 480 seconds]
22:46<tyger86>I have 2 root and deployer and I did it for both of them. I read this blog post which is why I thought it had to do with connection attempts
22:49<Kyhwana>well, it's obviously not got the key setup if it's not letting you login with it
22:54<tyger86>thank you Kyhwana for the help that ssh-copy-id is a cool utility
22:58-!-tyger86 [] has left #linode []
23:00-!-bbeausej [] has quit [Quit: Leaving.]
23:00-!-Turl [] has joined #linode
23:03-!-virtualr1ot [] has joined #linode
23:05-!-gflanders [] has joined #linode
23:06-!-virtualroot [] has quit [Ping timeout: 480 seconds]
23:06-!-SelfishMan [] has quit [Quit: meh]
23:07-!-hfb [] has joined #linode
23:09-!-SelfishMan [] has joined #linode
23:12-!-Kuukunen [] has joined #linode
23:12-!-vraa [] has quit [Quit: Leaving]
23:14-!-john51 [] has joined #linode
23:14<john51>I'm trying to backup my linode with the instructions from the linode library:
23:15<john51>However, I dd-ed the disk image to my computer and tried to mount it, but no filesystem was recognized
23:15<john51>Output of fdisk:
23:15<bob2>fdisk looks for a partition table not a filesystem
23:15-!-Santo [] has joined #linode
23:15<bob2>so you'll need to say how you came to the conclusion that "no filesystem was recognized"
23:16-!-Turl [] has quit [Quit: . o O ( why am I quitting? )]
23:16<john51>@bob2: Mac's disk image utility says no FS is in the disk image
23:16-!-userme [] has joined #linode
23:16<bob2>this isn't twitter
23:17<bob2>also OS X has no ext2/3/4 support
23:17<bob2>so unsurprising
23:17<john51>oh, alright
23:17<bob2>there's some third party drive I think
23:17<bob2>no idea if it is any good or not
23:17<john51>ok, I'll try to see if there are any ext3 plugins for the mac
23:18<bob2>an image itself probably isn't super useful if you're an os x user
23:19-!-Adam2 [] has joined #linode
23:20-!-Turl [] has joined #linode
23:20<Adam2>question - I can putty into my linode server using the lish url/ip. But when I putty using the ssh url/ip, it says Network Error: connection refused.
23:21-!-fo0bar [fo0bar@2607:f740:0:d::face] has quit [Quit: leaving]
23:21<Adam2>anyone seen this before?
23:21<auraka>Adam2: firewall?
23:21<@Praefectus>did you change your ssh port?
23:21<bob2>== you broke ssh
23:22<Adam2>i'm new to this box. so I'm not exactly sure if the previous dev setup or broke something.
23:22<Adam2>how do I check the ssh port?
23:22<@akerl>Adam2: netstat -lpn
23:22<auraka>or lsof -n -iTCP
23:22-!-john51 [] has quit [Remote host closed the connection]
23:23<auraka>or lsof -n -P -iTCP might be even better to not resolve ports against /etc/services
23:23<auraka>but i'm betting firewall
23:25<Kyhwana>or changed ssh port
23:25-!-Ghost_ [~Ghost@] has quit [Quit: Leaving]
23:26-!-rurufufuss [] has quit [Ping timeout: 480 seconds]
23:26-!-alester [] has quit [Quit: alester]
23:28<Adam2>lsof -n -iTCP and lsof -n -P -iTCP don't return anything
23:28<Adam2>netstat -lpn returns a lot but nothing on port 22
23:29<Kyhwana>then it's not listening on port 22. What port does it say sshd is listening on?
23:29<Kyhwana>(under listen address)
23:29<@Praefectus>Adam2: grep Port /etc/ssh/sshd_config
23:32-!-fo0bar [fo0bar@2607:f740:0:d::face] has joined #linode
23:32<Adam2>awesome. that grep returned port a way different port. perfect
23:32<auraka>and i lose :-/
23:33<Smark>when using nginx load balancing (upstream) does all the data still pass through the balancer, or is it passed off to the other nodes?
23:33<Adam2>thanks auraka
23:34<bob2>not sure what you mean though
23:34<@akerl>Smark: How would the traffic know where to go if it weren't being passed through the balancer?
23:34-!-fayimora [] has quit [Quit: Busy…..zzzzz]
23:35-!-Kunda [] has quit [Quit: Kunda]
23:35-!-SelfishMan [] has quit [Quit: meh]
23:36-!-rurufufuss [] has joined #linode
23:36-!-SelfishMan [] has joined #linode
23:43-!-fayimora [] has joined #linode
23:46-!-Kunda [] has joined #linode
23:47-!-gflanders [] has quit [Quit: gflanders]
23:51-!-eyepulp [] has joined #linode
23:53-!-vraa [] has joined #linode
23:59-!-VS_ChanLog [] has left #linode [Rotating Logs]
23:59-!-VS_ChanLog [] has joined #linode
23:59<linbot>Point (0.93822994, 0.50680314) falls outside of the unit circle. Hits: 110515 of 140290 (π ≈ 3.151044265450139 - 0.009451611860346). http://π
23:59<linbot>Kyhwana: Point (0.14528390, 0.31814150) lies within the unit circle. Hits: 110516 of 140291 (π ≈ 3.151050316841423 - 0.009457663251629). http://π
23:59<Kyhwana>still not 3.14
---Logclosed Wed Mar 28 00:00:19 2012