Back to Home / #linode / 2012 / 08 / Prev Day | Next Day
#linode IRC Logs for 2012-08-03

---Logopened Fri Aug 03 00:00:07 2012
---Daychanged Fri Aug 03 2012
00:06-!-dhubbard [] has joined #linode
00:06-!-Kane` [] has quit [Quit: Leaving]
00:08-!-userme [] has quit [Ping timeout: 480 seconds]
00:08<dwfreed>chesty: nop
00:11<Takyoji>Any reason for these errors at boot?
00:11<Takyoji>regarding init
00:12<dwfreed>Takyoji: how much stuff do you have in /tmp ?
00:13<Takyoji>Nothing is present in /tmp at all
00:13<dwfreed>and I imagine you're not out of space in /
00:13<orudie>caker, are you there ?
00:14<Takyoji>I'm not out of space
00:14<@heckman>orudie: doubtful
00:14<@heckman>!to orudie date
00:14<linbot>orudie: 12:14 AM, August 03, 2012
00:14<@heckman>orudie: what do you need?
00:14<Takyoji>I'm also having issues with two additional IPs that aren't getting routed properly, which were transferred between Linode instances
00:14<dwfreed>Takyoji: did you reboot both Linodes?
00:15<orudie>heckman, I need to say hello to him
00:15<@heckman>orudie: ah, he may be lurking but probably not.
00:15<Takyoji>The other Linode is decommissioned, but yes, I guess I could
00:15<staticsafe>If I wanted an extra IPv6 address for a node what prefix length should I request?
00:16<dwfreed>staticsafe: if you don't want to deal with routed blocks, just ask for a /116
00:16<orudie>I had been a linode customer for a while, then introduced linode to my prior company which still uses linode
00:16<staticsafe>cool thanks dwfreed
00:16<orudie>but I had to move on ...
00:17<orudie>learned a great deal from mesing with linodes
00:17<orudie>currently working with red had mostly and some centos
00:17-!-mcinerney [] has quit [Ping timeout: 480 seconds]
00:19-!-mcinerney [] has joined #linode
00:20-!-arooni-mobile [~arooni-mo@] has joined #linode
00:23-!-efo [] has joined #linode
00:23-!-efo [] has quit [Remote host closed the connection]
00:25<@Praefectus>"red had"?
00:26-!-jhanjon [] has joined #linode
00:26-!-hfb_ [] has quit [Remote host closed the connection]
00:27<jhanjon>switch user) to some accounts
00:28<arooni-mobile>where i do need to put my generated ssl keys? in /etc/ssl? (for nginx)
00:28-!-danblack [~danblack@] has joined #linode
00:28<jhanjon>oops I'm unable to su to some users but able to others, what gives? centos 6 using ssh
00:29-!-jhanjon [] has quit [Remote host closed the connection]
00:29-!-jhanjon [] has joined #linode
00:29-!-jhanjon [] has quit [Remote host closed the connection]
00:30-!-zivester [] has quit [Quit: Leaving]
00:30-!-jhanjon [] has joined #linode
00:30-!-Drone4four [] has quit [Read error: Operation timed out]
00:31<jhanjon>In SSH I can su to some users but not others, what do I need to do for the others? Centos 6
00:35-!-_j45 [] has left #linode []
00:40-!-mcinerney [] has quit [Ping timeout: 480 seconds]
00:41-!-jhanjon [] has quit [Remote host closed the connection]
00:44-!-tyrel [] has joined #linode
00:45<JoeK>jscholl: 'to' or 'from'?
00:45<JoeK>er, nevermind i guess
00:46<tyrel>Hello I am trying to deploy a rails app with nginx and unicorn. I am almost there the unicorn daemon starts up with no complaints as does nginx but when I make a request I get a 502 response and this error in nginx's error log 2012/08/03 00:41:59 [error] 4970#0: *19 connect() failed (111: Connection refused) while connecting to upstream, client:, server: , request: "GET /pages/home HTTP/1.1", upstream: "", host:
00:47-!-Drone4four [] has joined #linode
00:48<arooni-mobile>i'm seeing "The site's security certificate is not trusted!" after following a guide on slicehost. i generated a ssl cert with startssl... not sure what i need to check next. nginx and ubuntu 10.04
00:50<kyhwana>tyrel: are you sure unicorn is running/listening?
00:50<kyhwana>arooni-mobile: whats your site?
00:52-!-vraa_ [] has joined #linode
00:53<tyrel>well there are 5 proses named ruby one that matches the unicorn ps id however nmap shows none are listening
00:55<tyrel>my nginx.conf is and my unicorn.rb is
00:56<kyhwana>arooni-mobile: ?
00:56<kyhwana>don't /msg me
00:56<@Praefectus> /msg kyhwana HAI!
00:56<kyhwana>tyrel: hmm, so you're using a unix socket to conenct to unicorn, is that how you have unicorn setup?
00:57<kyhwana>It'll be looking for /home/CannaPages/tmp/sockets/unicorn.sock to exist
00:57*kyhwana baps Praefectus
00:57*Praefectus wangs kyhwana
00:58<arooni-mobile>sorry kyhwana i was trying to give yout he url without making it public
00:58<@heckman>tyrel: where does unicorn listen by default?
00:58<kyhwana>arooni-mobile: why?
00:58-!-vraa [] has quit [Read error: Operation timed out]
00:58<@heckman>arooni-mobile: oh fun, redacting information already
00:58<arooni-mobile>cuz you asked "arooni-mobile: whats your site?"
00:58<@heckman>Which means put it in the channel.
00:58<kyhwana>and if I go idle and you only tell me, I can't help you, someone else in channel might though, etc
00:58<tyrel>listen "/home/CannaPages/tmp/sockets/unicorn.sock", :backlog => 64
00:59<@heckman>Ah wow, glazed over that
00:59<@heckman>gist: (as root) netstat -pln
01:00<@heckman>Looks like unicorn is listening
01:01-!-Bass10 [] has quit [Read error: Operation timed out]
01:01<@heckman>You configured it to listen on a unix socket, which wouldn't show up in nmap.
01:01<@heckman>Also, using if statements in nginx.conf is frowned upon
01:02<@heckman>You should really be doing try_files there.
01:02*dcraig frowns upon heckman (all over)
01:02*heckman slaps dcraig around a bit with Ubuntu.
01:03<tyrel>haha maybe proxy_pass http://localhost; instead of proxy_pass http://;
01:03<@heckman>why localhost?
01:03<@heckman>Unicorn isn't listening on localhost.
01:04<tyrel>loop back then?
01:04<tyrel>right now it says proxy_pass http://unicorn_server;
01:04<@heckman>Yeah...which is defined by the upstream block above.
01:05-!-bchristensen [] has quit [Remote host closed the connection]
01:06<@heckman>tyrel: no, a unix socket is not loop back either.
01:06<@heckman>Also, what the heck is at the end of line 46 in that gist?
01:06<@heckman>fail_timeout=0 should be on the same line, and should have a space between the socket file and that directive.
01:07<@heckman>I would like to suggest using try_files instead of that funky if statement
01:08<tyrel>alright I will look for another nginx.conf to try I am not good enough to write my own
01:08<@heckman>I mean, that one is just weird. Have you looked at the documenation for Unicorn?
01:09<@heckman>I bet they have an Nginx configuration example available.
01:09<@heckman>If not, I know the Python fork of Unicorn (GUnicorn) does.
01:09<@heckman>And it should be applicable.
01:09<@heckman>Reading up on what you are doing in the Nginx configuration, instead of blindly pasting stuff, might be helpful for future endeavours as well.
01:10<tyrel>Thank you for the help. I was planning on just getting this site up and slowing learning and customizing over time
01:12-!-seanh-ansca [] has joined #linode
01:13-!-SnapALoop [] has joined #linode
01:13<SnapALoop>Never drop your cell phone again. Avoid Costly Mistakes. Watch the video its cool. Ill say anything .. click the link so I can send my kids to college... Actually all I want is a beer.. Just click it.
01:13-!-SnapALoop [] has quit []
01:14-!-bchristensen [] has joined #linode
01:14-!-fisted [] has joined #linode
01:14-!-danblack [~danblack@] has quit [Read error: Operation timed out]
01:15<@heckman>Sounds legit.
01:15<SleePy>Smells like spam, tastes like spam, looks like spam.. It must be chicken
01:15<dcraig>I've dropped my phone a few times
01:15<dcraig>it was really a pain to have to bend over and pick it up
01:19-!-Andrei [] has joined #linode
01:20-!-fisted_ [] has quit [Ping timeout: 480 seconds]
01:20-!-seanh-ansca [] has quit [Ping timeout: 480 seconds]
01:21<Andrei>Hi All, I am trying to setup postfix using this guide
01:21<Andrei>Everything looks good, but emails won't make it to the server from outside.
01:21<Andrei>If i email to users on the server from the server, it works perfectly, but from outside the server, in it does not.
01:21<Andrei>Any ideas?
01:22<kyhwana>Andrei: what do your log files say? is postfix actually listening on your public interface? is it firewalled?
01:22<kyhwana>huh, the guide even has a "check your logs" bit
01:22<Andrei>the logs are fine
01:23<Andrei>they match the guide
01:23-!-Kane` [] has joined #linode
01:23<Andrei>I am thinking it may be the firewall, but I do not know where to logs are for iptables
01:23-!-seanh-ansca [] has joined #linode
01:23<Andrei>I checked the authlog and the syslog and everything seems ok in there
01:24<kyhwana>so pastebin "sudo iptables -L" ?
01:24<kyhwana>and "sudo netstat -antp"
01:24<Andrei>syslog showed iptables denying the packets
01:24<Andrei>guess i didn't check them closely enough lol
01:25<Andrei>sorry to have wasted your times! Thanks
01:25<squircle>we're here to help
01:25<SleePy>your logs match the tutorials logs?
01:25<Andrei>yes but the tutorial logs do not mention IP tables
01:25<Andrei>and i had forgotten that I set it up
01:26-!-triplei [] has quit [Ping timeout: 480 seconds]
01:26<SleePy>I curse you for taking away that 26 seconds I spent reading your text.. I will never get that back
01:27-!-danblack [~danblack@] has joined #linode
01:28-!-bchristensen [] has quit [Remote host closed the connection]
01:32-!-mcinerney [] has joined #linode
01:37-!-mpkossen [~mpkossen@] has joined #linode
01:40<tyrel>Yea thank you and damn do I feel stupid I thought I had a symbolic link from my sites enabled pointing to the file I was editing turns out there wasn't. Now I am precompiling assets and I should be good.
01:41-!-mcinerney [] has quit [Ping timeout: 480 seconds]
01:45-!-tyrel [] has quit [Remote host closed the connection]
01:47-!-Gnintendo [] has joined #linode
01:49-!-_j45 [] has joined #linode
01:57-!-_j45 [] has quit [Quit: Leaving.]
01:58-!-_j45 [] has joined #linode
02:00-!-fuchs_fuchs [] has joined #linode
02:01-!-hfb [] has joined #linode
02:08-!-fuzzman442 [] has joined #linode
02:09-!-fuzzman442 [] has quit [Remote host closed the connection]
02:10-!-fuzzman4_ [] has joined #linode
02:11-!-endzyme [] has joined #linode
02:14-!-message144 [] has quit [Remote host closed the connection]
02:15-!-duckxx [] has quit [Ping timeout: 480 seconds]
02:19-!-Andrei [] has quit [Remote host closed the connection]
02:19-!-message144 [] has joined #linode
02:25-!-sivy [] has quit [Remote host closed the connection]
02:41-!-seanh-ansca [] has quit [Read error: Operation timed out]
02:43-!-danblack [~danblack@] has quit [Ping timeout: 480 seconds]
02:45-!-seanh-ansca [] has joined #linode
02:45-!-sbattey [] has joined #linode
02:51-!-arooni-mobile [~arooni-mo@] has quit [Ping timeout: 480 seconds]
02:52-!-fuzzman4_ [] has quit [Remote host closed the connection]
02:52-!-danblack [~danblack@] has joined #linode
02:53-!-message144 [] has quit [Remote host closed the connection]
02:58-!-MarkJ [] has quit [Quit: leaving]
02:59-!-_j45 [] has quit [Quit: Leaving.]
03:03-!-MJCS [] has joined #linode
03:16-!-fuchs_fuchs [] has quit [Quit: fuchs_fuchs]
03:18-!-danblack [~danblack@] has quit [Ping timeout: 480 seconds]
03:19-!-ben____ [~ben@] has joined #linode
03:20-!-ben____ is now known as ziph
03:20-!-seanh-ansca [] has quit [Quit: Leaving.]
03:21<ziph>Hey there, my system won't boot after an apt-get upgrade where initscripts, sysv-rc, sysvinit-utils, libudev0 and udev were upgraded, is there any way I can mount the original install disk to downgrade the older versions of those?
03:22<ziph>(It's Ubuntu 12.04 LTS)
03:23<chesty>but don't downgrade, find and fix the problem
03:25-!-Gnintendo [] has quit [Quit: Leaving]
03:25<ziph>All I'm seeing in the kernel log is "mountall: Disconnected from Plymouth".
03:25<ziph>And I don't get a console (even in single user mode).
03:26<ziph>Is there some way to poke in deeper?
03:27<ziph>e.g. is it possible to start a second console from lish?
03:27<chesty>boot with the rescue image (finnix) and look at the log files. also google error messages
03:28<stevepiercy>where in my account can I find the number of cores for my linode?
03:28<@Praefectus>you have access to 4 cores
03:29<@Praefectus>(regardless of plan)
03:29<chesty>Praefectus: that didn't answer the question
03:29<stevepiercy>yeah, i scoured the website looking for the answer
03:29<@Praefectus>chesty: since it's on the main page of the site, it is fairly pointless to list it in your account
03:29<ziph>chesty: Thanks, I'll keep plugging away. Worst case I can download the packages through Finnix as you suggest.
03:30<stevepiercy>Praefectus: i missed that
03:30<@Praefectus>features box in the middle of the main page: 4 processor Xen instances
03:30<stevepiercy>argh! i was looking for the word "core" or "CPU"
03:30<stevepiercy>but i understand that they are virtualized, thus not "real" cores or cpus
03:31<stevepiercy>anyway, that will get me along. thank you
03:31<chesty>they are real, you just share them
03:33<chesty>every host has 2 x 4 cores, so 8 cores in total, you get access to 4 at a time
03:33<stevepiercy>chesty: are those details written up somewhere? i am writing up a spec for a client, and i would like to provide a link to read more
03:34<chesty>nah, linode are more secretive than the nsa
03:34<stevepiercy>yeah, but you can hack the NSA ;)
03:35<chesty>and you can hack linode
03:35<stevepiercy>sorry, was attempting humor
03:35<@Praefectus>so whats the difference? oh, you have nice men in generic black suits knocking on your door if you hax0r the nsa
03:35<Peng>stevepiercy: Well, you can look at /proc/cpuinfo in a node. We assume that Linode isn't spending money on dual-capable CPUs and then only using one... :P
03:36<Peng>Praefectus: Can you officially confirm that Linode hosts have two CPUs?
03:36<@Praefectus>Peng: No.
03:36-!-Ehtyar [] has quit [Quit: There are two major products that come out of Berkeley: LSD and UNIX. We don't believe this to be a coincidence.]
03:36<Peng>Praefectus: I see. It's not the biggest trade secret, you know.
03:37*Praefectus is following the rules.
03:37<chesty>Praefectus: can you officially confirm linode has computers?
03:37<@Praefectus>chesty: No.
03:37*stevepiercy gives Praefectus a cookie
03:37<Peng>Praefectus: I'm not blaming you.
03:38<@Praefectus>Peng: i know
03:38<chesty>reminds me of that manning trail transcript
03:38<Peng>Praefectus: <3
03:38<@Praefectus>manning is lucky he got a trial
03:39<chesty>pffft, what about watergate?
03:39<@Praefectus>releasing docs from a politcal party office is not the same
03:40-!-TimTim [] has quit []
03:40<chesty>what about your constitution and the law about not being help for more than 180 days without charge?
03:41<Peng>That doesn't apply to criminals.
03:41<@Praefectus>especially not traitors
03:42<chesty>your country is a police state, you have no freedom, you are spied on by the government, you have poor education, poor health, poor economy, corruption in all ranks
03:43<@Praefectus>chesty: i have dual citizenship, which country are you referring to?
03:43<Peng>Well you're a terrorist!
03:43<Peng>Dual citizenship? You're an un-American terrorist too!
03:43<chesty>dinner time, I'll troll you all in a bit
03:44<@Praefectus>Peng: my family has been defending this country since WW2, and i have a lot of family in government agencies
03:45<@Praefectus>("this country" being the USA)
03:45<@Praefectus>Peng: PROVE IT
03:45<Peng>Praefectus: NO NEED
03:46<Peng>Wait, if you're a family of spies, why are you working for Linode? Investing your paycheck in SPYING? Selling corporate secrets? :D
03:46<Peng>Mmm, stealing user data?
03:46<@Praefectus>Peng: if I told you, I'd have to kill you
03:46<Peng>God I hope that doesn't get quoted out of context by some nut.
03:46*Praefectus looks up Peng's account
03:48<Peng>Praefectus: You already know my account, from the triple-charging. :)
03:48<@Praefectus>ya, but i dont have it bookmarked, so i have to look it up when i need yer info
03:49<Peng>You should bookmark it, then.
03:49<Peng>That's what they're *for*.
03:49<@Praefectus>are you a spy?
03:50<Peng>Or do you already have so many bookmarks of other accounts that you can't fit any more?
03:51<Takyoji>Anyone know of an emailer webapp that I won't want to kill myself over?
03:52<Takyoji>PHPlist is an abomination and crime against humanity, and requires magic_quotes, and stores passwords in plaintext (and shows your current password, when you want to change it)
03:52<Takyoji>PoMMo hasn't been touched for 4 years, also requires magic_quotes, has SQL syntax errors in the installer, and even petty spelling errors in the installer..
03:53<Takyoji>I suppose that'll have to be my option.
03:53<ziph>Can you do apt-get into a file system you've mounted on /media/xvda, or would one just download the packages manually and use dpkg?
03:59<ziph>Heh, "cp /media/xvda/etc/apt/sources.list /etc/sources.list;apt-get update;apt-get download ..." works well enough.
04:03-!-Garbee [] has quit [Read error: Connection reset by peer]
04:03-!-Garbee [] has joined #linode
04:11<chesty>Praefectus: and, your government hates women
04:12<@Praefectus>chesty: again, which government
04:12-!-karstensrage [] has quit [Remote host closed the connection]
04:12<chesty>Praefectus: <@Praefectus> Peng: my family has been defending this country since WW2,
04:13<@Praefectus>chesty: so the us government hates you too?
04:14<chesty>it's pretty clear they do
04:19<ziph>The US government loves women.
04:19<ziph>You see scantily clad ones on Fox News all the time.
04:25-!-sivy [~sivy@2002:62a7:ded1:0:2d51:b5c9:3fe0:59c6] has joined #linode
04:27<meskarune>thats one reason whhy I don't watch fox news. but the mexican news station is worse :P
04:27<hawk>meskarune: Are they fully nude? Are you sure that was the right channel?
04:28<meskarune>I was in texas, half hte stations are spanish :)
04:28<meskarune>and it was def the news
04:28<meskarune>chick had on a bikini doing the weather
04:28-!-Kane` [] has quit [Quit: Leaving]
04:28<meskarune>that is pretty normal in their game shows too
04:28<chesty>it might have been hot
04:28*meskarune is a latina
04:29<chesty>I'd love to visit latin america and learn latin
04:29<ziph>Sigh, downgrading the packages doesn't fix it.
04:30<meskarune>you should. latin is an awesome language ... :P
04:30<ziph>I think he means learn a derivative of vulgar latin.
04:30<chesty>ziph: downgrading isn't supported for one, pastebin /etc/fstab
04:30<meskarune>e pluribus unum is on our money, and latin is used for scientific names
04:31<@Praefectus>they removed latin from the curriculum my freshman year of high school
04:31<chesty>are you that old?
04:31-!-hipsters_ [] has joined #linode
04:32<linbot>New news from forum: getting reports of SPAM from my vBulletin website in General Discussion <>
04:32<chesty>ziph: is that it?
04:32<ziph>chesty: Yeap.
04:32<chesty>you're missing the devramfs
04:32<Anomie>Is there anywhere I reliable I can 'rent' a sysadmin from for a few hours to look into a few issues I can't solve?
04:32<chesty>none /dev devtmpfs rw 0 0
04:32*Praefectus will rent you chesty
04:33<chesty>Anomie: what do you needs attention?
04:33<chesty>what needs attention
04:33<chesty>i rewrote that sentence part way through
04:34<Anomie>I've always had a problem with the mail on my box
04:34<Anomie>people dont receive access keys
04:34<Anomie>i dont receive contact forum submissions etc
04:34<Anomie>Ive spent hours looking into it and I'm still unsure if its a misconfiguration or spam. Either way its a pain and I need to fix it :S
04:35<chesty>i'll pass, but there's a place on linodes forum for soliciting sysadmins
04:35<Anomie>cool, thanks.
04:37<ziph>chesty: Hmm, no luck with that one. "Automount devtmpfs" was enabled in the Linode configuration too.
04:38<chesty>ziph: where you looking at the right /etc/fstab?
04:38<ziph>chesty: Yeap, I booted with init=/bin/bash to make that change, not Finnix.
04:39<@akerl>ziph: What's broken?
04:39<ziph>akerl: I get "mountall: Disconnected from Plymouth" as the last message in the kernel log and no console.
04:40<@akerl>And the above paste is the full contents of your /etc/fstab>
04:41<@akerl>:< do you have anything else odd going on? (pv-grub, custom diskiness, etc?)
04:41<ziph>Nope, it's a stock standard Ubuntu 12.04 LTS
04:42<@akerl>SpaceHobo: Most times I've had to mess with it have led me to want to stab something
04:42<ziph>Just did an apt-get upgrade today to get some security patches and it died on the reboot.
04:42<chesty>ziph: what kernel?
04:42<ziph>(initscripts, sysv-rc, sysvinit-utils, libudev0 and udev were upgraded)
04:43<ziph>Latest 3.4 (3.4.2-linode44)
04:44<@akerl>ziph: Is there anything of note in your logs, or any other errors on LISH besides the 12.04 pty/no space errors?
04:47<ziph>akerl: Nope, is the kernel log
04:47<@akerl>ziph: Pastebin `df -h` and `df -i`?
04:48<@akerl>Well, no need to pastebin them, just make sure neither looks like you're out of space/inodes
04:48<ziph>Yeah, that's the first thing I checked, but I forgot to look at inodes, let me look now.
04:49<@akerl>Most of the things that sneak up and eat your inodes are cleaned on reboot, so it's unlikely, but worth checking
04:50<ziph>(-i) /dev/xvda 4176256 54394 4121862 2% /media/xvda
04:50<ziph>(-h) /dev/xvda 39G 3.2G 35G 9% /media/xvda
04:50<chesty>ziph: it's not ext4, is it?
04:50<ziph>Nope, ext3
04:51-!-SelfishMan_ [] has joined #linode
04:51<@akerl>:< unfortunately, you've now reached the limit of my suggestions. You're probably hitting other issues that would be good to investigate, but the 12.04 logging thing masks them
04:52<ziph>Is upstart run by init, or does it replace it?
04:52<meskarune>it replaces it i believe
04:52<chesty>replaces it
04:52<ziph>So the "init: ..." messages are actually from upstart?
04:53-!-SelfishMan [] has quit [Ping timeout: 480 seconds]
04:54<ziph>Remind me to alias apt-get upgrade to something that checks if it is a Friday afternoon and exits without upgrading if it is. ;)
04:55<@jchen>akerl: go back to sleep
04:55<chesty>he was awake?
04:55<@akerl>jchen: wat
04:56<@akerl>It's time to wake up and go fishing
04:56<@akerl>not a joke; I'm getting my gear together to go fishing :)
04:58<meskarune>aw, I'm jealous
04:58-!-sbattey [] has quit [Quit: Computer has gone to sleep.]
04:58<meskarune>but I'm going up to NY again this weekend
04:59<meskarune>I have a fishing pole and some rigs in my storage thingy
04:59<chesty>ziph: stupid question, is your linode actually down after you boot it?
04:59<ziph>chesty: Yeah, it doesn't start networking.
05:00<chesty>i could see where that might be a problem
05:01<Anomie>I'm reconfiguring postfix using this guide; It says the third option is 'NONE'. This is the place where I enter the FQDN on the postfix config
05:01<Anomie>should I not enter a FQDN?
05:01<ziph>And since I have no console I can't even poke around and see the state of anything.
05:02<Anomie>also I send mail from both / - Does it matter which I enter?
05:03<hawk>Anomie: I don't know about that guide but neither of those alternatives look like they would be the fqdn
05:05<Anomie>hawk: │ This name will also be used by other programs. It should be the single, fully qualified domain name (FQDN).
05:05<Anomie> │ Thus, if a mail address on the local host is, the correct value for this option would be
05:05<Anomie>excuse the formatting :s
05:07<hawk>What is it that they are actually requesting?
05:07<hawk>(What does "this" refer to?)
05:09<chesty>ziph: do you have 0:2345:respawn:/sbin/getty 38400 hvc0 in /etc/inittab ?
05:10<hawk>Anomie: Ok, well... then I guess those would work. But the first part explains how it is used, not sure how much I can add to that explanation.
05:10<chesty>ziph: btw, those errors at the end of the boot are normal with the linode kernel. it's to do with having no initramfs to mount the pty's or something
05:10<ziph>chesty: I don't have an inittab, but I assumed that was some weird feature of upstart.
05:11<chesty>ziph: right, it's probably in /etc/init
05:11-!-TecnoBrat [] has quit [Read error: Connection reset by peer]
05:11<Anomie>hawk: Just wondering if it matters if I use or .com here?
05:11-!-TecnoBrat [] has joined #linode
05:12-!-kingdutch [~kingdutch@] has joined #linode
05:12<hawk>Anomie: Attempt att clarification: What it says is that if you are a crazy person and send mail to eg just "anomie", what should it add for a domain?
05:13<ziph>chesty: /etc/init/hvc0.conf has "exec /sbin/getty -8 38400 hvc0"
05:14<hawk>Anomie: The guide suggested not accepting that at all? (Was there such an option?)
05:15<Anomie>hawk: Well the guide isnt very clear. Im not sure if the order of the questions has changed since it was written
05:15<Anomie>but you can leave it blank yeah
05:16<ziph>I presume upstart is trying to attach stderr to some pty's so that it can log the output of jobs somewhere fancy?
05:17<hawk>Anomie: Either way it's only how to deal with email addressed to just a name with no domain (which really is an edge case that you may not want to deal with at all)
05:19<hawk>But I guess it would be appropriate to have it be the domain where your actual system users exist, if there is such a domain
05:20-!-HorizonXP [] has quit [Read error: Operation timed out]
05:22<chesty>ziph: what does /var/log/syslog look like?
05:22<ziph>chesty: There's nothing in it since I shutdown just after the upgrade.
05:24-!-Chukwumah [] has joined #linode
05:24<ziph>Is there any way to specify kernel options?
05:25<ziph>(Without going whole hog and doing a custom kernel)
05:25<chesty>i thought you could with the linode manager, but I've never needed to do it
05:26<@Praefectus>you can pick your kernel through the manager, but if you need options we dont provide you need to run your own kernel
05:27<chesty>ziph: btw, is this a 32bit install?
05:28<ziph>chesty: Yeap.
05:31<chesty>so what happens if you boot with /bin/bash then do an exec /sbin/init ?
05:31<ziph>The same error messages.
05:32<chesty>actually after you boot with /bin/bash, mount /dev
05:33<ziph>It's already mounted (
05:34<ziph>Interestingly so is /dev/pts, but upstart still bleats about it. :)
05:35<ziph>Oh, that could be wrong due to /etc/mtab being out of date.
05:36<hawk>"cat /proc/mounts" maybe? (If mount just reads mtab?)
05:40<Anomie>What should my reverse DNS be?
05:40*ziph If I manually mount /dev/pts I get: init: ureadahead main process (1049) terminated with status 5
05:40*ziph plymouthd: ply-terminal.c:630: ply_terminal_set_mode: Assertion `terminal != ((void *)0)' failed.
05:40<chesty>ziph: can you mount -o remount,rw /
05:43<ziph>chesty: Yeap.
05:44<chesty>try starting udev
05:44<chesty>/sbin/udevd --daemon
05:45<Peng>It weirds me out that chesty is a more reliable alternative to upstart.
05:45<Peng>Slower, though.
05:47<ziph>udevd[1018]: error: runtime directory '/run/udev' not writable, for now falling back to '/dev/.udev'
05:47<ziph>udevd[1019]: error disabling OOM: No such file or directory
05:47<chesty>i have no idea what I'm doing Peng, but thanks for the nod
05:47<@Praefectus>chesty: neither does upstart, so yer even
05:47<chesty>do you have a /run dir?
05:47<ziph>Yeap, but no /run/udev
05:47<chesty>something must run before udev to mount /run
05:49<ziph>It mounts /run to a ramdisk or something?
05:50<chesty>have a look in /etc/init/mount*
05:56<HoopyCat>Anomie: if your hostname is 'squid' and you have a domain '' that you want your servers, etc, to live within, '' is a good choice
05:57<ziph>Anomie: Your reverse DNS has to match your forward DNS by the way, if you don't want to get blacklisted by e-mail servers.
05:57<Anomie>HoopyCat: yeah I have it set to but when I use to do a lookup on it tells me the reverse dns doesnt match
05:57<HoopyCat>ziph: ... or if you want it to actually be set :-)
05:59<HoopyCat>Anomie: i don't see a problem... resolves to, resolves to, resolves to
05:59<HoopyCat>Anomie: looks like it matches to me
05:59<Anomie>HoopyCat: cool, ta
05:59<ziph>Even with working logging, upstart doesn't log anything useful.
06:00<HoopyCat>Anomie: when i do the "SMTP Diag" button on there, it does tell me: Warning - Reverse DNS does not match SMTP Banner
06:01<HoopyCat>Anomie: what's 'hostname -f' say? it should say
06:07<ziph>Hmm, exec /sbin/init -v gets the same as if you do --verbose as a kernel option:
06:08<HoopyCat>ziph: is also a bug you may have run into at some point as well... haven't had enough coffee to determine if that's the case tho
06:08<Anomie>HoopyCat: yeah it does
06:09<ziph>HoopyCat: Yeap, I can fix that by manually mounting /dev/pty, which then gives me the output (with -v for verbosity) that I just pastebinned.
06:09<HoopyCat>Anomie: your mail server seems to think it isn't ''
06:10<HoopyCat>ziph: cool, just wanted to make sure. (remember to 'this bug affects me' so that maybe it gets fixed before 2014)
06:11<Anomie>HoopyCat: Here's my - This is where that would be set, (right?) --
06:11<HoopyCat>When this post is 1hr old, reddit will go down for maintenance (self.announcements) (19|5) submitted 10 minutes ago by alienth [A]
06:11<Anomie>"myhostname = cravenpublishing"
06:11<Anomie>should I change that to ?
06:11<HoopyCat>Anomie: either comment out myhostname, or set it to
06:12-!-ukitazume [] has quit [Ping timeout: 480 seconds]
06:12*HoopyCat rummages around for a banana
06:16-!-ukitazume [] has joined #linode
06:17<Anomie>I'm getting this message from wordpress - Your message was sent successfully. Thanks. // But nothing at all is appearing in my mail.log :/
06:19<HoopyCat>Anomie: does this work? echo "test message blah blah" | mail -s "test message" ?
06:20<linbot>New news from forum: Hostname usage without FQDN in Linux Networking <>
06:21<Anomie>HoopyCat: Yeah, comes through fine (it wasn't last week so that change to must've fixed that - ta!)
06:21<HoopyCat>Anomie: then wordpress and/or PHP are probably broken
06:21-!-Duke [] has quit [Quit: K-lined]
06:22-!-ukitazume [] has quit [Read error: Operation timed out]
06:22-!-John [] has joined #linode
06:24<Anomie>HoopyCat: Hmm, it started happening across all my sites at the same time though. And that echo test wasnt coming through either. Now thats fixed, hopefully wordpress will follow suit
06:26<ziph>Fixed it.
06:26<chesty>ziph: what was it?
06:26<ziph>Probably dying because it was having trouble finding a frame buffer.
06:27<chesty>how did you fix it?
06:31<HoopyCat>Additional Info: CRITICAL - load average: 31.65, 27.62, 17.96
06:31<Anomie>root@server1:~# ps aux | grep sendmail
06:31<Anomie>root 29990 0.0 0.0 3096 728 pts/0 S+ 10:31 0:00 grep sendmail
06:31<ziph>Removed --attach-to-session from /etc/init/plymouth.conf; which doesn't make a lot of sense. I'm checking the plymouth source now.
06:31<Anomie>Does that indicate sendmail is running fine?
06:32<ziph>Is sendmail ever running fine?
06:32<HoopyCat>Anomie: no, it's not running (the only thing matching 'sendmail' is your grep command). also, i thought you said you were running postfix?
06:32<Anomie>no :(
06:32<Anomie>HoopyCat: Yeah I am, wasn't sure if sendmail was a seperate MTA or something I needed in addition
06:33<HoopyCat>Anomie: Sendmail-brand sendmail is a different MTA; you're using Postfix-brand sendmail. Sendmail-brand sendmail is mostly used by folks who don't understand the difference :-) postfix should work just fine.
06:34<chesty>ziph: i don't think that's the root cause, that flag is in the linode ubuntu image
06:34<ziph>chesty: Yeah, neither do I.
06:35<ziph>chesty: What version of plymouth does it have in the image?
06:36<HoopyCat>golly, never have i ever seen iowait hit 99%/
06:37<HoopyCat>(is not a linode; 95% of the time, when i'm bitching about problems, it's unrelated to linode.)
06:38-!-ukitazume [] has joined #linode
06:45<Anomie>Any more information I could provide in this post to make it easier? :
06:45<Anomie>*easier for people to determine whats wrong
06:45<ziph>Hmm, spoke too soon.
06:46<ziph>Switched it to init=/sbin/init and it doesn't work. :)
06:46<ziph>init: plymouth main process (1024) terminated with status 69
06:47<ziph>It's hard not to be annoyed that some wanky splash screen is stopping a headless server from running.
06:47<linbot>New news from forum: Wordpress not sending out emails. in Email/SMTP Related Forum <>
06:53<Anomie>I just installed a wordpress plugin
06:53<Anomie>and it fixed everything
06:53<Anomie>But it now sends all mail over SMTP - anyone wanna tell me why thats probably stupid to do? lol
06:53-!-danp [] has quit [Ping timeout: 480 seconds]
06:53-!-Y_Ichiro [] has quit [Read error: Connection reset by peer]
06:53-!-Y_Ichiro [] has joined #linode
06:53-!-mkasu [] has quit [Ping timeout: 480 seconds]
06:53-!-danp [] has joined #linode
06:53-!-mkasu [] has joined #linode
06:53<meskarune>do you have an issue with smtp?
06:54-!-zoxie [] has joined #linode
06:54-!-zox [] has quit [Read error: Connection reset by peer]
06:54-!-zoxie is now known as zox
06:54-!-branko_ [] has quit [Read error: Connection reset by peer]
06:54-!-stdaro [~sroylance@] has joined #linode
06:54-!-branko [] has joined #linode
06:54<Anomie>meskarune: Nope, but I know very little (compared to others in here). One of those 'too good to be true' fixes, I suppose.
06:54-!-sroylance [~sroylance@] has quit [Read error: Connection reset by peer]
06:54-!-ukitazume [] has quit [Ping timeout: 480 seconds]
06:55-!-madbytes [~madbytes@] has joined #linode
06:55<rnowak>it will be slower than using your local machine's mail queuer
06:56<rnowak>and you lose the log entry that a certain email was queued and/or sent
06:56<HoopyCat>Anomie: that's probably totally fine. it'll work, that's for sure.
06:56<HoopyCat>rnowak: it's not like PHP was returning a fault when it wasn't queued and/or sent in the first place
06:57<ziph>dpkg --force-depends -P plymouth
06:57<ziph>fixes that little problem.
06:57<Anomie>HoopyCat: rnowak: I can live with that. woo
06:58<ziph>chesty, akerl, thanks for your help!
06:59<Anomie>I'm getting about 1 email per minute via my contact form though now
06:59<Anomie>all blank
06:59<Anomie>all weird, spammy email addresses
06:59<Anomie>not sure whats going on
06:59<HoopyCat>Anomie: ok! sounds like everything is working normally
07:01<Anomie>Think its just people setting spam bots to enter competitions actually
07:01<chesty>ziph: you uninstalled plymouth?
07:02<ziph>I broke it (by moving the binary) and it booted.
07:02<ziph>Removing the package stops it from booting.
07:02<ziph>Despite other people reporting success with doing so.
07:02<ziph>Does debian still ask a bunch of stupid questions for every package you install? ;)
07:03<chesty>debian is the one true distro
07:03<ziph>I haven't bothered with it since I watched someone spend over an hour answering questions for each package it was installing.
07:04<ziph>That someone thought that was a good idea showed a bit of a lack of taste.
07:04-!-steveg [] has joined #linode
07:04<HoopyCat>ziph: linode has already gone through the installer, in any case
07:05<chesty>they probably had question priority on low, so it asks everything
07:05<chesty>but most packages install with no questions
07:05<ziph>This was over a decade ago, by the way.
07:05<rnowak>and the questions that get asked should be in post install, so at worst you'd screw up the package and nothing else
07:05<ziph>I'm assuming that it hasn't improved in that time. ;) ;) ;)
07:06<linbot>Today is Debian Appreciation Day! \o/
07:06<linbot>HoopyCat: Point (0.06742550, 0.22884444) lies within the unit circle. Hits: 512 of 647 (π ≈ 3.165378670788253 - 0.023786017198460). http://π
07:06<rnowak>good job linbot
07:06<chesty>linbot is slow tonight
07:06<chesty>waiting for the is is not a command message
07:06-!-D[a]rkbeholder [] has joined #linode
07:06<chesty>got it
07:07<HoopyCat>linbot was blocking on network I/O for !debian
07:07<rnowak>too much awesome
07:07<Peng>It checks a web server to verify how much Debian should be appreciated today?
07:08<chesty>it checks the calendar
07:08<rnowak>Peng: it actually runs a differential equation weighing in the tears of all arch users
07:08<linbot>New news from forum: getting reports of SPAM from my vBulletin website in General Discussion <>
07:08-!-Anomie [] has left #linode []
07:10<HoopyCat>it uses the Wronskian of the tears of all arch users and the deviation function for ubuntu as the determinant for whether or not it is Debian appreciation day
07:11<chesty>dude, it simply looks up the calendar over ical
07:12-!-darkbeholder [] has quit [Ping timeout: 480 seconds]
07:12<rnowak>it only does that to make sure we still have days in the calendar, so that it can call it an appreciation day, instead of an appreciation moment in time in space
07:13-!-DephNet[Paul] [] has joined #linode
07:13<HoopyCat>yeah, which is itself generated by cloning onto a new linode, booting it, and then consulting it for the answer
07:14<HoopyCat>"She was bicycling with a friend on the shoulder of the road just south of Bainbridge Lane around 7:45 a.m. July 29 when she was struck by an alleged reckless motorcyclist and thrown into the path of an alleged drunken driver."
07:15<HoopyCat>thank god there's no bus service out there
07:18<meskarune>sounds like my daily commute
07:18<meskarune>NJ drivers FFFFUUUUUUUuuuu
07:18-!-mcinerney [] has joined #linode
07:19<@Praefectus>only place thats worse that ive been was TX
07:20<meskarune>I just came from texas
07:20<meskarune>I have to disagree
07:20<meskarune>unless you mean houston
07:20<meskarune>which is its own thing
07:20<@Praefectus>thats exactly where i mean, i was there once or twice a month for nearly 2 years
07:20<meskarune>I lived in houston, then moved to dallas :)
07:20<meskarune>yeah, houston drivers are nuts
07:21<Peng>Why Houston?
07:21<meskarune>but houston is nothing like the rest of texas
07:21<@Praefectus>amarillo was quiet
07:21<Peng>You were in Dallas? Did you say hi to the Linode servers?
07:21<ziph>chesty: It seems to work only if I manually rw remount /
07:21<Peng><- only values cities by data centers
07:21<ziph>chesty: (It's hanging just after starting up-start-udev-bridge now)
07:22<meskarune>I did not peng :P
07:22<meskarune>I did say hi to lots of mexican food though
07:22<Peng>What kind of weird being values food over servers? :(
07:22<@Praefectus>i wanna go to houston just to stop by goode & co again
07:23<meskarune>my family is still trying ot sell a house in houston
07:24<meskarune>got hit with 2 hurricans so they had to do lots of fixing to get it ready to sell
07:24<HoopyCat>wow... xbox support is giving linode support a run for their money. 4 minutes:
07:25-!-dubenstein [~dubenstei@] has quit [Quit: leaving]
07:26<path>why do they have xboxsupport[1-5]
07:26<rnowak>1-12 actually
07:26<HoopyCat>probably API limits
07:27<path>seems kinda hacky
07:27-!-skcin7 [] has quit [Quit: Computer has gone to sleep.]
07:29<HoopyCat>path: ... "if you’re unaware, we use a number of overflow accounts to tweet from when we’ve reached our primary account’s limit of tweets for the day"
07:30<path>is this something twitter automagically does, or do their reps have multiple accounts configured in their clients?
07:30<path>"oh shoot, we hit the limit.. time to switch to lamer2"
07:30<rnowak>I'd imagine the latter... otherwise twitter could just allow them more tweets per account?
07:31<HoopyCat> <--- the client they appear to be using
07:31-!-Bryen [~bryen@] has quit [Ping timeout: 480 seconds]
07:31<rnowak>looks very... enterprise
07:32<path>ooo.. power point slides in a webpage..
07:32<Peng>MS uses something that's not in-house?
07:32<HoopyCat>i suspect it's a client feature... tweet gets rejected due to limiting -> switch to the overflow
07:32<HoopyCat>the specific product: (it is enterprise-class)
07:33<path>i think i hit the limit of time allowed in my house.. time to overflow to work.. later
07:33<rnowak>"To schedule a personalized demo of Attensity Respond, contact Attensity today.", let's do this
07:34<HoopyCat>Ha! Ha! I have Fridays off.
07:35<HoopyCat>i still have to wake up at 5:15am to transport the spouse to work so i can have the car for the day's tasks, tho
07:42<linbot>New news from forum: Install Virtual Server on a Linode - Is this possible in General Discussion <>
07:46*Kyh_ installs qemu then installs windows then instals sheepshaver
07:46<@jchen>openvz on xen
07:46<@jchen>it's been done before
07:46<Peng>Can you qemu on qemu?
07:47<@jchen>yo dawg i heard you like virtualization
07:47<HoopyCat>so has urmom, but it doesn't mean it is a good idea
07:47<@jchen>urmom is always a good idea
07:47<Peng>Virtualization is always an even better idea.
07:48<ziph>try_mount: / waiting for device
07:49*ziph ponders FreeBSD hosting.
07:50<HoopyCat>indeed, back before the tax evasion situation, was running on a linode running openvz, because he liked virtualization
07:52<HoopyCat>ziph: disregard ubuntu, acquire debian. really -- and this might just be the coffee talking -- 12.04 is a pile of shit if you aren't on an octo-core tablet with 16 GB of RAM
07:52<ziph>I would, but this thing runs all kinds of things that I only just finished setting up. :)
07:53<@Praefectus>so it will be easier doing it a second time
07:53<HoopyCat>ziph: that's why i say debian... all the "do-stuff" things are pretty much the same between the two
07:53<ziph>Not really, the bit that slows me down is all the silly differences between distributions.
07:53<HoopyCat>ziph: that's why i say debian
07:54<ziph>The fs layout is the same?
07:54<ziph>(And the packages?)
07:54<Peng>...The fs layout is always mostly the same on Linux.
07:54<Peng>ziph: Ubuntu is a derivative of Debian. They're quite similar.
07:54<HoopyCat>Peng: well, some distros are always mostly the same than others
07:54<ziph>That hasn't been my experience. :)
07:55<ziph>Things like different ways of chopping up the apache configuration files drives me nuts.
07:55<Peng>Yeah, Apache. That's why I said "mostly". :X
07:56<HoopyCat>ziph: the sites-available / sites-enabled thing is inherited from debian
07:56<ziph>It also doesn't help that I'm old, and have been playing with the stuff long enough that I no longer care about the details and just want things to work. ;)
07:57<HoopyCat>see, i love ubuntu on the desktop for precisely that reason. but it's shit like this upstart "fixing bugs is hard :-/" thing
07:57<ziph>I have a feeling that I'm going to have to kludge in something to get this working and then move everything to a Debian VM down the road to clean it up.
07:58<ziph>It's actually mountall hanging on udev I think.
07:58<ziph>Which might make sense since apt-get upgrade also upgraded libudev
07:59<ziph>But then I guess mountall is part of upstart.
07:59<ziph>(Or at least has been patched to talk to it.)
08:00<HoopyCat>part of the reason i like my Real Job(tm) is because i get to find edge-case bugs like this, AND THEY GET FIXED
08:00<HoopyCat>needless to say, it's not a linux distro
08:00<ziph>Part of the reason I avoid having IT as a job is because I get to find edge-case bugs like this and have vendors ignore them when reported. ;)
08:01-!-Cromulent [] has joined #linode
08:01<HoopyCat>ziph: my job is also not IT-related :-)
08:01<HoopyCat>well, ok, so it involves technologies relating to information
08:01<ziph>What is it?
08:03<HoopyCat>ziph: adaptive networking wideband waveform PHY lab technomancer
08:04<Peng>How small's the print on your business card?
08:04<ziph>HoopyCat: You're an EE?
08:04<marius>13pt, but the business card it self is as big as his overcoat
08:05<HoopyCat>Peng: fortunately, we have a lot of acronyms
08:05<marius>HoopyCat works in ANW^2PHYLT
08:06<HoopyCat>ziph: used to be in IT (particularly IP network engineering), now in EE. still do sysadmin for fun tho :-)
08:06<HoopyCat>marius: we just call it ANW2
08:06<marius>How boring
08:08<HoopyCat>marius: srsly, i actually had to google to remember which order the Ws came in
08:09<ziph>HoopyCat: You've probably got some of the toys I regularly lust over then. ;)
08:09<marius>technomancer must be the most awesome part of atitle ever
08:09<ziph>HoopyCat: Ever deal with the EM modelling side of things?
08:09<marius>I immediately picture osme guy in a heavy cloak protecting the city from technological evil masterminds
08:10<Peng>marius: Hey, Chief Lizard Wrangler is close.
08:10<meskarune>I think of a person who magically conjures tech
08:10*Praefectus thinks of Babylon5
08:10<marius>magically counouring tech makes no sense
08:10<Peng>I think of Kingdom of Loathing. And what meskarune said. :P
08:10<marius>if you do magic yo udon't need tech!
08:10-!-ukitazume [] has joined #linode
08:10<marius>wow, does KoL still exist?
08:10<@Praefectus>Peng: i havent played that in years lol
08:11<Peng>marius: No idea.
08:11<HoopyCat>ziph: the tektronix rep was around a few weeks back to demo a new scope, and got all wide-eyed when we casually mentioned we had an RSA6000 and really liked it
08:11<meskarune>a necromancer makes the dead live and kills the living
08:11<meskarune>a technomancer could also create techno music
08:11<ziph>HoopyCat: Because he realised you could afford the commissions that were buying him a new boat?
08:11<meskarune>like, daft punk
08:12<ziph>HoopyCat: What do you think of the MDO4000? Seems like the kind of thing that would be perfect for your application.
08:12<Peng>marius: Magically conjuring tech makes sense if it's the most convenient way to accomplish something.
08:12<marius>nothing is more convenient than conjuring a giant meteor to crush the world.
08:13<meskarune>I guess mancer/mancy = prophet or phophesize
08:13<meskarune>(probably spelled that wrong)
08:13<Peng>marius: I'm pretty sure that's *not* the most convenient way of making pizza dough.
08:14<marius>Are you sure? flattest pizza dough EVER
08:14<HoopyCat>ziph: i don't directly deal with EM modeling, although i'm pretty sure we have folks who do. much of my day is spent wrangling spaghetti in/out of a channel emulator and seeing how things react
08:14<Peng>marius: Yes, but when your pizza oven and the rest of the planet explode, how are you gonna finish the pizza?
08:15<Peng>Chief Spaghetti Wrangler?
08:15*Peng imagines someone fighting to control spaghetti the size of an alligator.
08:15<marius>the meteror is hot when it hits the athmospehere, I'm sure that'll bake it for you
08:15<HoopyCat>ziph: the MDO4000 is actually the scope he was demoing :-) i do think it is nifty. we can do a lot of what it does with what we have (thanks PyVISA and external triggers), but having it all in one box, with a handle...
08:16*Praefectus replaces marius' pizza oven with a chef-in-the-box
08:16<Peng>marius: So you put the toppings on before flattening the dough?
08:16<marius>Of course
08:16<marius>gets the flavor realyl deep in there
08:17<Peng>This is sounding inconvenient.
08:17<ziph>I think it died because the udev upgrade triggers weren't running properly.
08:18<HoopyCat>ziph: speaking of boxes with handles, rest assured that linode's servers have handles on them, so you can visualize yourself defenestrating one of them :-)
08:18<HoopyCat>the handles are small, tho, so don't get your fingers stuck
08:18<ziph>Nah, couldn't do that.
08:19<ziph>Any other VM provider and I would've had to just lodge a ticket and wait two weeks for them to tell me that they're reinstalling the VM and that they hope I had backups.
08:19<ziph>One of my clients just waited 2 months for some DNS records to be changed.
08:19-!-mcinerney [] has quit [Read error: Connection reset by peer]
08:20<HoopyCat>Area man plummets to death from 86th floor: witnesses describe 'terrible mess' as man frantically attempted to use a rackmount server as a glider
08:20<hawk>ziph: And they accept that?
08:20<ziph>hawk: They're using Australia's first and best Domain Name registrar. :)
08:20<ziph>"best", I mean.
08:21-!-stas [~stas@] has joined #linode
08:21<hawk>ziph: Who is also providing dns hosting services? Or were they changing the delegation specifically?
08:22<stas>hi, I'm having troubles with apt-cache server in dallas, is it still in production?
08:23*stas pings JshWright
08:23<ziph>DNS hosting with a company purchased by Melbourne IT I think.
08:24<ziph> is a very special place.
08:24<ziph>It's only fairly recent that you can pick up a domain for less than $200.
08:25<ziph>Excluding tax..
08:26<HoopyCat>stas: i know he's been busy the last few whiles with a project, but it's probably worth dropping him an e-mail
08:26<HoopyCat>speaking of which, off to the bathysphere
08:27<stas>hmm, didn't want to bother anybody, but thanks
08:34-!-BarkerJr [BarkerJr@2002:1802:e05a:1:2c0f:dc91:cee5:5193] has quit []
08:36-!-ashaffer [] has joined #linode
08:40-!-minerale_ [~minerale@] has quit [Ping timeout: 480 seconds]
08:44-!-minerale [~minerale@] has joined #linode
08:44-!-Bass10 [] has joined #linode
08:44-!-ukitazume [] has quit [Ping timeout: 480 seconds]
08:44<kingdutch>Google is telling me someone signed in to my account. The time and location match my linode and it's logs (down to the second), but it's a different IP than Linode says my box has, is that possible?
08:44-!-pyruvate [] has joined #linode
08:45<kingdutch>I have tried to get fetchmail to sign in to my gmail yesterday so I think it's legit, but the IP being different puzzles me.
08:45<chesty>kingdutch: was it your phone?
08:46<chesty>btw, two factor all the things
08:46<Kyh_>kingdutch: whats your home/etc IP/ISP vs your linode IP?
08:46-!-John [] has quit [Quit: John]
08:47<kingdutch>chesty: Nope, because the location was UK and I'm in the NL
08:47<kingdutch>My home is 186.*.*.* and linode is 176.*.*.*
08:47<kingdutch>Reported is 217.*.*.*
08:47<chesty>kingdutch: whois the ip
08:47<kingdutch>good idea
08:48-!-MrGeneral [] has quit [Quit: Saindo]
08:50<Kyh_>wtf 224
08:50<Kyh_>what are the actual IPs?
08:50<hawk>kingdutch: You said 217 here but the pastebin seems to be for 227?
08:50<kingdutch>sorry yeah
08:51<kingdutch>did that from memory, I was 1 digit off :$
08:51<hawk>kingdutch: Which one was off?
08:51<kingdutch>It's 227 instead of the 217 I said
08:51<hawk>Ok, so multicast?
08:51<kingdutch>The IP in whois is what google's telling me
08:53<kingdutch>The WhoisIp doesn't give me any more details, still think it's legit though
08:53<hawk>The whois data and I both say that it's an IP reserved for multicast use
08:54<hawk>What exactly are you saying it was used for?
08:57<kingdutch>I signed in using fetchmail from my linode box
08:57<kingdutch>At that exact timestamp
08:57<kingdutch>From the given location
08:57<kingdutch>I just don't know why google says a diff IP than my linode box has
08:58<hawk>Sounds absurd that it would show a multicast IP
08:58<Peng>Sure you don't want That seems to be...Credit Suisse.
08:59<linbot>New news from forum: Citadel permission denied in Email/SMTP Related Forum <>
09:01-!-smed_ [] has joined #linode
09:03-!-John [] has joined #linode
09:03-!-kingdutch [~kingdutch@] has quit [Quit: kingdutch]
09:03-!-kingdutch [~kingdutch@] has joined #linode
09:04<kingdutch>Hit the X instead of anywhere else on the tab to show me chat >.>
09:04<kingdutch>Anyway google doesn't guarantee that domain/ip info is accurate, so since location and time is correct with what I was doing yesterday I just gave google the green light : )
09:05*rnowak blink
09:06<hawk>What is "location" based on, then?
09:06<kingdutch>Hmmm you're right, probably the IP
09:06<hawk>If the IP is wildly inaccurate
09:07<kingdutch>Still, I somehow doubt that someone would succesfully log-in to my gmail account at the exact same time that my fetchmail does it
09:07<hawk>I kind of doubt that someone would be able to connect from that address at all
09:07<rnowak>'.'.join(str(randint(0,255)) for n in range(4)) problem users? (:
09:08-!-linville [] has joined #linode
09:08<kingdutch>hawk: No clue, I would hardly call myself a linux sysadmin, I'm not going to say I have any knowledge of networking, the ip-protocol and other related protocols :')
09:09<kingdutch>rnowak: Quite a big problem, haha
09:09<kingdutch>Anway, coffee time, thanks for the help : ) bbl
09:11<rnowak>'.'.join(str(randint(0,256)) for n in range(4)) rev2, commit message: rtfm
09:15-!-skcin7 [] has joined #linode
09:16-!-Alan_ [] has quit [Ping timeout: 480 seconds]
09:17<Garbee>Has anyone setup SPDY on their Linode?
09:18<rnowak>SPDY is an application layer protocol
09:18-!-pyruvate [] has quit [Quit: leaving]
09:18<Kyh_>there's an apache module, I think maybe an nginx one as wel?
09:19<rnowak>there is a patch for nginx
09:21-!-skcin7 [] has quit []
09:21-!-Finlay [] has joined #linode
09:22<Finlay>g'morn! I woke up to mysql crashed and 100% swap, but only 30% mem usage on ubuntu 12.04 /var/log/mysql .log and .err are completely empty. Where can I look for the cause next?
09:23<Finlay>also, the .err being completely empty is confusing . grrr
09:23<hawk>Finlay: fun OOM messages in dmesg?
09:23<Kyh_>Finlay: whats your disk usage like? ( as in apce)
09:23<rnowak>if mysql got sniped by the oomkiller, it wouldn't be given a chance to write an error entry
09:28<Finlay>hawk: if I did a reboot would that clear all my dmesg?
09:29<Finlay>kyh_ dunno what apce is, but my usage on / is 13% and I have a 249M swap
09:29<Finlay>mowak: what's the oomkiller?
09:29<rnowak>!to Finlay mowak
09:29<linbot>Finlay: mowak is not a nick. Please see
09:30<rnowak>and the oomkiller is a code path that gets triggered when the machine is running out of memory, and linux wants to save itself, killing other applications
09:30<Finlay>rnowak: sorry, this web client font is non-ideal.
09:30<Finlay>I did a dmesg | grep memory and only got a message about freeing unused kernel memory
09:31<Finlay>but this is post-reboot
09:31<rnowak>dmesg is a circular buffer, old entries get overwritten by new -- the bootup process probably managed to do that
09:32<rnowak>but if you had full swap, and mysql dead, it probably was an oom situation
09:32-!-Bullrush [] has left #linode []
09:33-!-Bullrush [] has joined #linode
09:33<linbot>New news from forum: Redis on its own Linode? in General Discussion <>
09:35<rnowak>5-10 requests per second, don't stress it out (:
09:37<hawk>Finlay: Aha, if it's post-reboot dmesg(1) won't be of any use... what about /var/log/messages or /var/log/kern.log or something like that?
09:37<auraka>maybe I'm totally using this wrong but amazon ebs with dedicated iops sucks
09:37-!-Bullrush [] has left #linode []
09:37<Garbee>rnowak, Thanks for the info. I'm looking into setting it up soon once we get SSL for a site I'm working on.
09:37-!-Dokujisan [] has quit [Read error: Connection reset by peer]
09:38-!-Dokujisan [] has joined #linode
09:39<Finlay>hawk: yeah I checked /var/log/messages being debian but it doesn't exist on ubuntu. kern.log does though, and boy, is it full of OOM messages!
09:39<Finlay>including a OOM: kill process mysqld
09:39-!-seanh-ansca [] has joined #linode
09:40<Finlay>so that answers that. I searched the linode library for a mysql memory guide for linode
09:40<Finlay>so going to try to set those now
09:40-!-storrgie [] has joined #linode
09:41<rnowak> run it, do what it says, and keep running it now and then -- should help a bit if you do not know how to configure it specifically for your workloads
09:42<hawk>Finlay: There's your confirmation, then...
09:42-!-seanh-ansca [] has quit []
09:43-!-_j45 [] has joined #linode
09:48<linbot>New news from forum: Is redis needed for this? in Web Servers and Web App Development <>
09:51-!-userme [] has joined #linode
09:53<Finlay>aw crap, I had already set the low mem settings from the linode-library. What now?
09:54<Finlay>eeek sitting at 11mb free right now
09:54<rnowak>have you optimized the schemas and queries? if you have, time to upgrade the server, or configure mysql to use less memory
09:54<Peng>Finlay: 11 MB free?
09:54<Finlay>I followed the low memory guide on linode-library
09:55<Finlay>and I run wordpress
09:55-!-E1ven [] has joined #linode
09:55-!-Bry8Star-{GB [] has quit [Quit: Quitting (For IRC Bouncer/BNC/Reflector service, visit GeekBouncer -]
09:56<Finlay>Peng: thank you. now I know how to read free and I have 212 free.
09:57<Finlay>but yeah I'm running nginx+php-fpm+mysqld+bind9
09:57<Finlay>on my linode 512
09:57<rnowak>how many php workers are you running?
09:57<rnowak>and how much memory is each allowed to allocate?
09:58-!-sivy [~sivy@2002:62a7:ded1:0:2d51:b5c9:3fe0:59c6] has quit [Remote host closed the connection]
09:58-!-sivy [] has joined #linode
09:59<Finlay>sec doing research, no idea how to find out currently. but ps waxu | grep php-fpm gave me > 40 processes
09:59<Finlay>actually, it gave me 148 lines
09:59<Finlay>now that I added wc -l
10:00<rnowak>you'll need to look at the configuration which sits somewhere like /etc/php*
10:00<Finlay>yeah I already looked in /etc/php5/fpm/php-fpm.conf
10:00<Finlay>but the whole file is comments really
10:01<rnowak>is there no pool.d configuration directory?
10:01<Finlay>I see a 'process.max' with a no limit default
10:01-!-R5Chris [] has joined #linode
10:01<R5Chris>morning, folks
10:01<hawk>Finlay: that sounds less than ideal
10:02<Finlay>the line to uncomment has a max of 128
10:02<rnowak>try a max of about 4
10:02<Finlay>but this is total processes to fork
10:02<Finlay>and there is a big warning on it saying use with caution
10:02<Finlay>so I'm looking if maybe the configuration is with my nginx files?
10:03-!-userme [] has quit [Ping timeout: 480 seconds]
10:03<rnowak>max_children is what you want to configure
10:03<Finlay>yeah fastcgi_params isn't really what I want
10:04<Finlay>there is a pool.d directory yes
10:04<rnowak>look for files in it
10:04<Finlay>lots of files there
10:04<Finlay>max_children in each is 50
10:04<Finlay>start servers 20
10:04<Finlay>min spare 5
10:04<Finlay>max spare 35
10:05<Finlay>sorry I should of dpasted that
10:05-!-endzyme [] has quit [Ping timeout: 480 seconds]
10:05<@heckman>That's a lot of servers ...
10:05<Finlay>there is a file there for each of my virtualhosts
10:05<@heckman>Sound right.
10:05<rnowak>it does?
10:06<@heckman>If each VirtualHost has different users, or if you expect different resources required, kind of makes sense to do it that way
10:06<Finlay>my fastcgi has a 'children' setting of 8
10:06<rnowak>yeah, but it doesn't for him on a single 512 linode
10:07<@heckman>Well, on a resource standpoint no. :P
10:08<rnowak>you will need to ensure that you're not running more than 4 php workers, and that's probably too much for you as well, if we say that each is allowed 100MB RAM each
10:08<Finlay>k googling for some better settings now
10:08<rnowak>and wordpress is not light (:
10:08-!-endzyme [] has joined #linode
10:11<Finlay>hm I have a feeling this is going to be frustrating, because I have a seperate pool.d file for each virtualhost
10:11<Finlay>I can configure each but I can't really do something for overall
10:11<rnowak>do the pools have security separation of sorts? different user?
10:11<rnowak>time to upgrade the linode then, or get more of them
10:12<Finlay>I have a few friends with domains and stuff, and they all can configure their own dns / websites
10:12<Finlay>nothing major
10:12<Finlay>it was just convenient so they could do their own stuff for their personnal blogs
10:12<rnowak>running like you do right now, you will oom as soon as enough requests come in
10:12-!-Cydus [] has joined #linode
10:12<Finlay>I tried webmin but it was way too complicated for my skill level to set up, and way too complicated for my friends to use
10:12-!-skvare [] has joined #linode
10:13<rnowak>it wouldn't have helped
10:13<Finlay>so I am using ispconfig to let them manage their stuff right now
10:13<Finlay>it is the one generating the pool.d files
10:13-!-skvare [] has left #linode []
10:13<rnowak>you're just requiring more resources for the way you've set things up than are available to you
10:13<Finlay>I can tweak the settings per website though
10:13<Finlay> seems to have some settings, I can try those, but on a per site basis.
10:14*rnowak hugs his uWSGI
10:15<hawk>Finlay: How many sites are there?
10:16<Finlay>hawk: that get >50 uvisitors a day? 4?
10:16<hawk>That may get any request at all at any given time
10:16<Finlay>it's really just a mess around box, with small little personnal blogs
10:17<rnowak>do your users have a mutual trust?
10:17<hawk>Finlay: I mean, if all of your sites were to get a single request at the same time, would you oom?
10:18-!-David [] has joined #linode
10:18-!-Cydus [] has quit [Read error: Connection reset by peer]
10:18<Finlay>hawk: ugh, I dunno?
10:18<hawk>Finlay: What is the memory limit for php set to?
10:19<Finlay>rnowak: I'd prefer not to have to deal with that. but the biggest problem was even if I gave them access to their own virtualhost files, they couldn't create new ones, and they couldn't restart the services... so giving them a GUI was the simplest way
10:19<Finlay>maybe I should not be using nginx + php-fpm, and go back to apache2+mod_php?
10:19*rnowak sighs
10:20<Finlay>same with their bind9 files
10:20<Finlay>not that I would trust them not to introduce errors in any of said files and have the service not come back up
10:20<rnowak>so you're running DNS for them on a single machine?
10:20<Finlay>yes I run bind9 on the same box, but we don't serve queries from it
10:20<Finlay>I use it as a unlisted master
10:20<Finlay>and I use linode's dns
10:20<Finlay>in the actual domain NS records
10:21-!-stafamus [] has joined #linode
10:21<Finlay>this meant I didn't need to create them linode accounts etc for the dns manager
10:21<Finlay>and they can use ispconfig for dns and www setup
10:22<Finlay>okay I changed all the specific site settings to be lower, and restarted the services, hopefully that will be good
10:22<Finlay>but I see what you mean that if that's the max settings for the box... if multiple sites get hit... I'm in trouble
10:22<rnowak>not if, when
10:22<Finlay>it's really sounding like maybe I should go back to apache2+mod_php
10:22<Finlay>unless there is a non fpm I can use with nginx? I'm not sure
10:22<rnowak>what did you use to ensure your security separation when you ran mod_php?
10:22<Finlay>I just recently switched
10:23<Finlay>to try and get some experience with it
10:23<Finlay>boy, did I >_<
10:25<linbot>New news from forum: Adding an 'O-Line' to ircd.conf in General Discussion <> || Clamscan whole disk or certain places? in General Discussion <>
10:26<Finlay>thanks for all the guidance / help as usual #linode <3
10:27-!-Cromulent [] has quit [Quit: WeeChat 0.3.7]
10:29-!-dubenstein [~dubenstei@] has joined #linode
10:30<linbot>New news from forum: getting reports of SPAM from my vBulletin website in General Discussion <>
10:30<Finlay>rnowak: probably my best course of action then, neh?
10:31-!-hfb [] has quit [Quit: Leaving]
10:32<rnowak>it doesn't change anything, so no
10:33<Finlay>rnowak: wouldn't apache2+mod_php at least use a single set of conf parameters?
10:34<rnowak>you could just as well alter php-fpm then
10:34<Finlay>maybe I can override the pm settings AFTER I include the pool.d stuff....
10:34<Finlay>my problem is each vhost sets it's own pm settings
10:34<Finlay>yes that is what I will do
10:34<rnowak>your problem is using a crappy "control panel"
10:35<Finlay>rnowak: what's a better alternative though?
10:35<rnowak>not using a crappy control panel
10:35<Finlay>is always my question when people cringe when I say I am using ispconfig
10:35<Finlay>everyone cringes and yet nobody has a better solution
10:36<Finlay>a multi-user setup is just not manageable without it.
10:36<rnowak>a better solution? not using a crappy control panel
10:36<rnowak>I already gave you one
10:36<Finlay>and the only decent/popular alternative I see out there is webmin which is sooo sysadmin heavy that I would never trust it my users
10:37<rnowak>you could stop trying to be another webhost, there's already thousands of them out there
10:37<Finlay>rnowak: that isn't a solution. suggesting how I could provide the same service without using a control panel... that would help!
10:38<rnowak>the same service? a service that ooms? quite easily, actually
10:38-!-eyepulp [] has joined #linode
10:38<Finlay>rnowak: you have a point there, but then the point of this linode was just to mess around and learn things especially wrt linux
10:38<rnowak>I hope you're at least honest to your users, that you have little idea what you're doing
10:38<Finlay>rnowak: if I had a real production website that had any real traffic I would find myself a webhost. and yes very honest.
10:38<Finlay>and yes, on the experience front as well
10:39-!-HorizonXP [] has joined #linode
10:39<Finlay>I have made it clear they are much better off paying 5-10$/mo for a webhost or somesuch
10:39<Finlay>for anything they really care about
10:40<Finlay>to them this is a playground too to mess around with website and php and stuff
10:40<Finlay>have a shell with screen and irssi
10:45-!-DrJ [] has quit [Read error: Connection reset by peer]
10:45-!-DrJ [] has joined #linode
10:50-!-Gnintendo [] has joined #linode
10:53-!-stafamus [] has quit [Ping timeout: 480 seconds]
10:54-!-fisted [] has quit [Quit: leaving]
10:55-!-mpkossen [~mpkossen@] has quit [Quit: Leaving.]
10:55-!-fisted [] has joined #linode
11:00-!-Austin__ [~austin@] has joined #linode
11:00-!-MrGlass [] has quit [Ping timeout: 480 seconds]
11:00-!-MrGlass [] has joined #linode
11:03-!-seanh-ansca [] has joined #linode
11:04-!-_j45 [] has quit [Read error: Connection reset by peer]
11:04-!-_j45 [] has joined #linode
11:13-!-kik [] has joined #linode
11:13-!-hfb [] has joined #linode
11:13-!-hfb_ [] has joined #linode
11:13-!-hfb_ [] has quit [Remote host closed the connection]
11:13-!-Finlay [] has quit [Remote host closed the connection]
11:14-!-kik [] has quit [Remote host closed the connection]
11:14*EugeneKay gives mowak a beer
11:15-!-Michael [] has joined #linode
11:16-!-Michael is now known as Guest1729
11:17<Guest1729>anyone have experience with game servers on a linode?
11:17<Peng>Some people do. Not me.
11:18<Guest1729>Is it ok?
11:18<Guest1729>to host them
11:18<Guest1729>I assume they would be classified as noisy neighbors
11:18<Guest1729>a few hundred probs and a garrys mod server is chewing through cpu
11:18<meskarune>well, if you expect to be using a ton of system resources, size your plan accordingly
11:19<Guest1729>Im unsure about the cpu capacity
11:19<Guest1729>I can look at the ram
11:19<Guest1729>but what about the cpu?
11:21<bss>Guest1729: i run l4d2 and hl2dm servers on a 1024 without problems
11:21<bss>only rarely at the same time, but still
11:21-!-bchristensen [~bchristen@] has joined #linode
11:22<Guest1729>my current dedicated is
11:22<Guest1729> DUAL (2) Intel Xeon 3.4GHz 800MHz 1MB Cache
11:22<Guest1729>in the cpu
11:22<Guest1729>how does a 512 compare?
11:22<XReaper>you get 4 virt cpu's
11:22<XReaper>on all linodes
11:23<XReaper>that gets shared with everyone else on your host box
11:23<XReaper>but normally, everyone id idling
11:24<Guest1729>im looking at source engine games
11:24<Guest1729>and garrys mod may be a cpu hog
11:24<MajObviousman>it's Quakecon this weekend, less Source more iD
11:24<XReaper>yeah, you will have plenty of cpu
11:24*MajObviousman wishes he could go this year, but vacation
11:24<XReaper>for game server a 1024 would be best
11:25<XReaper>a 512... might run out of hdd space before ram
11:25-!-devcomp [] has joined #linode
11:25<MajObviousman>also, somebody had the bright idea to plan the Texas Linux Fest on the same weekend as Quakecon
11:25<Guest1729>How does the bandwidth tie into the equation
11:25<Guest1729>will I bog down the hardware or use up the bandwidth first
11:25<XReaper>50Mbit uplink
11:25<XReaper>that's plenty
11:25<Guest1729>I mean gb limit
11:26<Guest1729>like total
11:26<XReaper>400GB on a 1G
11:26<Guest1729>and yes burst is nice
11:26<Guest1729>I currently host on my 4 mbit home line
11:26<Guest1729>so 50 is an improvment
11:26<XReaper>50Mbit will be much nicer
11:26<MajObviousman>50MBit is more than you need
11:27<Guest1729>but no one donates
11:27<Guest1729>im not sure about paying
11:27<Guest1729>for a server
11:27-!-duckxx [] has joined #linode
11:27<XReaper>Guest1729: yeah. Well...
11:27<Guest1729>I might do it in the future
11:27<XReaper>If you had enough income you could justify :P
11:27<XReaper>or had a reason to have
11:27<Guest1729>I have zero income
11:28<Guest1729>I want to work in IT
11:28<XReaper>(i put two 512's on an annual... $555 gone :P)
11:28<Guest1729>but im 15
11:28<XReaper>yeah, linode is a bit expensive
11:28<rnowak>XReaper's mental age is 11, so that's fine
11:28<XReaper>works really nice though
11:28<XReaper>rnowak: ??
11:28<rnowak>sup?! (:
11:28<XReaper>been awhile
11:28<Guest1729>but employers look at calendar age
11:29<meskarune>it depends on what you are doing
11:29<meskarune>mowing lawns is 11 years old and up
11:29<XReaper>There are -ways- to get linodes
11:29<XReaper>amirite meskarune?
11:30<EugeneKay>Linode isn't expensive, it's priced for what you get - a server that doesn't randomly go down.
11:30<XReaper>EugeneKay: precisely
11:30<XReaper>Well... I run arch...
11:30*EugeneKay points and giggles
11:30<XReaper>pacman can kill
11:31<Guest1729>I have actually started to put ads in the motd of my servers
11:31<Guest1729>Can you resize linodes
11:31-!-Bry8Star-{GB [] has joined #linode
11:31<Guest1729>I might try to host for cs go
11:31<XReaper>resize means migrate to a new host
11:31<XReaper>so :P
11:31-!-Bry8Star-{GB is now known as Guest1731
11:32<XReaper>just downtime
11:32<XReaper>each host only carries one plan
11:32<Guest1729>how much
11:32<XReaper>depends on size of disks
11:32<XReaper>into the minutes
11:33<XReaper>not long
11:33-!-jhanjon [] has joined #linode
11:33<linbot>New news from forum: crontab -e no longer working; old crontab running in General Discussion <>
11:33-!-David [] has quit [Quit: Leaving...]
11:34<Guest1729>how is bandwidth overage handled?
11:35-!-Cydus [] has joined #linode
11:36<XReaper>A 20G linode will handle everything
11:36<Guest1729>Linode 512?
11:36<XReaper>you want a 5`12 :P
11:37<XReaper>20G is the top plan
11:37<Guest1729>no i mean
11:37-!-Bass10 [] has quit [Ping timeout: 480 seconds]
11:37<XReaper>A low $799.95/month
11:37<EugeneKay>Speaking as someone who manages several 12Gs, no, not really.
11:37<XReaper>EugeneKay: but that's you
11:37<Guest1729>I am stressing over $20 a month
11:37*heckman has seen a 20G OOM
11:37<XReaper>heckman: ha
11:38*EugeneKay has accidentally OOMed a 12G
11:38<XReaper>what were they/you doing on it?
11:38*EugeneKay has purposefully done it, too
11:38<@heckman>No idea.
11:38<XReaper>heckman: um... how many 20G's on a host... going by math: 1
11:38<rnowak>you act as if 20GB of RAM is a lot these days ;p
11:38<EugeneKay>MySQL >_
11:38<XReaper>Yeah servers can sit on 100's of GB
11:38<XReaper>EugeneKay: oh yes, the memory hog
11:39<Guest1729>so do linodes plans have larger cpus
11:39<Guest1729>as they go up
11:39<EugeneKay>A basic dual hex-core Xeon nowadays can have what.... 16 slots with 8GB sticks in each?
11:39<EugeneKay>At something like $50/stick
11:39<hawk>Guest1729: Only lower number of guests, afaik
11:39<XReaper>EugeneKay: then you get ECC models...
11:40<XReaper>Guest1729: basiaclly, less density == more cpu time
11:40<EugeneKay>Ah, ECC sticks are more like $80/stick
11:40<EugeneKay>Still, that is a LOT of ram.
11:41<XReaper>they can handle terabytes on some platforms...
11:42<EugeneKay>4-way CPUs are $$$$
11:42-!-flip_digits [] has joined #linode
11:43<XReaper>EugeneKay: $$$$ == :D for the IT people. D:< for accounting
11:43-!-fuchs_fuchs [] has joined #linode
11:43<EugeneKay>Well, if you need to scale up an Oracle DB it's still cheaper to buy a $100k server than it is to rewrite your application for sharding
11:44<EugeneKay>And you WILL need to scale up an Oracle DB, because it's enterprisey.
11:44<XReaper>Paypal does sharding with mysql?
11:44<rnowak>some funding that we get needs to be spent, or it is lost; the few weeks before this happens is known as hardware upgrade time
11:44<XReaper>i think they do...
11:45<XReaper>EugeneKay: that's when you buy a pre-made oracle server
11:45<EugeneKay>You can shard with any of the SQLs, if you have a good sharding key.
11:45<EugeneKay>For a dataset like PayPal's, where you have a lot of users whose data isn't connected in any real fashion, MySQL should work fine.
11:46<rnowak>mysql, work fine, ha ha ha
11:47<XReaper>EugeneKay: yeah, was talking to a DBA, apparently they have a nice system :P
11:47<XReaper>but he hates mysql... so yeah
11:48<@heckman>Starting to love Postgres moar and moar
11:48<rnowak>the only people that do not hate it are the ones ignorant or unaware of better
11:48<rnowak>and no, the opinion of a "I run my own blag mang", does not count
11:49-!-Gnintendo [] has quit [Quit: Leaving]
11:49<linbot>New news from forum: crontab -e no longer working; old crontab running in General Discussion <>
11:49<Guest1729>is the linode to linode network gigabit?
11:49<Guest1729>in a single facility
11:50-!-storrgie [] has quit [Quit: Leaving]
11:50<XReaper>your upstream is capped at 50Mbit
11:50<Guest1729>even internal?
11:51-!-bchristensen [~bchristen@] has quit [Remote host closed the connection]
11:51<XReaper>the upstream is rate-limited
11:51-!-mbreslin [] has quit [Server closed connection]
11:51-!-bchristensen [~bchristen@] has joined #linode
11:51<XReaper>i've asked before
11:52<@heckman>Same interface
11:52<KyleXY> 50Mbit is still fast none the less.
11:53<XReaper>mind you, the downstream is uncapped
11:53<rnowak>fast is relative \o/
11:53-!-stdaro [~sroylance@] has left #linode []
11:53<XReaper>but that doesn't really help much
11:53<EugeneKay>Unless you're downloading massive quantities of.... linux isos.
11:54-!-jhanjon [] has quit [Remote host closed the connection]
11:55-!-_j45 [] has left #linode []
11:57-!-Bdragon [] has quit [Read error: Operation timed out]
11:57<XReaper>EugeneKay: you are on efnet...
11:57<XReaper>you beat me up on idlerpg :(
11:57<EugeneKay>I usually don't have Xchat connected to that network. My ZNC sits in #idlerpg
11:58<EugeneKay>Good. You deserved it.
11:58<chesty>what's #idlerpg?
11:58<EugeneKay>!to chesty google #idlerpg
11:58<linbot>chesty: MultiNet Idlerpg: <>; #IdleRPG Idle RPG: Game Info: <>; AfterNET #IdleRPG: Home: <>; OperaNet Idle RPG - Stu's Opera Stuff: <>; IdleRPG on deviantART: <>; Channel search > idlerpg: <>; IdleRPG Contest! (1 more message)
11:58<XReaper>on efnet
11:59<@akerl>chesty: It's a good thing you have such good friends to teach you about googles
11:59-!-Bass10 [] has joined #linode
11:59<XReaper>i hate soldering irons
11:59<linbot>New news from forum: Redis on its own Linode? in General Discussion <>
11:59<chesty>i don't need to know bad enough to click a link
12:00<EugeneKay>Apparently I'm a level 40 EugeneKay.
12:02-!-ukitazume [] has joined #linode
12:08-!-Chukwumah [] has quit [Quit: changing servers]
12:09-!-DephNet[Paul] [] has quit [Read error: Connection reset by peer]
12:11-!-triplei [] has joined #linode
12:12<XReaper>i need to ditch this hdd out of my laptop...
12:13<XReaper>the fan almost never kicks in
12:13-!-Vadtec [~Vadtec@2001:470:1f06:13e0::1337] has quit [Server closed connection]
12:15-!-arooni-mobile [~arooni-mo@] has joined #linode
12:15-!-Vadtec [~Vadtec@2001:470:1f06:13e0::1337] has joined #linode
12:15-!-Bdragon [] has joined #linode
12:16<MajObviousman>woo documentation day
12:18*XReaper undocments all MajObviousman's documentation
12:19*MajObviousman documents XReaper's undocumentation so he doesn't look like he's slacking off
12:20<XReaper>trying to find a cheap broken 360 that isn't really broken :P
12:21<KyleXY>XReaper: The ones that have the dvd drive broken are fun
12:21<XReaper>i have one of the original 360's... can't do anything with it
12:21<XReaper>no RGH :(
12:21<XReaper>KyleXY: that's what i'm looking for
12:22<XReaper>i have a working drive
12:22<XReaper>and a powersupply etc :D
12:23<XReaper>KyleXY: only issue is, i don't want another xenon
12:23<MajObviousman>I bet most of the RROD versions are out of the supply pool by now
12:24-!-jakechapa [~jakechapa@] has joined #linode
12:24-!-John[a] [] has joined #linode
12:24<KyleXY>XReaper: A friend of mine got a nice deal the other day
12:24-!-John [] has quit [Read error: Connection reset by peer]
12:24<KyleXY>XReaper: Fairly large xbox library, plus xbox 360, for around a hundred bucks
12:24<XReaper>I got this one for $240 3-4 years ago
12:24<KyleXY>(Another friend didn't want it anymore.)
12:24<EugeneKay>eBay has a good number of Live banned ones
12:30-!-bchriste_ [~bchristen@] has joined #linode
12:31-!-wrf [] has quit [Server closed connection]
12:31-!-wrf [] has joined #linode
12:32<KyleXY>Gah, /me kills Comcast
12:35<XReaper>EugeneKay: i don't use live
12:35<EugeneKay>Then they're perfect for you?
12:36-!-HedgeMage [] has quit [Quit: brb]
12:36-!-bchristensen [~bchristen@] has quit [Ping timeout: 480 seconds]
12:37-!-HedgeMage [] has joined #linode
12:37-!-seanh-ansca [] has quit [Ping timeout: 480 seconds]
12:37-!-seanh-ansca [] has joined #linode
12:41-!-tailf [] has quit [Server closed connection]
12:42-!-tailf [] has joined #linode
12:44-!-arooni-mobile [~arooni-mo@] has quit [Read error: Operation timed out]
12:48<XReaper>ha, this one is a classic
12:55-!-vraa_ [] has quit [Read error: Operation timed out]
12:57-!-Deathvalley122 [] has quit [Server closed connection]
12:58-!-Deathvalley122 [] has joined #linode
12:58-!-bchriste_ [~bchristen@] has quit [Remote host closed the connection]
12:58-!-bchristensen [~bchristen@] has joined #linode
13:06-!-seanh-ansca [] has quit [Quit: Leaving.]
13:07-!-Cromulent [] has joined #linode
13:09-!-richardddddd [] has joined #linode
13:09<richardddddd>Anyone have experience with this module?
13:09<richardddddd>(using the .conf from their example fudges my site)
13:10<cruxeternus>Is there a way to boot Finnix in 32-bit mode?
13:13<cruxeternus>nm, found Rescue option :)
13:15<purrdeta>squircle: what happens if spamd isn't available when a mail comes with spamass-milter?
13:15<squircle>purrdeta: I believe the default action is to just pass the mail back to postfix without modifying it
13:16<purrdeta>hmm ok. That might be interesting too. I'm thinking of offloading spamd to another server which is less reliable (it isn't terribly by any means but it isn't great). I haven't decided yet but that is something I was curious about. If amavis dies all your mail just stops.
13:18<squircle>yeah, that's why I stopped using it
13:18<squircle>I'm 98% confident that it's default-pass (or configurable in the milter's config)
13:18<purrdeta>It probably is configurable after all.
13:19<purrdeta>I'm about to see if opendkim works and if it does I'll move on to spamass-milter
13:20<squircle><3 milter
13:20<linbot>New news from forum: Adding an 'O-Line' to ircd.conf in General Discussion <>
13:20<Cromulent>everytime I think about setting up email servers I cry a little bit
13:20<squircle>setting up a mail server has made me a stronger person :P
13:21<purrdeta>I know my opinion is less popular, but I have a lot of fun playing with it lol
13:22<EugeneKay>That which does not kill me only gives me cancer
13:22<hawk>squircle: You'll need that strength over the coming years :P
13:22<squircle>hawk: we're switching to hosted exchange next week
13:23-!-_j45 [] has joined #linode
13:23<purrdeta>squircle: you will need strength for that. I sort of like exchange, but strength needed. Lots of it.
13:24<squircle>Microsoft's Office 365-frontend to exchange configuration is a dream
13:24<hawk>I guess Exchange can be ok if it does what you need. It's not very good for mail specifically, though.
13:24<squircle>well my parents need it because they use blackberries and it needs to interface with another exchange server in their office
13:24-!-stevepiercy [] has left #linode []
13:25<squircle>I spent years administering an exchange server on Win2k, I know how hellish it is haha
13:25-!-fod [] has joined #linode
13:25<linbot>New news from forum: Linode resolvers vs own resolver in Linux Networking <>
13:25-!-fuchs_fuchs [] has quit [Quit: fuchs_fuchs]
13:27-!-_j45 [] has left #linode []
13:28<purrdeta>I think that Exchange is good at what it does. However, a lot of people think they need that and do not.
13:28<squircle>I agree
13:28<squircle>I wouldn't use Exchange if I didn't feel I had to
13:29-!-hipsters_ [] has quit []
13:31-!-fod [] has quit [Quit: Leaving]
13:31-!-Cydus [] has quit [Read error: Connection reset by peer]
13:31-!-fod [] has joined #linode
13:36-!-Gnintendo [] has joined #linode
13:39-!-richardddddd [] has quit [Remote host closed the connection]
13:39-!-richardddddd [] has joined #linode
13:39<richardddddd>any nginx config bosses?
13:41<rnowak>read the documentation
13:41<linbot>New news from forum: Two Factor Authentication in Feature Request/Bug Report <>
13:41-!-Cromulent [] has quit [Quit: WeeChat 0.3.7]
13:41<rnowak>unless you cut off a vital part of the configuration, it suggests you have not read the documentation for the module
13:42<richardddddd>ive read the github docs (the relevant parts)
13:43-!-fod [] has quit [Quit: Leaving]
13:43<rnowak>you've not even included the vital parts from the example configuration you supplied
13:44<richardddddd>rnowak: that was just to show my current base conf file
13:44<rnowak>which is wrong
13:44<richardddddd>well, no, the site is online working so that's the conf for it currently
13:44-!-alester [] has left #linode []
13:44<richardddddd>this line "listen default;" alone breaks the current config
13:45<richardddddd>Reloading nginx configuration: nginx: [emerg] a duplicate default server for in /
13:45<rnowak>you're trying to load both of the configuration files?
13:46<richardddddd>i was just taking parts of the document config so see which was causing the various errors
13:46<rnowak>then you better explain exactly what you're doing because it isn't evident from what you've typed up
13:48-!-message144 [] has joined #linode
13:49<richardddddd>rnowak: ok, here is my attempt at adding it to my conf
13:49<richardddddd>which results in the page just showing "No web site at this address"
13:49<rnowak>start by indenting the file correctly
13:50<richardddddd>what do you mean by that?
13:50-!-arooni-mobile [~arooni-mo@] has joined #linode
13:50<purrdeta>staticsafe: so like... I would think that I have opendkim right but it isn't signing. HALP
13:52<rnowak>richardddddd: I mean, make the file tidy, one indentation level per block scope, so that it can easily and cleanly be followed, instead of looking at a mess
13:53<rnowak>other than that, what error does nginx give you when you try restarting?
13:55<richardddddd>rnowak: heres it tidier nginx logs give no error
13:56-!-jakechapa [~jakechapa@] has quit [Ping timeout: 480 seconds]
13:56<purrdeta>staticsafe: sorry to bother, fixed it :P
13:57<rnowak>richardddddd: so nginx is running?
13:58<rnowak>what's the ip address of your linode?
13:59<richardddddd>rnowak: cant post that publically, but the 'No web site at this address' page it shows, is the same page that shows when you visit http://the linode ip
13:59<rnowak>ok, good luck!
14:01-!-ukitazume [] has quit [Ping timeout: 480 seconds]
14:02<purrdeta>squircle: sorry to bother you, but I've moved on to spamass-milter now that dkim is straightened out. I can't remember where you said to put it (non_smtpd or smtpd) to not check outgoing mail. And is that wise?
14:08<richardddddd>rnowak: fixed that problem, now getting nginx error: 1024 worker_connections are not enough while connecting to upstream, client:
14:08<rnowak>you've probably created a loop
14:09<richardddddd>rnowak: well i had this same conf except listen line i had as "listen 80 default"
14:09<linbot>New news from forum: Most distracting noise in your office in /dev/random <>
14:10-!-Cydus [] has joined #linode
14:10<rnowak>you're proxy_passing any request besides /progress to
14:11<richardddddd>don't i want to be passing all requests?
14:11<rnowak>which comes back in on, finds location /, awwww yeah, time to proxy pass to again
14:12<rnowak>it is not done correctly
14:12<richardddddd>the only difference i see is the server name in their example is "server_name _ *;" while mine is
14:13-!-vraa [] has joined #linode
14:13-!-hipsters_ [] has joined #linode
14:16-!-CornishPasty_ [] has joined #linode
14:16<arooni-mobile>if i enter a challenge password when genearting a CSR using "openssl req -nodes -newkey rsa:2048 -keyout myserver.key -out server.csr".... must i enter it on every nginx restart? is it better to leave challenge password blank?
14:18-!-Jonis_ [] has joined #linode
14:18-!-hawk_ [] has joined #linode
14:18<EugeneKay>You can strip the passphrase from the key using `openssl rsa -in foo.key -out bar.key`
14:19-!-jakechapa [~jakechapa@] has joined #linode
14:19-!-Netsplit <-> quits: phyber, descender, encode, Null_, Gnewt, decklin_, igufi, Frools, fulg0re, kethry, (+109 more, use /NETSPLIT to show all of them)
14:19-!-CornishPasty_ is now known as CornishPasty
14:19-!-Netsplit over, joins: XDS2010
14:19<arooni-mobile>so i dont need a pssphrase for the key
14:19-!-rnowak [] has joined #linode
14:19<EugeneKay>The convenience of your server being able to start unattended is usually worth the risk of having the key sit in plaintext on the disk.
14:19<arooni-mobile>since only i have access to the server
14:20<EugeneKay>Just make sure nobody roots your box and you have good perms set on the key+keydir
14:20-!-Netsplit over, joins: Consdale, fulg0re
14:20<arooni-mobile>what would that allow them to do
14:20<MajObviousman>well, if somebody roots your box you're fucked in a lot of different ways
14:20<EugeneKay>Steal your SSL cert
14:21<arooni-mobile>if i only allow key based login
14:21<arooni-mobile>i suppose thatd be a lot harder; at a not standard ssh port; and use fail2ban
14:22<arooni-mobile>+firewall... i mean only so much i can do
14:22<rnowak>>at a not standard ssh port
14:22-!-jsh [] has joined #linode
14:22*rnowak throws a tree at arooni-mobile
14:22-!-Netsplit over, joins: jpg
14:23-!-Netsplit over, joins: MrYiff
14:23*MajObviousman throws a forest at rnowak
14:23<rnowak>I can take it, COME AT ME
14:23*MajObviousman then runs away from the tree huggers
14:23-!-bacon [~rasher@2a01:7e00::f03c:91ff:feae:d964] has quit [Server closed connection]
14:23-!-SNy [] has joined #linode
14:23<rnowak>on it!
14:23-!-Dokujisan [] has quit [Read error: Operation timed out]
14:23-!-Netsplit over, joins: tomaw, bigjust, richardddddd, bchristensen, Deathvalley122, tailf, HedgeMage, dubenstein, kingdutch, smed_ (+28 more)
14:23-!-Netsplit over, joins: stan_theman, tres__, mwalling, daj_, tonyyarusso, decklin_, BrianHV_, ajmitch_, windsurf1r, tharkun
14:23-!-Netsplit over, joins: irgeek, BlandSau1e, encode, dr_jkl
14:23-!-Netsplit over, joins: mcone, bd_, staticsafe, G, praetorian, shedammit, kenyon, PrgmrBill^, lkb_, carnage_ (+11 more)
14:23-!-Netsplit over, joins: @deaton, marius, chesty, rasengan, spaam, kyhwana, Chris___, eshlox, avorntur, Frools (+11 more)
14:23-!-ServerMode/#linode [+oov irgeek deaton deaton] by
14:23-!-Netsplit over, joins: Myon, pharaun, JStoker, zibri, Dave, GLaDOSDan, Neal, LiquidAtom, bliblok, graq (+4 more)
14:23-!-Oloryn [] has quit [Read error: Connection reset by peer]
14:23-!-wrf [] has quit [Max SendQ exceeded]
14:23<rnowak>my paper is 100% environtment friendly, made purely from environment!
14:23-!-wrf [] has joined #linode
14:23-!-wrf [] has quit [Max SendQ exceeded]
14:24<MajObviousman>hello netsplit, how are you
14:24-!-bchristensen [~bchristen@] has quit [Write error: connection closed]
14:24-!-wrf is "(unknown)" on (unknown)
14:24<MajObviousman>yeah, the vegan movement never made much sense to me
14:24<EugeneKay>I am AWESOME
14:24-!-mode/#linode [+o stan_theman] by ChanServ
14:24-!-wrf is "(unknown)" on (unknown)
14:24-!-wrf [] has joined #linode
14:24<MajObviousman>you're eating things from the environment still, idiots
14:24-!-mode/#linode [+v irgeek] by ChanServ
14:24-!-jsh_ [~jsh@] has joined #linode
14:25<EugeneKay>I love vegetarians. They taste great, flame broiled and covered in A-1.
14:25<MajObviousman>bah hahaha
14:26-!-Oloryn [] has joined #linode
14:27-!-jsh [] has quit [Remote host closed the connection]
14:27<jsh_>Hi everyone - I am looking to run a centos 6 server with nginx, php-fastcgi, MySQL, as well as a mail server (Dovecot?). None if these will be particularly stressed (1k web hits daily, 50 emails/day, and pretty low DB use.). How much ram
14:27<EugeneKay>!apropos ram
14:27<linbot>EugeneKay: linuxatemyram, ram, and scramble
14:27<jsh_>Do I probably need?
14:27<tharkun>jsh_: do you plan on recievin mail on that server?
14:27<richardddddd>rnowak: someone here using a different way
14:28<EugeneKay>Hrm. Thought there was a factoid for that.
14:28<rnowak>richardddddd: the network split, I probably lost a message or two inbetween if you wrote anything else
14:28<jsh_>@ tharkun - Probably under 15 emails daily
14:28<EugeneKay>jsh_ - I'd start with a 512, that really is nothing for load.
14:28<linbot>MajObviousman: (scramble <text>) -- Replies with a string where each word is scrambled; i.e., each internal letter (that is, all letters but the first and last) are shuffled.
14:28<MajObviousman>!scramble EugeneKay has a very long nose
14:28<linbot>MajObviousman: EKanegeuy has a very lnog nsoe
14:28<MajObviousman>needs longer words to have an impact
14:29<MajObviousman>!scramble eschew obfuscation
14:29<linbot>MajObviousman: ecehsw otocsufiban
14:29<KyleXY>!scrable MajObviousman loves to pick his nose.
14:29<KyleXY>!scramble MajObviousman loves to pick his nose.
14:29<tharkun>jsh_: Think of it as a nice jigsaw. One piece at a time. For sending and recieving you have several options. google mta
14:29<linbot>KyleXY: MavjsuaobmOin loves to pcik his nose.
14:29<MajObviousman>that's way more than 7 letters man. You suck at scrabble
14:29<KyleXY>MajObviousman: lol
14:29<jsh_>EugeneKay: I figured and knew it was very little. I am mainly looking for security, which shared doesn't give me at the level I want.
14:30<richardddddd>rnowak: i didn't send any new ones except the last, i didnt understand your msg about directive
14:30<rnowak>you're missing a directive; upload_progress <something> <something>
14:30<EugeneKay>Keep in mind that you're only as secure as your host anywhere you go. With Linode you get a Xen VPS, but it's still not a dedicated box with hardware disk encryption
14:30<jsh_>tharkun: I knew my MTA's yesterday from reading the Linode Library. But that knowledge has been dumped.
14:30<rnowak> read through that, use your base configuration if it works, and adapt it to what this needs
14:30<MajObviousman>jsh_: your box will be as secure as you know how
14:30<EugeneKay>You /can/ do pv_grub+truecrypt or such, but that's a PITA to boot(involves typing in a password)
14:30<tharkun>jsh_: you will also need sometipe of firewall research iptables
14:30<richardddddd>rnowak: that is in my nginx.conf main file which this domain conf is then added
14:30<MajObviousman>and it's not a thing you can do casually. It will take some reading
14:31<richardddddd>rnowak: upload_progress proxied 1m;
14:31<staticsafe>purrdeta: \o/
14:31<rnowak>richardddddd: it needs to be in a http block, is it?
14:31<jsh_>MajObviousman: I've been reading quite a bit of security stuff.
14:31<MajObviousman>there's lots of moving pieces in a system, and any one can be vulnerable to attack
14:31<richardddddd>rnowak: yes 100%
14:31<rnowak>richardddddd: alrighty
14:31<jsh_>EugeneKay: I am not that paranoid ;)
14:31<MajObviousman>jsh_: cool =) Practicing on a real server is the best way to take academic knowledge and convert it to experience
14:31-!-fcoury [] has quit [Max SendQ exceeded]
14:31<rnowak>but yes, the error you're getting suggests nginx is infinitely looping requests back to itself
14:32<MajObviousman>even better, practicing on a real server with other people's data is >> with your own data
14:32<jsh_>tharkun: I am looking at a service called dome9. Anyone know much about them?
14:32<tharkun>MajObviousman: volunteering something?
14:32<purrdeta>staticsafe: you gave up on spamass-milter right? :P
14:32<MajObviousman>I do something like this, "Hmm, that service isn't working right. Meh, nobody uses it anyway, I'll fix it later."
14:32<staticsafe>purrdeta: I did
14:32<richardddddd>rnowak: here is my main nginx.conf to the point where it pulls in my other domain conf at the bottom
14:33<tharkun>jsh_: I personally use iptables without a glitch. Ok maybe one over ipv6 but that was easily solved
14:33<purrdeta>why do the perms of a sock need to be so damn closed anyway lol
14:33-!-seanh-ansca [] has joined #linode
14:33-!-jakechapa [~jakechapa@] has quit [Read error: Operation timed out]
14:34<staticsafe>purrdeta: I made opendkim use a inet socket cause i'm lazy
14:34<purrdeta>its probably easier. damn
14:34-!-SelfishMan_ [] has quit [Remote host closed the connection]
14:34<rnowak>richardddddd: ok
14:34<jsh_>Oops.. tharkun: I've been contemplating that. But iptables didn't like me when I used it in a VM (and I have confidence that my rules were correct).
14:35-!-SelfishMan [] has joined #linode
14:35<richardddddd>i dont see anything wrong there, other than my indents :p
14:35<rnowak>well, some things are questionable, but they will work
14:35<jsh_>tharkun: I don't *need* the GUI or anything though. I am very comfortable in a CLI environment
14:35<tharkun>jsh_: If you use a recent enough kernel then there is no problem at all. What else can I say there is a very decent work done here to have it run extremely smooth.
14:36<tharkun>jsh_: IS THERE ANY OTHER WAY?? :-)
14:36-!-jgornick [] has joined #linode
14:36<rnowak>richardddddd: stop nginx and start it again, not doing a restart -- paste the exact error you get
14:36-!-fcoury [] has joined #linode
14:36<richardddddd>rnowak: ok 2mins
14:37<jsh_>tharkun: well yes, there is cPanel (expensive), or Dome9... Those offer GUI firewall building, but I don't *need* either. Just so you feel better, I will not be running X, Gnome, or any other desktop GUI on a server
14:39<tharkun>jsh_: I can tell you that there is no need for that. I've only found a tiny problem with a few ip6tables modules because the OS I'm running is not precisely up to date
14:39<richardddddd>rnowak: 2012/08/03 19:37:58 [alert] 7439#0: *84287 1024 worker_connections are not enough while connecting to upstream, client:, server:, request: "GET / HTTP/1.0", upstream: "", host: "", referrer: ""
14:39<rnowak>yeah, quite likely that
14:39<jsh_>tharkun: Yeah, I just need more iptables practice...
14:39<rnowak>can you link to your full base configuration?
14:39<rnowak>the vhost one that is
14:40<richardddddd>rnowak: the one for that specific domain?
14:40<rnowak>because I am not convinced you need to proxy_pass back to nginx there
14:41-!-Dokujisan [] has joined #linode
14:41<tharkun>jsh_: feel free to ask there are lots of people here that will happily lend a helping hand
14:41<richardddddd>rnowak: here you go
14:42<jsh_>I can tell. So much is happening in the channel that I am scrolling up and down to catch stuff ;). On another note, have you heard much about Percona server? It's supposed to be a drop-in replacement for MySQL, just more efficient.
14:42<rnowak>ok, you can just put your track_uploads in the location ~* .\php$ block starting on line 64
14:42-!-John[a] [] has quit [Quit: John[a]]
14:42-!-jakechapa [~jakechapa@] has joined #linode
14:43<rnowak>and you can remove the proxy_pass and proxy_redirect in the block on line 17
14:43<richardddddd>rnowak: it'll try that out now
14:44<rnowak>as a side note, your enforce no www should be done in its own vhost for rather than using the if statement
14:45<richardddddd>rnowak: got a 500 error, same in the log as previously :/
14:45<rnowak>paste the current configuration
14:46<richardddddd>1sec saw a typo will retest
14:46<MajObviousman>which is ironic because the admin before me used it everywhere in nginx configs
14:46<rnowak>that does not make it ironic
14:47<purrdeta>staticsafe: well I got spamass-milter past permission denied, but it appears to do some very funky stuff with my LDAP setup. For instance, after I enabled it everything sent told me "SMTP AUTH failed" and dovecot logged trying to connect to the remote SMTP server for LDAP.
14:47<purrdeta>so I quickly undid that.
14:47<staticsafe>purrdeta: O_o
14:47<MajObviousman>I suppose not
14:47<richardddddd>rnowak: heres the current conf
14:48<rnowak>"WARNING: this directive must be the last directive of the location. It must be in a proxy_pass or fastcgi_pass location. Repeating the directive in a location will results in segfaults."
14:48<purrdeta>staticsafe: so basically I seem to be stuck with it. Not entirely sure what the heck happened or even how to fix it because that is just really bizarre.
14:49<richardddddd>rnowak: site now loads!
14:49<richardddddd>no errors in logs either
14:49<rnowak>I thought you said you read the documentation (:
14:50<richardddddd>well i added at the start since you mentioned line 64, oh well. ill go and remove that IF now also
14:53-!-John [] has joined #linode
14:58<linbot>New news from forum: Is redis needed for this? in Web Servers and Web App Development <>
15:01-!-Cydus [] has quit [Quit: Leaving...]
15:02-!-jsh_ [~jsh@] has quit [Remote host closed the connection]
15:03<richardddddd>rnowak: because its in the php area it now only returns "new Object({ 'state' : 'starting' })"
15:04-!-kingdutch [~kingdutch@] has quit [Quit: kingdutch]
15:04<richardddddd>if you see 3rd post down :s
15:04<rnowak>it shouldn't -- I use very similar, using uwsgi_pass (python application)
15:05<rnowak>are you actually passing the unique identifier with the upload?
15:05<richardddddd>passed about 20times during this test upload
15:05<richardddddd>returns "starting" everytime
15:06-!-ansch [~ansch@] has joined #linode
15:07-!-jakechapa [~jakechapa@] has quit [Ping timeout: 480 seconds]
15:07-!-stas [~stas@] has quit [Ping timeout: 480 seconds]
15:08<rnowak>I can't check the configuration that I used right now, but I'm fairly sure there was nothing in location / regarding it
15:09<rnowak>what is the form's action set to?
15:10<rnowak>(I have a guess why it works for my application and not for ~\.php -- I send everything to the application that does *not* follow any other location block )
15:10<richardddddd>rnowak: action is (controller)
15:11<richardddddd>rnowak: relevant?
15:11<rnowak>ah, so no .php in the location
15:11<richardddddd>no .php anywhere on the site
15:12<arooni-mobile>hi folks; do i have a dedicated ip with linode?
15:12<arooni-mobile>or is it a shared IP
15:12<Peng>arooni-mobile: Dedicated.
15:12<MajObviousman>arooni-mobile: Dedicated.
15:12<arooni-mobile>a little unclear why my cert isnt working :(
15:12<Peng>arooni-mobile: Actually, you have two dedicated IPs, since Linode provides both IPv4 and IPv6. :P
15:12<Peng>arooni-mobile: I'm curious, which CA did you use?
15:13<arooni-mobile>Peng, went with comodo cuz it was $2 through namecheap
15:13<rnowak>richardddddd: this will be some odd issue by how try_files is invoked probably
15:13<Peng>arooni-mobile: StartSSL is $0 through StartSSL. :X
15:13<rnowak>richardddddd: the http progress upload module is a really, really ugly hack, and it is pretty fragile
15:13<arooni-mobile>Peng, i know but you have to revalidate it every 30 days i think
15:13<Peng>arooni-mobile: No you don't?
15:13<purrdeta>arooni-mobile: once a year
15:14<purrdeta>the free ones expire yearly.
15:14<arooni-mobile>but it said something about proving you own the domain every 30 days
15:14<purrdeta>thats only if you need to add new stuff after 30 days
15:14<richardddddd>rnowak: try_files in location / ?
15:14<purrdeta>I only revalidate my domain once a year, when its time to fix all the expiring certs.
15:14<arooni-mobile>i did try startssl first
15:14<arooni-mobile>i couldnt get it working
15:15<rnowak>richardddddd: afraid I can't offer any advice on how to continue without knowing about the application
15:15<arooni-mobile>little wonder i cant get the $2 cert working either
15:15<MajObviousman>rnowak: it's a secret to everybody
15:15*MajObviousman gives rnowak 30 rupees
15:15<rnowak>richardddddd: actually ignore that try_files part, it is irrelevant -- but the matching for \.php is probably throwing it off
15:15<rnowak>MajObviousman: I wasn't implying that I want to know more, fyi ;p
15:15<MajObviousman>it's hard to tell with you
15:16<purrdeta>*shrug* it has never been an issue for me :/
15:16-!-jsh [~jsh@] has joined #linode
15:16*rnowak stares at MajObviousman
15:16*MajObviousman holds up a mirror, points the stare back at rnowak
15:16<arooni-mobile>you win. i follwed ... and when i navigate to it just hangs... i'm tailing the log files in nginx but not seeing anything
15:17-!-vinic [] has quit [Server closed connection]
15:17<arooni-mobile>so to make ssl work properly do i need one domain per ip address
15:17-!-vinic [] has joined #linode
15:18<MajObviousman>arooni-mobile: check to see that your nginx binary supports SNI
15:18<richardddddd>rnowak: how else could i do it? since none of the php ends with .php
15:18<MajObviousman>I think it's nginx -V to get it to print out modules
15:18<MajObviousman>if it does support SNI, then your'e good to go, none of this 1 IP, 1 SSL cert bullshittery
15:18<rnowak>richardddddd: how does nginx manage to route your requests if you're matching for \.php$ but no .php is seen in your urls?
15:19<arooni-mobile>MajObviousman, nginx: TLS SNI support enabled
15:19<MajObviousman>then you're golden
15:19<MajObviousman>just do a bit of reading on nginx and SNI so you don't step in shit
15:19<MajObviousman>also, keep in mind by using SNI you're going to lock out some of your more ancient users. I think it is unsupported by browsers older than what, 4 years old? 5?
15:19<richardddddd>rnowak: codeigniter does the routing i think
15:19<MajObviousman>less of a problem now than it was. Every modern browser supports SNI
15:20<arooni-mobile>namecheap is saying, "Unfortunately one SSL Certificate can be installed on one dedicated IP-address. As I can see there are several domains are hosted there, which means that IP-address is rather shared, than dedicated. Moreover there is probably installed an SSL Certificate in this IP-address for another domain, which causes the issue."
15:20<rnowak>richardddddd: at application level, but I don't see how your requests ever reacy the location ~* \.php$ block
15:20<MajObviousman>well namecheap are a bunch of uneducated asshats then
15:20<MajObviousman>send them an email with a link to this:
15:21<MajObviousman>then find a vendor that isn't stupid
15:21<rnowak>ha ha
15:21<rnowak>ladies and gentlemen, MajObviousman, give him a big applause
15:21<richardddddd>rnowak: maybe this part? try_files $uri /index.php;
15:21<rnowak>duh, yes ok, that would be the one
15:21<richardddddd>within the location ~* \.php$
15:22<rnowak>I am not sure how I would deal with this
15:22*MajObviousman scrolls up looking for the pastebin
15:22<richardddddd>an if? :o
15:22-!-joecool [] has quit [Read error: Operation timed out]
15:22<arooni-mobile>any suggestions on what to look at if just hangs
15:23<rnowak>the if statement in nginx does not work how one would think it does
15:24<Yaakov>Anyone in striking distance of downtown Chicago, please accept my personal invitation to attend a Linode-sponsored F/OSS community event:
15:24<MajObviousman>arooni-mobile: have you verified the simple stuff?
15:24<MajObviousman>like is port 443 open in your firewall?
15:24<richardddddd>rnowak: this one has a proxy_pass linked in the last comment on
15:24<MajObviousman>is your web server listening on port 443?
15:24<richardddddd>not sure if his comment wording means that worked for him or not
15:24<arooni-mobile>MajObviousman, i just ran openssl s_client -connect .. and it appears the startssl attempt i installed earlier is messing up the validation chain
15:25<arooni-mobile>how do i uninstall a given cert
15:25<arooni-mobile>just delete the cert in /etc/ssl/certs ?
15:25<arooni-mobile>anything else?
15:25<arooni-mobile>MajObviousman, obviously with not godaddy
15:25<MajObviousman>I believe so yes
15:25<MajObviousman>and yes of course with the mydomain =)
15:25<MajObviousman>if you've got a bad cert getting rid of it and everything after it in the chain will let you recreate it
15:25*MajObviousman chucks a comma in there
15:27<rnowak>richardddddd: but you see, he has the proxy_pass and track_uploads in a normal location /, not one which the request gets rewritten to by try_files, and one that is rather specific using ~*\.php$ as its qualifier
15:27<rnowak>I am not sure, but I am guessing the latter part confuses the module
15:27<arooni-mobile>simply removing the cert i'm still seeing, "depth=1 CN = OpenVPN Web CA 2012.01.15 16:11:48 PST li25-146 verify error:num=19:self signed certificate in certificate chain"
15:27<arooni-mobile>i removed it form /etc/ssl/certs ... do i need to reboot ubuntu or something?
15:30<richardddddd>rnowak: so the progress url being checked needs a .php or?
15:30<rnowak>I'm not sure if it would work then either -- I do not know how the module works internally, other than the few seconds I skimmed through it, and came to the conlcusion that it is a horrible hack -- but it works well enough for me
15:32<richardddddd>rnowak: changed it to request /progress.php?X-Progress-ID=85d9f75cdd85f8e7742725280058ce9e
15:32<richardddddd>still returns starting throughout :/
15:32-!-jsh [~jsh@] has quit [Ping timeout: 480 seconds]
15:32<rnowak>maybe it cannot be in a location block like you have it, which complicate things with php
15:33<richardddddd>how can i have it in / then?
15:33<rnowak>as I said before, the way for requests to get through to my application is for them to not match any specific location block, and that means just a location / block for the application, and that could very well be why it works
15:34<rnowak>any other, more specific *
15:34<arooni-mobile>MajObviousman, any ideas ? im still getting " Verify return code: 19 (self signed certificate in certificate chain)"
15:34<rnowak>I will now take a second to urge you to switch to something more awesome like python for your future projects (:
15:35*rnowak waves python flag
15:36<auraka>rnowak is also a fan of arch
15:36<richardddddd>rnowak: ive actually been reading up for my next web app, but id like this one to work xD
15:36<rnowak>he is not, he is a fan of debian though
15:36<linbot>Today is Debian Appreciation Day! \o/
15:37-!-duckydan [] has quit [Quit: Bye!]
15:37-!-steveg [] has quit [Remote host closed the connection]
15:38<rnowak>richardddddd: you want it to work? rewrite it in python! (:
15:38<rnowak>there's some incentive to get started right now!
15:38<auraka>what is it written in now
15:38<richardddddd>if i had a time machine
15:38<richardddddd>php with codeigniter
15:39<arooni-mobile>having trouble with sslcerts. i have this error message when i run: openssl s_client -connect ==> "depth=1 CN = OpenVPN Web CA 2012.01.15 16:11:48 PST li25-146; verify error:num=19:self signed certificate in certificate chain" how do i find and delete this cert?
15:39<rnowak>it isn't really php's fault this time, but still!
15:39<rnowak>arooni-mobile: it will be referenced in your httpd configuration
15:39<richardddddd>i bothered re-compiling nginx for this crappy plugin so need it to work :(
15:40<arooni-mobile>rnowak, so its not in /etc/ssl/certs
15:40<arooni-mobile>its referenced in my nginx config?
15:40<auraka>richardddddd: recompiling nginx takes 5 min so we aren't talking that much work
15:40<rnowak>I don't know, it could be, or maybe it isn't, but it could be, but it might not be, but it could be
15:40-!-duckydan [] has joined #linode
15:40<richardddddd>when its in the location / , could the try_files be messing it up? :s
15:40<rnowak>it would be referenced in your httpd's configuration for it to load it
15:41<auraka>arooni-mobile: he is saying it should be refernced in whatever your webserver configuration is
15:41<rnowak>richardddddd: *maybe* but the test you did means probably not
15:41<rnowak>it probably does not cope with location ~*\.php$
15:41<arooni-mobile>is there a way to test my chained cert to see if i chained it / validates correctly?
15:41-!-duckydan [] has quit []
15:41<rnowak>(or it *really* wants it to be location / and nothing else, I have not checked that this is not the case)
15:43<rnowak>the documentation does not suggest that would be the case
15:43-!-stas [~stas@] has joined #linode
15:44<richardddddd>rnowak: tried / again, nice 500 error
15:44-!-DanishRolls [] has joined #linode
15:45<rnowak>well, you cannot do the way you wanted to, or the way the example suggests, by using proxy_pass back to itself, as it most certainly will create an infinite loop, as the error message seems to confirm
15:46<richardddddd>rnowak: new error if i move the progress block above location / "*21409 recv() failed (104: Connection reset by peer) while reading response header from upstream,"
15:47<richardddddd>that's a 502 error this time
15:47<arooni-mobile>in trying to track down my problems; i removed all ssl config from my nginx config. restarted nginx, and ran "openssl s_client -connect" still getting "Verify return code: 19 (self signed certificate in certificate chain)" ... really curious how nginx is serving ssl if im not telling it to
15:48-!-jakechapa [~jakechapa@] has joined #linode
15:48-!-binaryseed [] has joined #linode
15:48<rnowak>richardddddd: no idea I'm afraid, out of ideas for this -- other than using two chained nginx instances, or configurations (might work), but that's crazy
15:50-!-duckydan [] has joined #linode
15:51<arooni-mobile>i even turned off nginx and somehow openssl is saying i am still serivng ssl somehosw
15:52-!-devcomp [] has quit [Quit: devcomp]
15:53<MajObviousman>arooni-mobile: sorry got some calls there
15:54<arooni-mobile>MajObviousman, hey no problem; you're not obligated to help; i'm just pretty frustrated haha
15:54<MajObviousman>ssl management can get a bit hairy
15:54-!-John [] has quit [Quit: John]
15:54<arooni-mobile>i find it fascinating that turning off nginx doesnt cause the opensssl cert check to do anything different
15:54<arooni-mobile>whats serving ssl certs if nginx isnt even running
15:55-!-JoeK [] has quit [Server closed connection]
15:55-!-richardddddd [] has quit [Remote host closed the connection]
15:55<MajObviousman>you're sure nginx is dead?
15:55<MajObviousman>I've had nginx get locked and stop responding to signals
15:55-!-JoeK [] has joined #linode
15:56<arooni-mobile>MajObviousman, ps -lA | ack nginx ==> nothing
15:57<arooni-mobile>its saying the first (and bad cert is ) 1 s:/CN=OpenVPN Web CA 2012.01.15 16:11:48 PST li25-146
15:57<MajObviousman>what platform is this?
15:57<arooni-mobile>openvpn was running; so i stopped it; re ran the test... same error
15:57<arooni-mobile>i'm thinking its openvpns fault
15:57<arooni-mobile>MajObviousman, ubuntu 10.04 lts
15:57<MajObviousman>would be my guess as well
15:57<MajObviousman>can you purge the certs and reinstall the package to get it back to a baseline?
15:57<linbot>New news from forum: easiest way to override foreign MX records for outgoing mail in Email/SMTP Related Forum <>
16:00-!-fisted [] has quit [Quit: leaving]
16:00<arooni-mobile>MajObviousman, well i think i made some progress; killed the openvpn processes; and now at least openssl s_client -connect is returning connection refused
16:00-!-fisted [] has joined #linode
16:02-!-jsh [~jsh@] has joined #linode
16:10-!-binaryseed [] has quit [Remote host closed the connection]
16:11<MajObviousman>anyone here running debian squeeze and iptables-persistent?
16:13<arooni-mobile>MajObviousman, purged firewall rules to be safe; sitll getting connection refused using openssl tool
16:13<MajObviousman>arooni-mobile: I dunno man, I'm out of my depth here
16:14-!-Guest1729 [] has quit [Quit: Guest1729]
16:14-!-Whoop [] has joined #linode
16:16-!-bacon [] has joined #linode
16:16<bacon>Good day to you sirs!
16:17*MajObviousman nom noms
16:18<EugeneKay>GET IN MY SAMMICH
16:18-!-jsh [~jsh@] has quit [Ping timeout: 480 seconds]
16:18-!-Cydus [] has joined #linode
16:22-!-joecool [] has joined #linode
16:23-!-Cromulent [] has joined #linode
16:23-!-jakechapa_ [~jakechapa@] has joined #linode
16:26-!-sbattey [] has joined #linode
16:28-!-jakechapa [~jakechapa@] has quit [Ping timeout: 480 seconds]
16:28-!-duckydan [] has quit [Read error: Operation timed out]
16:30-!-duckydan [] has joined #linode
16:32-!-hawk_ is now known as hawk
16:35-!-smed_ [] has quit [Ping timeout: 480 seconds]
16:36-!-userme [] has joined #linode
16:40-!-squircle [~squircle@2607:f2c0:a000:138:216:76ff:feb4:8027] has quit [Remote host closed the connection]
16:40<linbot>New news from forum: crontab -e no longer working; old crontab running in General Discussion <>
16:41-!-squircle [~squircle@2607:f2c0:a000:138:216:76ff:feb4:8027] has joined #linode
16:44-!-sbattey [] has quit [Quit: Computer has gone to sleep.]
16:46-!-John [] has joined #linode
16:46-!-zed173 [~zedsdead@] has joined #linode
16:48-!-zed173 [~zedsdead@] has left #linode []
16:49-!-zed173 [~zedsdead@] has joined #linode
16:49-!-skvare [] has joined #linode
16:49-!-skvare [] has left #linode []
16:50<Yaakov>Has there been any trouble with the credit card run?
16:50-!-zed173 [~zedsdead@] has quit []
16:51<@heckman>Yaakov: mine got charged
16:53<auraka>heckman: yours got charged?
16:56-!-logichole [] has joined #linode
16:56<auraka>why would you be charged?
16:56<@heckman>Because I pay for a Linode o_O
16:57<auraka>that is like the owner of Chic-fil-a throwing a gay wedding just doesn't make sense
16:57<hawk>And here we all assumed you got paid in Linodes
16:58<rnowak>auraka: so if you work at a diamond cutter, you expect all the diamonds you want in return?
16:58<auraka>rnowak: comparing linodes to diamonds is dumb
16:58<rnowak>they are both resources
16:59<auraka>one is physical and very valuable...the other is virtual and cheap to give to employees
16:59<auraka>it is like comparing apples to volvos in driving ability
16:59<rnowak>I can't compare linodes to food, because you can't stuff your face with infinite amounts, but you could want all the linodes
16:59<rnowak>unless it is urmom
17:02<Yaakov>heckman: Mine didn't.
17:02<linbot>New news from forum: Linode Sponsored: The Perl Foundation Community Symposium in General Discussion <>
17:02<auraka>rnowak: I think you have proven you can stuff your face with infinite amounts of food fatty
17:05-!-seanh-ansca [] has quit [Quit: Leaving.]
17:09<Yaakov>heckman: Are you here?
17:10<@heckman>Yaakov: yes I am
17:10<Yaakov>heckman: /msg?
17:13-!-Drone4four [] has quit [Remote host closed the connection]
17:16-!-Drone4four [] has joined #linode
17:19-!-patcito [~123@] has joined #linode
17:25-!-ashaffer [] has quit [Quit: Leaving.]
17:29-!-patcito_ [~123@] has joined #linode
17:32-!-patcito [~123@] has quit [Read error: Operation timed out]
17:34-!-ashaffer [] has joined #linode
17:34-!-Bass10 [] has quit [Quit: Leaving]
17:37-!-io_ [] has joined #linode
17:39-!-R5Chris [] has quit [Ping timeout: 480 seconds]
17:43-!-vraa [] has quit [Quit: Leaving]
17:48-!-John[a] [] has joined #linode
17:48-!-John [] has quit [Read error: Connection reset by peer]
17:49-!-John[a] [] has quit []
17:52-!-John [] has joined #linode
17:59-!-seanh-ansca [] has joined #linode
18:00-!-mode/#linode [+o mdcollins] by ChanServ
18:03-!-dwhite [] has joined #linode
18:05<dwhite>So I've built a PPTP VPN on a CentOS 2 Linode in London. I'm able to connect to it from my Windows 7 machine, but my client gateway, according to ipconfig (on Windows) is set to, and I'm unable to browse anywhere. I'm googling around a little, but thought I'd stick my head in here as well to see if folks might have any ideas
18:06<dwhite>s/CentOS 2/CentOS 6.2
18:06<EugeneKay>PPTP is a terrible, insecure protocol, disavowed by Microsoft. Please don't use it.
18:07<EugeneKay>Use openvpn with "redirect-gateway def1" instead.
18:07<dwhite>I'm shutting it down in 2 weeks and don't plan on using it for anything important. Just playing with it, and setting it up to stream some olympics from the BBC website. =)
18:07-!-DrJ [] has quit [Quit: Bye]
18:07<dwhite>I realize it's not secure.
18:08-!-ashaffer [] has quit [Quit: Leaving.]
18:11-!-plastics [~plastics@2600:3c02::f03c:91ff:feae:6a27] has quit [Server closed connection]
18:11-!-plastics [~plastics@2600:3c02::f03c:91ff:feae:6a27] has joined #linode
18:12-!-ponas- [~chiisai@] has joined #linode
18:12<bacon>dwhite, tut, tut, tut
18:13-!-ponas [~chiisai@] has quit [Ping timeout: 480 seconds]
18:13<bacon>there are some sopcast streams running
18:14<bacon>And there are a number of open proxies in the uk too, you just need to look.
18:16<@mikegrb>mmm bacon
18:16-!-ponas [~chiisai@] has joined #linode
18:17<bacon>jeremyb, :*
18:17<dwhite>Can't a man just work on a project for the sake of learning? =) (first time setting up a VPN server, we're geeks... you get the point. On the flip side, I attempted OpenVPN but decided to scratch it after running into some complications - I'm DEFINITELY not going to use pptp in the future if I need a real VPN box, and will take the time to learn OpenVPN)
18:17<dwhite></end soap box rant>
18:17<bacon>and pptp is very much a pain in the ass when it doesn't work - which is most of the time, IME
18:17<dwhite>heh... but thanks for the suggestions
18:18<dwhite>lol. I'll take a look at OpenVPN again with the option suggested by Eugene
18:18-!-arooni-mobile [~arooni-mo@] has quit [Ping timeout: 480 seconds]
18:18-!-DanishRolls [] has quit [Read error: Connection reset by peer]
18:18-!-hipsters_ [] has quit []
18:19<dwhite>I admittedly didn't spend a lot of time on it... 1-2 hours max. I'll give it another shot - would be better to figure it out and do it right from the start, I'll admit (and agree)
18:19-!-arooni-mobile [~arooni-mo@] has joined #linode
18:19-!-arooni-mobile [~arooni-mo@] has quit [Remote host closed the connection]
18:19-!-ashaffer [] has joined #linode
18:20-!-DanishRolls [] has joined #linode
18:20-!-ponas- [~chiisai@] has quit [Ping timeout: 480 seconds]
18:22-!-arooni-mobile [~arooni-mo@] has joined #linode
18:22-!-jgornick [] has quit [Quit: jgornick]
18:23-!-danblack [~danblack@] has joined #linode
18:25-!-mcinerney [] has joined #linode
18:26-!-Cromulent [] has quit [Quit: WeeChat 0.3.7]
18:27-!-Austin__ [~austin@] has left #linode []
18:27<genku>why is linode
18:28<bacon>Interference from a Geo-Magnetic storm.
18:28-!-linville [] has quit [Quit: Leaving]
18:31-!-jakechapa_ [~jakechapa@] has quit [Remote host closed the connection]
18:35-!-flip_digits [] has quit [Ping timeout: 480 seconds]
18:36<dwhite>So using OpenVPN with Windows, do I still have to use client configuration files (like those found in /usr/share/doc/openvpn-2.2.1/sample-config-files/client.conf), or can I just download the Windows OpenVPN client from and just connect with a username and password?
18:36<GLaDOSDan>dwhite: fwiw OpenVPN-AS is very easy to set up, perhaps look into that
18:37<GLaDOSDan>it's basically openvpn with a max of 2 user limit and a web ui that's easy to use
18:37<dwhite>I ran into some issues even connecting to the server last night.... I've just finished setting it up from scratch again (you're right, It does seem pretty easy to set up)
18:37<dwhite>But that's a client, and not the server, right?
18:37<GLaDOSDan>it'll even give you some bundled thing where it embeds your connection configurations and certificates into a .exe
18:37<GLaDOSDan>it's both
18:37<GLaDOSDan>openvpn-as runs on the server
18:38<GLaDOSDan>you hit it with your browser like yo let me connect
18:38<EugeneKay>dwhite - you're downloading the wrong openvpn windows client
18:38<GLaDOSDan>and it's like here's a nice .exe with all your stuff in it
18:38<EugeneKay>You want the Community one
18:38<EugeneKay>You feed it a client.conf, but you have to call it .ovpn(windows extensions yay)
18:38<dwhite>ahh, gotcha. From here, I presume. (
18:38<dwhite>I'll play around with that
18:39<GLaDOSDan>you can load .ovpn files into the one dwhite linked if it's the same one i'm thinking of
18:39<GLaDOSDan>i... think
18:40<dwhite>there wasn't an option to do that, so I'm guesing you're not thinking of the one I was actually using. =)
18:40<arooni-mobile>having trouble installing a ssl cert and getting nginx to serve ssl... would love any help.
18:41<hawk>arooni-mobile: What have you done and what happens?
18:41<arooni-mobile>hawk, one sec
18:42<GLaDOSDan>dwhite: The one I'm talking about looks like this:
18:43-!-Oloryn1 [] has joined #linode
18:43<dwhite>hmm... that's the one (or very similiar to the one) I tried to use last night
18:47<arooni-mobile>hawk, im having trouble on two problems. 1) when i run "openssl s_client -connect localhost:443 -showcerts" ... i get an error that: " Verify return code: 19 (self signed certificate in certificate chain)" .. secondly when i try to access ; nginx doesnt seem to do anything ... running unbuntu 10.04 lts
18:48-!-Oloryn [] has quit [Ping timeout: 480 seconds]
18:48<hawk>arooni-mobile: Connection refused
18:48-!-ashaffer [] has quit [Quit: Leaving.]
18:48<arooni-mobile>right thats the second issue; even though the iptables doesnt have any rules; and locally i can telnet to port 443
18:49<arooni-mobile>i have no idea why an outside user cannot access 443 port
18:49<hawk>arooni-mobile: What address has nginx bound to for port 443?
18:49<hawk>arooni-mobile: netstat -plnt might be a good start for figuring that out
18:49-!-JoeK [] has quit [Ping timeout: 480 seconds]
18:49<arooni-mobile>thanks i was just going to ask you how ic an find that out
18:53-!-DanishRolls [] has quit [Quit: Red Hat: Make random *ds and force it on everyone. Fun!]
18:53-!-DanishRolls [] has joined #linode
18:53<arooni-mobile>hawk, tcp 0 0* LISTEN 7071/nginx
18:54<hawk>Ok, then I would suspect some kind of firewall nonsense
18:54-!-bacon_ [~rasher@2a01:7e00::f03c:91ff:feae:d964] has joined #linode
18:55<arooni-mobile>this is my iptables:
18:55<arooni-mobile>i mean its weird that i can connect to 443 locally
18:55<arooni-mobile>but not from outside
19:04-!-danblack [~danblack@] has quit [Ping timeout: 480 seconds]
19:06<swaj>oh my god
19:06<swaj>ordered a korean IPS LCD from eBay
19:06<swaj>and it is incredible
19:07<arooni-mobile>am i going to be able to buy those in a year
19:07<arooni-mobile>for the prices youre getting now
19:07<arooni-mobile>im outside the country till 2013 sometime
19:07<swaj>no idea :)
19:07<swaj>but I got a 27" 6ms IPS display for $299.90, shipped :)
19:07<swaj>tempered glass screen, LED backlit
19:08<arooni-mobile>how could i be able to connect to 443 on my server (from my server) but not be able to do it from my dev box... ubuntu 10.04 lts server. and nginx is hadnling 443: tcp 0 0* LISTEN 7272/nginx
19:08<@heckman>Firewall rules?
19:08<arooni-mobile>my firewall rules:
19:08<arooni-mobile>[i.e. there are none]
19:09<swaj>wish I could read Korean :)
19:09<swaj>so I could see what these buttons do
19:09<swaj>but damn it's nice :)
19:11<swaj>2560 x 1440 -- so much real estate!
19:12<staticsafe>I could use another monitor
19:12<swaj>3 now -- varying resolutions :)
19:12-!-hipsters_ [] has joined #linode
19:13<swaj>1920 x 1080 on the left, 2560 x 1440 center, and 1680 x 1050 right
19:13<staticsafe>swaj: i need another 1920x1080
19:17-!-John [] has quit [Quit: User divided by zero]
19:17<dwhite>I'm getting there... I think I'm almost done (with openvpn). Troubleshooting initial connectivity issues now (I've got some juicy info in the logs)
19:18-!-danblack [~danblack@] has joined #linode
19:18<dwhite>hmm... well, maybe
19:18<dwhite>I'm seeing this error when I try to connect: openvpn[4912]: read UDPv4 [ECONNREFUSED]: Connection refused (code=111)
19:19<swaj>client side?
19:20<dwhite>that's on the server... on the client, I see.....
19:20<dwhite>ah, this might be the issue.
19:20<dwhite>Fri Aug 03 19:15:12 2012 TAP-Win32 adapter 'MyTap' not found
19:23-!-taquter [] has quit [Server closed connection]
19:23<linbot>New news from forum: Adding an 'O-Line' to ircd.conf in General Discussion <>
19:23-!-taquter [] has joined #linode
19:23<dwhite>this is a Windows box.... do I "need" the "tap" driver? Readme / install tutorials I've read like to say I do
19:23<dwhite>I'll try commenting that out, see what happens
19:25<dwhite>That worked
19:25<dwhite>I'm in
19:25<arooni-mobile>according to chrome seeing, " However, this page includes other resources which are not secure. These resources can be viewed by others while in transit, and can be modified by an attacker to change the look of the page." ... any idea on how to fix it so user doesnt see it?
19:25<dwhite>and Google things I'm in the UK. :)
19:26<staticsafe>arooni-mobile: all resources on that page should be loaded via HTTPS
19:26<arooni-mobile>staticsafe, whats best way of checking that
19:27-!-skvare1 [] has joined #linode
19:27<staticsafe>arooni-mobile: i would check things like JS and CSS being loaded via HTTP and switch them to HTTPS
19:27<arooni-mobile>staticsafe, so its a nginx config change
19:27-!-skvare2 [] has joined #linode
19:27-!-skvare2 [] has left #linode []
19:27<staticsafe>arooni-mobile: no, change your webapp/site to load everything via HTTPS
19:28<arooni-mobile>staticsafe, but doesnt the server choose if stuff is loaded https or http?
19:28-!-endzyme [] has quit [Remote host closed the connection]
19:28<staticsafe>no, you decide in your code (by specifying http/https)
19:29-!-eyepulp [] has quit [Quit: eyepulp]
19:30<staticsafe>arooni-mobile: use your browser's dev tools to determine which HTTP resources are being loaded and fix those
19:31<dwhite>Thanks, guys, for the help
19:33<arooni-mobile>staticsafe, fixed it; my logo was loading over HTTP
19:33<arooni-mobile>thanks so much for your help!
19:35-!-skvare1 [] has quit [Ping timeout: 480 seconds]
19:38<HoopyCat>t/opic #linode beer: what we're drinking, what you should be drinking, and the best ways to move it from the fridge to your mouth
19:41*EugeneKay pets the HoopyCat
19:43-!-ashaffer [] has joined #linode
19:43-!-hewbert [] has quit [Server closed connection]
19:43-!-hewbert [~josh@] has joined #linode
19:43-!-dwhite [] has quit [Quit: Leaving.]
19:44-!-ponas- [~chiisai@] has joined #linode
19:45-!-ponas [~chiisai@] has quit [Ping timeout: 480 seconds]
19:46<HoopyCat>arooni-mobile: also note that you can omit the scheme: <img src="//"> will choose either HTTP or HTTPS depending on what the page's base URL's scheme is. that can be <3 or it can be </3
19:46<HoopyCat>or gopher, etc
19:46<arooni-mobile>thats useful
19:47<EugeneKay>Or use a <base> element
19:47<HoopyCat>EugeneKay: a <base> would defeat that, if you're hoping to serve the same exact content via HTTP or HTTPS
19:47<EugeneKay>Though that doesn't work with static .html pages - your webapp has to be transport-aware
19:48-!-userme [] has quit [Ping timeout: 480 seconds]
19:48<EugeneKay>Indeed; luckily a simple check of $_SERVER['SERVER_PORT'] will(unless you're using some weird non-standard ports) give you http vs https rather reliably.
19:49<EugeneKay>Or the equivalent for nginx/$LANGUAGE
19:49-!-R5Chris [] has joined #linode
19:49-!-R5Chris [] has quit []
19:51<HoopyCat>EugeneKay: gets into a rabbit hole when you're dealing with caching, etc, tho :-)
19:51<EugeneKay>I'm sure.
19:51<EugeneKay>I'm lazy, though.
19:51<HoopyCat>but yes, try it various ways
19:52<HoopyCat>and indeed, srsly, afk for
19:53-!-triplei [] has quit [Ping timeout: 480 seconds]
20:02-!-danblack [~danblack@] has quit [Read error: Operation timed out]
20:06-!-Dataforce [] has quit [Remote host closed the connection]
20:06-!-Dataforce [] has joined #linode
20:07<Chris___>what about nginx now?
20:07-!-Guest1731 is now known as Bry8Star-{GB
20:08-!-E1ven [] has quit [Quit: Computer has gone to sleep.]
20:09-!-_j45 [] has joined #linode
20:09-!-_j45 [] has left #linode []
20:09<HoopyCat>when will nginx be now?
20:15-!-DrJ [] has joined #linode
20:17-!-trippeh [] has quit [Quit: bring out the duct tape!]
20:18-!-hipsters_ [] has quit [Ping timeout: 480 seconds]
20:23-!-hfb [] has quit [Quit: Leaving]
20:24-!-ansch [~ansch@] has quit [Quit: Bye]
20:25-!-hipsters_ [] has joined #linode
20:28-!-_j45 [] has joined #linode
20:28-!-_j45 [] has left #linode []
20:33-!-trippeh [] has joined #linode
20:39-!-JoeK [] has joined #linode
20:58<amitz>front end nginx, the one man army
20:58<staticsafe>nginx <3
20:59<amitz>nginx, safe for static
20:59<staticsafe>amitz <3
21:01<amitz>i love you too staticsafe with an, admitedly, less than huge lov!
21:03-!-_j45 [] has joined #linode
21:03-!-karstensrage [] has joined #linode
21:04<dcraig>all I see are boxes :(
21:06-!-logichole [] has quit [Read error: Operation timed out]
21:09-!-_j45 [] has quit [Quit: Leaving.]
21:12-!-Cydus [] has quit [Quit: Leaving...]
21:17-!-Cydus [] has joined #linode
21:20-!-userme [] has joined #linode
21:21<linbot>New news from forum: can linode vps support Curl? in General Discussion <>
21:22-!-hfb [] has joined #linode
21:23-!-DanishRolls [] has quit [Ping timeout: 480 seconds]
21:28<linbot>New news from forum: easiest way to override foreign MX records for outgoing mail in Email/SMTP Related Forum <>
21:29-!-danblack [~danblack@] has joined #linode
21:29-!-gadams [~gadams@2600:3c03::f03c:91ff:fe93:23dd] has quit [Server closed connection]
21:30-!-gadams [~gadams@2600:3c03::f03c:91ff:fe93:23dd] has joined #linode
21:33-!-DanishRolls [] has joined #linode
21:33<linbot>New news from forum: can linode vps support Curl? in General Discussion <>
21:40<linbot>New news from forum: Install Virtual Server on a Linode - Is this possible in General Discussion <>
21:40<praetorian>lol curl
21:41<praetorian>that got a pretty quick response.
21:43<Kyh_>lol vps inside a vps
21:43<GLaDOSDan>V P S C E P T I O N
21:45-!-ukitazume [] has joined #linode
21:45<linbot>New news from forum: Install Virtual Server on a Linode - Is this possible in General Discussion <>
21:45<retro|blah>ONLY one page?
21:45<retro|blah>I am disappoint.
21:49<rnowak>as a side note, lxc is awesome
21:58<linbot>New news from forum: can linode vps support Curl? in General Discussion <>
21:58-!-_j45 [] has joined #linode
21:59<retro|blah>Stop using CentOS.
22:04<linbot>New news from forum: can linode vps support Curl? in General Discussion <>
22:05<retro|blah>Palm, meet face.
22:05<staticsafe>[Errno 4] IOError: <urlopen error (-3, 'Temporary failure in name resolution')>
22:08-!-danblack [~danblack@] has quit [Quit: solitude begins]
22:11-!-hipsters_ [] has quit []
22:18-!-danblack [~danblack@] has joined #linode
22:18-!-A-KO [as@2601:a:f00:1f:117a:af26:9bab:3a26] has quit []
22:18-!-skvare [] has joined #linode
22:19-!-_j45 [] has left #linode []
22:28-!-arooni-mobile [~arooni-mo@] has quit [Ping timeout: 480 seconds]
22:29-!-saikat [] has joined #linode
22:29-!-saikat [] has quit []
22:34-!-wheatie [] has quit [Quit: leaving]
22:34<amitz>is the performance of hybrid drive (small SSD as cache + spinning drive) significantly better than spinning drive only and just slightly less than exclusive SSD?
22:35-!-Takyoji [~Takyoji@2602:100:18c5:b0e8:d0f9:eca:95a4:21b6] has quit [Quit: Leaving]
22:36-!-sary_ [~sary@] has joined #linode
22:37-!-ecksit [] has joined #linode
22:38<@mdcollins>A friend has a motherboard that uses his ssd as a cache for his hard drive.
22:38<ecksit>just looking for a vps host and stumbled upon linode. do you use the services?
22:38-!-sary [~sary@] has quit [Ping timeout: 480 seconds]
22:39<linbot>Users with ops are employees of Linode, and know what they're talking about. The rest of us are the ever-so-helpful(?) community. Official Linode contact information:
22:40<ecksit>ah, coolio. not sure how the channel provisioned ops. :p
22:40-!-skvare [] has left #linode []
22:40<ecksit>i develop ruby, rails and php apps. would it be sensible to get two packages or would one suffice?
22:41<swaj>hmm... pretty impressed with the windows virtio drivers so far. Getting great performance on this KVM guest running Windows Server 2008 R2
22:41<@mdcollins>Completely depends on your usage and needs.
22:42<swaj>ecksit: that's a really subjective question. Are you running a large site, or just something small and personal?
22:43<swaj>ecksit: if you're just looking for a personal server to hack around on without major traffic or huge memory demands, a single 512 is more than adequate for a few rails and PHP apps.
22:43<ecksit>one application is a white labelled ticketing application with just over 120 accounts but a little larger then normal db
22:44<swaj>how big is the DB? is hard drive space a concern?
22:44<amitz>mdcollins: noted
22:44<@mdcollins>amitz: Oh and to continue that, he loves it and it caches game files too.
22:44<swaj>amitz: I've heard quite good things about those hybrid drives. Obviously not quite SSD performance, but still quite a bit better than a standard platter.
22:44<ecksit>definitely not something that would be a space concern. it is all deployed to a single code base and the db is probably 2gb max
22:45<@mdcollins>Along with his usual applications and OS.
22:45<swaj>ecksit: rails is a bit of a memory whore sometimes... have you profiled your app to see what it uses generally?
22:46<ecksit>not just yet. as the base app is php and the rails implementation is still under development.
22:48<swaj>oh, so the app is still PHP atm
22:48-!-asedeno [~asedeno@2620:0:102c:0:bd6a:55b9:c520:eb82] has quit [Server closed connection]
22:48<swaj>you'd probably be fine with a 512 if you tuned the DB and used PHP-FPM
22:48-!-asedeno [~asedeno@2620:0:102c:0:e8a5:bd3:9a85:396d] has joined #linode
22:49<ecksit>correct swaj
22:49<swaj>yeah so my recommendation would be to order a 512. use nginx/php-fpm to serve the app, and test it out. You might be pleasantly suprised :)
22:49<swaj>you can always add more nodes if you need them, or make your 512 into something larger :)
22:50*mdcollins waits for ruby to compile
22:50<swaj>I see you have a snail :)
22:50<ecksit>does linode own the datacentre's it utilises?
22:51<rnowak>that's a cinnamon bun fyi, swaj
22:51<amitz>swaj: yeah, the degree of "bit" is what i'm curious of
22:51<swaj>ecksit: nope. They're all co-located at various facilities around the world. There are 5 you can choose from when you create your node.
22:51<@mdcollins>*cough* 6 datacenters */cough*
22:51<rnowak>and the 7th we do not talk about
22:51<swaj>the 7th is at caker's house :P
22:52<@mdcollins>Hey, the first rule of the 7th is we don't talk of the 7th.
22:52<rnowak>that's the 0th akshully
22:52<swaj>I heard jed stole a datacenter when he left... true story.
22:53<ecksit>the no remote ssh is a bit of a bummer though.
22:53<swaj>you can do whatever you want to your node... and they have SSH running by default when you create them :)
22:53<@mdcollins>All distros have SSH on the default image.
22:54<swaj>you have r00t, install anything you desire, so long as it doesn't violate the ToS :)
22:54<ecksit>orly? sorry, i read something about a "lish" and assumed ssh access was only available from the console.
22:54<@mdcollins>LISH is if you break something ;)
22:54<swaj>lish is out-of-band access in case you b0rk your network config... it's like a virtual console
22:54<ecksit>well, that changes everything.
22:55<swaj>it's equivalent to "physically" sitting in front of your node... except virtually... over the internet... or something :)
22:55<rnowak>except it is more like a serial console! which could be in front of it, but maybe not!
22:55*akerl serializes rnowak
22:55<@mdcollins>Oh the uncertainty!
22:55<rnowak>DUN DUN DUN
22:56<swaj>they even have an ajax lish interface that works in a web browser, although it kinda sucks, and I've been campaigning for them to use shellinabox as a replacement.
22:57-!-seanh-ansca [] has quit [Quit: Leaving.]
22:57<@akerl>swaj: We already implemented a badass replacement for AJAX: openssh
22:57<@mdcollins>Comes on every Linode!
22:57<swaj>akerl: except... I break shit sometimes, and work blocks SSH :(
22:58<@akerl>swaj: LISH listens on 2000 and 443 also. Are they doing DPI?
22:58<swaj>work firewalls are bastards
22:58<@akerl>well damn
22:58<swaj>yeah I work at MasterCard... we have really fancy firewalls that make my life difficult daily :(
22:59<@akerl>$$ is srsbsns or something :P
22:59<rnowak>protip: don't break your shit while at work
22:59<swaj>I mean... they think the global processing datacenter (in the same building I work in) is like... really important
23:00<GLaDOSDan>When is the Moon Linode DC coming
23:00<@mdcollins>*sigh*.. git was not installed and the script called git.
23:00<GLaDOSDan>I've been waiting ages and quite frankly I'm fed up of waiting
23:00<@akerl>mdcollins: Sounds like poor error checking
23:00<amitz>swaj: wait, so I can just bring scissors and cut mastercard from the rest of the world? nice!
23:00<@mdcollins>GLaDOSDan: I'll be happy to forward your feature request on! :P
23:00<rnowak>mdcollins: that's how I felt compiling xen shit; the reaction afterwards: oh you fucker
23:00<GLaDOSDan>Thank you.
23:01<swaj>amitz: yeah somebody did that once... and then we instantly failed over to kansas city and didn't lose a single transaction :)
23:01<swaj>amitz: AT&T decided to go digging once and was like "oops.. was that fiber important?"
23:01<amitz>swaj: damn, I guess I will need 2 scissors then!
23:01<@mdcollins>I was trying to use the gitlab set up script..
23:02<amitz>swaj: col
23:03<@jchen>its pretty cool
23:04<@jchen>why do you not have git installed
23:04<linbot>IRC supports complete sentences. Less <CR> more content, please.
23:05<@mdcollins>Fresh DomU on my home server :P
23:06<@mdcollins>Just had to manually do step 3.
23:07<auraka>jchen: hipster version control.....back in my day we swam upstream both ways and used svn!
23:07<@jchen>back in my day
23:07<dwfreed>mdcollins: you'd think that installing git would be an important first step to installing gitlab :)
23:07<dwfreed>auraka: I used SVN a while back (about 4-5 years ago)
23:08<@mdcollins>Thought the gitlab script would be smart enough to figure that one out :P
23:08<auraka>dwfreed: you aren't old...hush it youngin
23:08<auraka>grown ups are speaking!
23:08<dwfreed>auraka: define "old"
23:08<@mdcollins>*sigh* missing zlib..
23:08<auraka>dwfreed: did you watch thundercats as a child?
23:08<dwfreed>auraka: uh, I think so
23:08<auraka>no you didn't
23:09<auraka>because you would have remembered the awesomeness
23:09<@jchen>why are you missing so many prereq's
23:09<dwfreed>auraka: not original runs, sure, but iirc, there were reruns on
23:09<@jchen>also what crappy script are you using
23:09<@mdcollins>The crappy script provided by gitlab!
23:10-!-Duke [] has joined #linode
23:11<@jchen>sudo apt-get install -y wget curl gcc checkinstall libxml2-dev libxslt-dev sqlite3 libsqlite3-dev libcurl4-openssl-dev libreadline6-dev libc6-dev libssl-dev libmysql++-dev make build-essential zlib1g-dev libicu-dev redis-server openssh-server git-core python-dev python-pip libyaml-dev sendmail
23:11<@jchen>^ that should have grabbed all the stuff you need
23:11<@jchen>it's like the first thing.... i call PEBCAK
23:11<@mdcollins>Pssh. I'm fixing it nao
23:11<dwfreed>why does it need redis? ...
23:11-!-TecnoBrat [] has quit [Read error: Connection reset by peer]
23:11<@akerl>dwfreed: Because it's web scale?
23:11<@heckman>dwfreed: job processing
23:11<@jchen>dwfreed: cloud seo
23:11<@heckman>uses resque
23:11<swaj>jchen: lol sendmail
23:11<@akerl>heckman: Cloud?
23:12-!-TecnoBrat [] has joined #linode
23:12<@jchen>swaj: password reset
23:12<@heckman>mdcollins: you still breaking it?
23:12<swaj>jchen: exim works wonders :)
23:12*auraka gives akerl a cookie and will shut up about arch for one month
23:12<@mdcollins>I decided to manually do the recompile of ruby
23:12<@heckman>I'd be done by now. :3
23:12<@jchen>hey, i didnt make gitlab
23:12<swaj>mdcollins: rvm install ree
23:12<@mdcollins>I fixed the git part and did the commands that failed
23:12<@mdcollins>swaj: no rvm >_<
23:13<@heckman>mdcollins: what command?
23:14<@mdcollins>heckman: all the ones that required git to be installed and failed when it wasnt
23:14<@mdcollins>I blame the script :P
23:15<@heckman>You ran the script?
23:15<@heckman>I did it by hand
23:18<rnowak>when little mdcollins grows up, he aspires one day to be just like uncle heckman
23:18<linbot>New news from forum: easiest way to override foreign MX records for outgoing mail in Email/SMTP Related Forum <>
23:19<staticsafe> linode ftw \o/
23:25-!-sivy [] has quit [Remote host closed the connection]
23:25<@mdcollins>Why do it by hand when the script works (mostly).
23:26<rnowak>03:25:31 @mdcollins: ?hy do it by hand when the script works (mostly).
23:26<rnowak>you fixed it real good
23:30<linbot>New news from forum: Linode Sponsored: The Perl Foundation Community Symposium in General Discussion <>
23:31-!-_j45 [] has joined #linode
23:32<staticsafe>inb4 lolperl
23:32-!-XSleeper [] has quit [Server closed connection]
23:33-!-XSleeper [] has joined #linode
23:34<rnowak>what is this, 1995?
23:34<staticsafe>well rnowak is creative as always
23:34<rnowak>I never said I was, but at least you know I am consistent!
23:35<@mdcollins>Y u has so many bugz apartments?
23:35<rnowak>I'm sure it will be lots of fun and a success, too bad it is about perl!
23:36-!-A-KO [as@2601:a:f00:1f:c9ce:1c93:c4d1:5c8e] has joined #linode
23:37-!-ukitazume [] has quit [Ping timeout: 480 seconds]
23:38<@mdcollins>*sigh* time to compile ruby again..
23:38<@mdcollins>Missing yaml support.
23:39<@mdcollins>GitLab better be worth all this.
23:39<@heckman>mdcollins: what did you do to make it angry?
23:39<@heckman>All worked for me
23:39<@mdcollins>Barebones Deb6?
23:40<@mdcollins>I've literally installed nothing on this image.
23:40-!-userme [] has quit [Quit: userme]
23:41<rnowak>how about figuratively?
23:41-!-eyepulp [] has joined #linode
23:42-!-userme [] has joined #linode
23:50<@heckman>mdcollins: from the Linode image, so yes barebone.
23:51<@heckman>The image I lovingly built. :P
23:51<@mdcollins>Maybe I'll just steal a copy if this doesn't work :P
23:52<@akerl>Windows 7
23:52<@heckman>Wait, thats not a Linux ISO.
23:53<@mdcollins>My Win7 install works just fine. It's this ruby/gitlab stuff that's giving me problems.
23:54<@heckman>User error
23:54<@heckman>Follow the instructions step by step... :x
23:54<@mdcollins>I'll make it work!
23:55<@heckman>Need help?
23:55<@mdcollins>Not so far. I'm multi tasking.
23:56<@mdcollins>Found out why zombies wont spawn in daylight.
23:56<@heckman>Oh, what was it?
23:57<@mdcollins>From #bukkitdev "Mobs all have a canSpawn method and the zombie one checks the time of day"
23:57<@akerl>There's way too much op in this party
23:57<@heckman>Was kind of right
23:57-!-mode/#linode [-o akerl] by heckman
23:57<@heckman>Alright, that's better. :P
23:57-!-mode/#linode [+o akerl] by heckman
23:57<@mdcollins>I think it's slightly different cause they spawn in water, but yeah.
23:58-!-mode/#linode [-o akerl] by akerl
23:58<@mdcollins>Die netbeans!
23:59-!-VS_ChanLog [] has left #linode [Rotating Logs]
23:59-!-VS_ChanLog [] has joined #linode
23:59<linbot>Point (0.47870373, 0.07970900) lies within the unit circle. Hits: 513 of 648 (π ≈ 3.166666666666667 - 0.025074013076873). http://π
23:59<@mdcollins>2 mins early..
23:59<swaj>netbeans, lol.... I tried to use it a few times... Eclipse is just so much better
23:59*akerl dubs himself Mance Rayder and goes to rally the non-snailed wildlings
23:59<@Praefectus>laker: watch out for snow
23:59<@mdcollins>I disagree, but such is the nature of opinions.
---Logclosed Sat Aug 04 00:00:37 2012