Back to Home / #linode / 2008 / 07 / Prev Day | Next Day
#linode IRC Logs for 2008-07-10

---Logopened Thu Jul 10 00:00:22 2008
---Daychanged Thu Jul 10 2008
00:00-!-digx [] has joined #linode
00:01<pygmalion>there is a mikegrb in the #gopherproject channel on freenode. same one?
00:03<ninjaPo>SelfishMan: good idea. compared to other messages in my spam folder, the headers on the ones I sent look much better so i hope the problem will be corrected. i have not tested with yahoo or hotmail; that is among my next steps. I also tried to install DomainKeys Identified Mail (DKIM) but couldn't get it to install properly
00:05<bob2>you'd want DKIM, not domainkeys
00:06<ninjaPo>bob2: right; that's what i meant...
00:06<bob2>oops, sorry, just skimmed incorrectly :)
00:06-!-Schroeder [] has quit [Ping timeout: 480 seconds]
00:07<SelfishMan>DKIM will help but google won't care much about that
00:09<ninjaPo>SelfishMan: I think you may be onto something with the 'community' idea---i will give that a whirl, do some tests with other providers and report back.
00:10<SelfishMan>I haven't dealt with Google's mail systems for about a year. Every method you use to prove yourself they will use but in the end it is all about community.
00:10<SelfishMan>anyway, I have to run for a bit
00:13-!-pygmalion [] has quit [Quit: pygmalion]
00:18-!-SelfishMan [] has quit [Ping timeout: 480 seconds]
00:23-!-DennisTT [] has joined #linode
00:25-!-Schroeder [] has joined #linode
00:27-!-ninjaPo [] has quit [Remote host closed the connection]
00:31-!-mode/#linode [-q *!*] by FloodServ
00:35-!-euph [] has quit [Ping timeout: 480 seconds]
00:47-!-SelfishMan [] has joined #linode
00:50<SelfishMan>With all the people in this room I think I've maybe seen 15 unique people say something and that includes linbot and mikegrb(bot)
00:51*SelfishMan listens to the sound of crickets chirping
00:54<@mikegrb>mmm cake
00:54<exor674>(trigger fail :D)
00:56<SelfishMan>oh wow, totally wrong keys there
00:57<SelfishMan>i quit
00:58-!-TheFirst [] has quit [Ping timeout: 480 seconds]
01:00<booja>sorry SelfishMan
01:01<SelfishMan>sorry? why?
01:01<booja>for not talking more
01:01<SelfishMan>my keyboard is defective. I can't figure out how to use it.
01:01<exor674>SelfishMan: throw it under a bus
01:01<SelfishMan>it's attached to my laptop
01:01<booja>or throw yourself under the bus
01:02<SelfishMan>booja doesn't love me anymore
01:02<booja>that's what SHE said
01:02<SelfishMan>I still feel dirty every time I touch a ReiserFS filesystem.
01:03<exor674>SelfishMan: do you mean a disk image? because I don't think touch works on /dev entries <_<
01:04<SelfishMan>Funny you should say that because I am actually using a mounted ReiserFS disk image.
01:05-!-brtb [] has joined #linode
01:05-!-brtb is now known as Guest15
01:06<SelfishMan>Um...The counter on rapidshare is counting up, not down.
01:13<booja>oh no
01:13<booja>stand on your head and spin around 3 times
01:13<booja>else the internet will asplode!
01:14<SelfishMan>I'm too lazy for that. I'll get my pr0n somewhere else.
01:23-!-Talman [~ender@] has quit [Read error: Connection reset by peer]
01:27<CaptObviousman>how selfish!
01:28*CaptObviousman is going to keep doing that
01:28<CaptObviousman>just so you know
01:28<CaptObviousman>I'm all about the low-hanging fruit
01:29<SelfishMan>I don't get it
01:29*SelfishMan hates shellfish
01:37-!-Internat [] has quit []
01:37-!-jcn [] has quit [Ping timeout: 480 seconds]
01:49-!-getsmart [] has joined #linode
02:12-!-exor674 is now known as exor|zzz
02:12-!-Dreamr_3 [] has quit [Quit: This computer has gone to sleep]
02:14-!-Dreamr_3 [] has joined #linode
02:20-!-Dreamr_3 [] has quit [Read error: Connection reset by peer]
02:20-!-Dreamer3 [] has joined #linode
02:25-!-Kassah [] has quit [Ping timeout: 480 seconds]
02:36-!-Kassah [] has joined #linode
02:41-!-Dengar [] has joined #linode
02:41-!-Deetz [lithium@] has joined #linode
02:44<SelfishMan>Time to test some odd shell commands on a test linode
02:46-!-digx [] has quit [Quit: digx]
02:47-!-Rifkin [] has joined #linode
02:47<Rifkin>I just upgraded to a 540 and WOO HOO
02:48<Rifkin>I wouldn't have guessed the difference would be so great
02:54-!-Internat [] has joined #linode
02:55-!-Deetz [lithium@] has quit [Remote host closed the connection]
02:56-!-Kassah [] has quit [Ping timeout: 480 seconds]
03:06-!-Kassah [] has joined #linode
03:08-!-leslie [] has quit [Remote host closed the connection]
03:14-!-Deetz [lithium@] has joined #linode
03:22-!-getsmart [] has quit [Quit: Ex-Chat]
03:27-!-Deet1 [lithium@] has joined #linode
03:29-!-row [] has joined #linode
03:33-!-Deetz [lithium@] has quit [Ping timeout: 480 seconds]
04:02-!-Deet1 [lithium@] has quit [Ping timeout: 480 seconds]
04:12-!-Schroeder [] has quit [Ping timeout: 480 seconds]
04:13-!-Deetz [lithium@] has joined #linode
04:15-!-SelfishMan [] has quit [Ping timeout: 480 seconds]
04:26-!-Deetz [lithium@] has quit [Ping timeout: 480 seconds]
04:34-!-Deetz [lithium@] has joined #linode
04:35-!-spasmface [] has quit [Quit: Leaving...]
04:35-!-kupesoft [] has quit [Remote host closed the connection]
04:54-!-SelfishMan [] has joined #linode
04:57-!-Dengar [] has quit [Remote host closed the connection]
05:08-!-Kassah [] has quit [Ping timeout: 480 seconds]
05:18-!-Kassah [] has joined #linode
05:43-!-xitology [] has joined #linode
05:45<linbot>SelfishMan: Fremont360 - 0, Fremont540 - 0, Fremont720 - 0, Fremont1080 - 0, Fremont1440 - 0, Fremont2880 - 0
05:45<SelfishMan>well, so much for the 720's
05:45<linbot>SelfishMan: Newark360 - 38, Newark540 - 8, Newark720 - 4, Newark1080 - 3, Newark1440 - 2, Newark2880 - 1
06:20-!-pygmalion [] has joined #linode
06:21-!-pygmalion [] has quit []
06:40-!-ondrej [] has joined #linode
06:45-!-Navi` [~Navi`] has joined #linode
06:48-!-Deetz [lithium@] has quit [Quit: Leaving.]
06:49-!-Deetz [lithium@] has joined #linode
06:57-!-Deetz [lithium@] has quit [Ping timeout: 480 seconds]
07:20<Internat>wow.. quiet channel for once tongiht
07:20-!-TheFirst [] has joined #linode
07:20<SelfishMan>yes it is
07:21-!-Deetz [lithium@] has joined #linode
07:21-!-bonhoffer [] has joined #linode
07:22<chesty>everyone is out lining up for the iphone
07:22<Internat>lol. more then likely
07:22<SelfishMan>screw all things apple
07:22<bonhoffer>i am frustrated with ssmtp that i am trying to set up on linode -- using google. messages are in my sent box, but the address is not appending the domain. i.e. if this is for the email is being sent to jerry without his domain
07:23<mwalling>bonhoffer: the *envelope* MAIL FROM or the *message* From header?
07:24<mwalling>also, having cashflow back after both me and my wife changing jobs ++++
07:24<bonhoffer>the message is not sending
07:25<bonhoffer>it is in my sent, but not in inbox
07:26<mwalling>your google sent box or your local sent box?
07:26<Dave>I think im the only geek not running out to buy an iphone
07:27<mwalling>Dave: i'm not
07:27<mwalling>because if its your local, your mail client puts that message there, not your MTA
07:27<Dave>got one already?
07:27<mwalling>Dave: nope
07:27<mwalling>Dave: too cheap to pay that much
07:27<mwalling>i want an old treo and a cheap data plan
07:29-!-Navi` [~Navi`] has quit [Quit: Navi`]
07:33-!-bonhoffer [] has quit [Ping timeout: 480 seconds]
07:33<mwalling>guess he didnt want to answer
07:34-!-Akazawa [] has quit [Ping timeout: 480 seconds]
07:41-!-TheFirst [] has quit [Ping timeout: 480 seconds]
07:43-!-soul9 [] has joined #linode
07:43<soul9>linode doesn't support paypal, right?
07:54<chesty>any other questions you'd like to answer?
08:02<soul9>heh :-)
08:02-!-soul9 [] has quit [Quit: leaving]
08:12-!-r3z`` [] has joined #linode
08:16<mwalling>chesty: they *kinda* do, but not direct paypal transactions
08:16<mwalling>(i pay caker from my paypal account)
08:19-!-r3z` [] has quit [Ping timeout: 480 seconds]
08:19<chesty>i've seen other people say they don't support paypal because of fraud
08:19<chesty>how does yours work?
08:19-!-Deet1 [lithium@] has joined #linode
08:25-!-Deetz [lithium@] has quit [Ping timeout: 480 seconds]
08:26-!-Sputnik7 [] has quit [Quit: -=SysReset 2.53=-]
08:26<mwalling>chesty: i have a paypal debit card ;)
08:27<@jadoba>mwalling cheats
08:28<mwalling>imma cheating whore
08:28*SelfishMan saves the log for future blackmail
08:28-!-Deet1 [lithium@] has quit [Ping timeout: 480 seconds]
08:31*mwalling shrugs
08:36*mikegrb emails cwalling
08:40*mwalling intercepts
08:43*jadoba receives phone call
08:46-!-torn59 [~tulin@] has joined #linode
08:46-!-torn59 [~tulin@] has quit [autokilled: This host is believed to be a home to spammers. - Contact for help. (2008-07-10 12:46:47)]
08:57-!-Talman [~ender@] has joined #linode
08:58-!-kupesoft [] has joined #linode
09:03-!-Talman [~ender@] has quit [Read error: Connection reset by peer]
09:04*mwalling waves at jadoba
09:04*jadoba gets back to work
09:04<@caker>please, don't feed the animals
09:05<mwalling>you said you received a phone call :P
09:05<mwalling>i just made sure you werent lying
09:05<Nigel>mikegrb: oh btw, I don't expect you guys to mess with getting the io stats on host56 going, seems too much pain for IO Stats :)
09:05<@tasaro>Nigel: i think he has a fix
09:06-!-Sputnik7 [] has joined #linode
09:06<@caker>tasaro: nah, this is something different
09:08<Nigel>mikegrb: what you could do, is "IO Graphs are not available on this host" or something though :)
09:09<Nigel>caker: has anyone else reported incorrect times on the dashboard recently, mine are an hour out (although set correctly - +12)
09:10<@caker>Nigel: it's because it uses our daylight savings
09:10<Nigel>I thought it might have been that :)
09:12<Nigel>thanks guys!
09:12*Nigel heads off to bed
09:16-!-jcn [] has joined #linode
09:18-!-Bass10 [] has joined #linode
09:19-!-Deetz [lithium@] has joined #linode
09:25-!-sveiss [] has quit [Quit: sveiss]
09:31-!-linville [] has joined #linode
09:39-!-internat1 [] has joined #linode
09:39-!-Internat [] has quit [Read error: Connection reset by peer]
09:44-!-Talman [~Talman@] has joined #linode
09:44<Talman>Sup from Pidgin.
09:45<path->no funpidgin?
09:45<Talman>I don't even know what that is, this is a locked down terminal I'm using Portable APps on a Sony Memory stick.
09:45-!-robg [] has quit [Quit: Bye!]
09:46<Talman>And by locked down, I mean I can't even right click. :)
09:46<Dave>path-: you mean carrier
09:46-!-Bryanstein [] has quit [Ping timeout: 480 seconds]
09:46<Talman> How do I kill a process tree that was started in lish? Log in lish?
09:48<Dave>Talman: probably
09:53<Talman>Its fragged, postfix install, gonna have to format.
09:53-!-Bryanstein [] has joined #linode
09:53<path->carrier doesn't sound very fun
09:53<path->sounds like a lot of work
09:54<Talman>That's new.
09:54<Talman>No configuration profile exists.
09:55<Talman>Yet an OS is running?
09:55*Talman turns off, wipes drives, restarts.
09:56-!-r3z`` is now known as r3z
09:56<path->Talman: you going to move all your porn to your linode so the customs agents can look at your laptop?
09:57-!-mendel [] has quit [Quit: Leaving.]
09:58<Talman>Hell no.
09:58<Talman>Customs can go to hell.
09:59*Talman wonders how I would boot my Linode when the disks are being made? :)
10:00<Talman>Also, the bears must have cell phones. I'm in Glacier Bay, and I get spotty cell coverage.
10:01<scott>CAN YOU HEAR ME NOW?
10:02<Talman>AT&T: NO.
10:03-!-Internat [] has joined #linode
10:03-!-internat1 [] has quit [Read error: Connection reset by peer]
10:05-!-mendel [] has joined #linode
10:05<Talman>I do enjoy how they lock the internet cafe terminals down, the whole IE policy gimmick.
10:05<Talman>So, of course, they put card readers on every terminal.
10:06<Talman>I just put in ap ortable apps memory stick card, and boom, I get firefox and anything else I care to run.
10:06<path->you hacked IE?!?!
10:06<scott>who hasnt?
10:07<Talman>They disabled the run menu, etc.
10:07<Talman>Disabled file.
10:07<Talman>Put a memory stick in, though, and LOL it opens the "New Device Found" wizard.
10:08-!-Isvara [] has joined #linode
10:08<path->surprised they didn't disable autorun
10:08<path->i guess that's not related
10:09<Talman>Yeah, I'm surprised too.
10:09<Talman>People keep putting viruses as "anti-virus and anti-spyware" on the crew internet terminals.
10:09-!-agentbleubleu [] has joined #linode
10:09<Talman>THere's a "WebSheriff Toolbar" on Crew9.
10:12<@caker>mmm .. log colorizer: ccze
10:12<@caker>tail -f ....... | ccze -A <-- pretty
10:12<@caker>it also has some curses mode that I haven't messed with
10:12<Talman>hmm, looks like I might want that...
10:13<Talman>in ubuntu repos? :)
10:13<mwalling>homepage link is broken on FM
10:13<Talman>oh wtf.
10:13<Talman>wtf is "apparmor."
10:13<mwalling>home page is broken on
10:14-!-Rifkin [] has quit [Quit: leaving]
10:15<Isvara>Talman: Some Linux security thing
10:15<mwalling>damnit, where the hell does one get the source?
10:15<Talman>Great, its breaking my install of mysql.
10:17<Talman>oh for FUCK's sake. mysql DEPENDS on apparmor?!
10:17<mwalling>(automated dependancy resolution)
10:17<straterra>ubuntu ftl
10:17<Bdragon> ftw
10:17*Bdragon ducks
10:17<Talman>What the HELL is this. :|
10:18<straterra>Welcome to shitbuntu. Please enjoy your stay.
10:18<@mikegrb>! people still use slackware?!?!
10:18<mwalling>Talman: use slackware
10:18<Isvara>How hard can it be? sudo apt-get install mysql
10:18<Talman>7.10 didn't have this shit.
10:18<Talman>isvara: Error: App Armor loading... failed.
10:18<Talman>For some reason, mysql-server now has apparmor and apparmor-utils as dependencies.
10:19<mwalling>works for me
10:19<Talman>Good for you. :)
10:19<mwalling>no, thats what WFM is
10:20<Talman>I'm referring to Isvara's "how hard can it be."
10:21<Talman>Alright, after uninstalling and reinstalling the packages, apparmor is history. Twice it took. Weird.
10:21<guinea-pig>Talman: it has them as Recommends: not Depends:
10:22-!-lakin [] has joined #linode
10:23<scott>aptitude would grab everything
10:24-!-Navi` [~Navi`] has joined #linode
10:24<Talman>Allright... echo > /etc/hostname
10:24<Talman>and /etc/init.d/hostname start
10:24-!-lakin [] has quit []
10:25<Talman>hostname -f returns, not, and netstat reports getnameinfo failure.
10:26<Dave>Talman: hostname -F /etc/hostname
10:27<Talman>hostname -f still replies with the domain only, not the hostname + domain.
10:27<Talman>Is this the proper response?
10:27<Dave>mine does
10:27<Dave>well... it should respond
10:28<Talman>Yes, that's what I thought it should do.
10:28<Dave>then again that is a cname, so perhaps something funky is going on?
10:28<Talman>See, its because I'm on internet that has 200 minutes, and I've used 47 already.
10:28<Talman>See how this works? It won't work right the first time.
10:28<Talman>I am paying for internet.
10:28<Talman>Per minute.
10:29<Talman>Hence, nothing will work correctly the first time. To eat the minutes.
10:29<guinea-pig>did you also set it in /etc/hosts?
10:30<guinea-pig>whoops, gotta run. workies
10:30<Talman>AH, I see.
10:30<Talman>Is that under, the IP, or both?
10:30<Talman>(No, I forgot)
10:31<Dave>should be both
10:32<Dave>having said that, mine isnt
10:32<jetlag>not under 127...
10:32-!-Schroeder [] has joined #linode
10:34<Talman>It should be under the static assigned by linode?
10:34-!-Internat [] has quit [Read error: Connection reset by peer]
10:34<Talman>Also, hmm.
10:34-!-Internat [] has joined #linode
10:34<Talman>postconf -e 'myhostname =' or '' or 'tamaki...'?
10:36<StevenK>Talman: My MX records are set as mail, and the server answers as mangled
10:38<Talman>"server answers as mangled" doesn't parse for me.
10:38<Talman>I used the FQDN.
10:39<StevenK>Talman: The machine is called "mangled" :-)
10:39<StevenK>IE, the same as you, the FQDN
10:42<Talman>copy that. ok.
10:54-!-Sputnik7 [] has quit [Quit: -=SysReset 2.53=-]
10:56<Talman>I seem to remember something about needing to set something in xen's proc before using NTP?
10:57<path->ccze is nice
10:58<HoopyCat>Talman: odds are good you don't need to use NTP, but if you do, it's /proc/sys/xen/independent_wallclock or somewhere around there
11:00<HoopyCat>some clocks are held for charity, and some for fancy dress; but when they're held for, they're the clocks that i like best.
11:00-!-Sputnik7 [] has joined #linode
11:01-!-internat1 [] has joined #linode
11:01-!-Internat [] has quit [Read error: Connection reset by peer]
11:04<Talman>... getnameinfo failed.
11:04<Talman>I have /etc/hosts setup with the IP address and the fdqn + hostname.
11:04<Talman>ANd just made sure that its an a record in Linode DNS manager.
11:05-!-Sputnik7 [] has quit []
11:05-!-Deetz [lithium@] has quit [Quit: Leaving.]
11:08-!-Sputnik7 [] has joined #linode
11:16<Talman>Hmm, anyone know a (free) SMS message gateway gimmick that warns me via SMS of problems/conditions on the server?
11:17-!-xitology [] has quit [Quit: Ex-Chat]
11:19<tjfontaine>yes email->sms
11:19<@tasaro>create an sms "user":
11:19<@tasaro>and get the Linode alerts
11:19<mwalling>nice idea
11:20-!-ryan8403_laptop [] has joined #linode
11:21<Talman>That's one way, I was thinking more a package in repo that told me certain things. The linode could be running fine, but apache could be off, etc.
11:21<tjfontaine>you want nagios
11:21<anderiv>tasaro: I probably shouldn't have even pointed him to phplist. Not that phplist has the chutzpah to be a real spaminator, though...
11:21<anderiv>err Talman rather
11:21<Talman>He'd have to figure out how.
11:22<@tasaro>anderiv: he'll be gone with any abuse tickets anyway
11:25<Talman>I see FF3 hates self signed certificates, to the point of giving a huge warning.
11:25<Talman>And making you do 5 clicks.
11:25<anderiv>Talman: yes, discovered that recently as well.
11:25<Isvara>(As it should.)
11:25<Hobbsee>tasaro: pre-emptively filed ones, too?
11:26<tjfontaine>Isvara: 1 click was fine
11:26<Isvara>People do the first click without thinking.
11:26-!-bubblewrapz [] has joined #linode
11:26<@caker>Destroy system: cancel or allow
11:27<bubblewrapz>i cant log into my linode.
11:27<StevenK>Destroy system: yes or maybe
11:27<bubblewrapz>with putty
11:27<Isvara>"Don't not fail to succeed to not destory the system? Yes or no."
11:27<Isvara>Think fast!
11:27<anderiv>bubblewrapz: you have an sshd running?
11:27<StevenK>Isvara: Argh.
11:27<bubblewrapz>idk what that is :\
11:27<Talman>bubblewrapz, use lish.
11:28<StevenK>Isvara: No
11:28<bubblewrapz>it worked fine, but i deleted my linode, and now im trying to log in
11:28<bubblewrapz>and it wont connect
11:28<Talman>Did yuo make... another linode?
11:28<anderiv>bubblewrapz: did you re-deploy another?
11:28<bubblewrapz>like i redeployed a new distra and everything
11:28<anderiv>is it running?
11:28<bubblewrapz>its been up for about 2 minutes
11:28<bubblewrapz>mayb 5
11:28<Bdragon>(better go catch it)
11:28<Hobbsee>did you run rm -rf / ?
11:28<@caker>shoooo beee doo be deooooooo
11:28<Talman>use lish.
11:29<Bdragon>Anyhow, use lish, you might have to twiddle sshd
11:29<bubblewrapz>Hobbsee: no, i never had to do anything before, i just opened putty and conencted.
11:29<StevenK>Isvara: Did I sort out the horrid mess of double and triple negatives? :-)
11:29<Hobbsee>well, clearly that's your problem.
11:29<StevenK>I didn't think Linode's had sshd installed by default
11:29<bubblewrapz>well clearly how
11:29<Hobbsee>StevenK: they do
11:29<bubblewrapz>it worked the first time
11:29<bubblewrapz>why nto now...i did everything the same..same distro same swap
11:30<StevenK>Okay, so at this point you should be using lish and debugging.
11:30<Isvara>StevenK: I don't know, because I didn't ;-)
11:30<Talman>Did you delete the partitions?
11:30<Hobbsee>because someone's hacked your machine, and locked you out?
11:30<StevenK>Isvara: Haha
11:30<Talman>Or are you reusing the same disk images?
11:30<Hobbsee>or because you forgot to make the required sacrifices
11:30<StevenK>Hobbsee: In 5 minutes? Sounds like a record
11:30<StevenK>Maybe his co-workers are exacting revenge
11:31<Hobbsee>ISTR that 'doze machines can be hacked in way less time than that.
11:31<bubblewrapz>i deleted everything
11:31<bubblewrapz>and am starting fresh
11:31<StevenK>Hobbsee: Sure, but Linux is a bit more difficult
11:32<Hobbsee>they only give you a customised distro the first time. it's to encourage you to only build it once. you'll have to use your first chance of slackware.
11:32<bubblewrapz>who is he talking to?
11:33<Hobbsee>if you're meaning me, a) i'm not a he, and b) YOU.
11:33<@caker>bubblewrapz: using the correct IP?
11:33<Karrde>Your teeth have the strenth of the proletariat.
11:33<bubblewrapz>caker, i think so
11:33-!-Internat [] has joined #linode
11:33<@caker>bubblewrapz: well, one of them works, the other doesn't (as it should on a fresh system)
11:34<Talman>YOur rectum shall bleed under the might of the unbeliever.
11:34-!-internat1 [] has quit [Read error: Connection reset by peer]
11:34<Talman>So, how about those internets?
11:34<StevenK>bubblewrapz: Can you ping the IP?
11:35-!-bubblewrapz [] has quit [Quit: CGI:IRC 0.5.6 (2005/02/09)]
11:35-!-bubblewrapz [] has joined #linode
11:35<bubblewrapz>i died
11:35<StevenK>I suspect your connection did, and you didn't.
11:35-!-TJF [] has joined #linode
11:35<bubblewrapz>well um caker, ill try my second ip.
11:35<StevenK>Dead people tend not to reconnect.
11:37<bubblewrapz>ah brb
11:37-!-bubblewrapz [] has quit []
11:37<Isvara>StevenK: Dead people reconnect with the living all the time. Don't you watch TV?
11:38<StevenK>Isvara: I thought that required Jennifer Love Hewitt to work?
11:39<Talman>SOmehow I doubt that she's present.
11:39-!-euph [] has joined #linode
11:39<tjfontaine>look sir tjf!
11:40<Isvara>StevenK: She's fictional, though. There are real people out there who need only your belief. Oh, and your cash. Your sweet, sweet cash.
11:42-!-bubblewrapz [] has joined #linode
11:43<StevenK>Isvara: Ah, but Jennifer Love Hewitt isn't. The character she portrays is ...
11:45-!-lakin [] has joined #linode
11:45<Talman>Windows Mobile 6 is designe to look like Vista?
11:45<Talman>I didn't even notice.
11:46<bubblewrapz>i was using the wrong ip
11:46<bubblewrapz>its good now :X
11:46<bubblewrapz>the first thing i should do is apt-get update; apt-get -u upgrade?
11:47<Talman>Go use the perfect server for your ubuntus.
11:47<Talman>its on howto forge.
11:48-!-jimcooncat [] has joined #linode
11:51-!-bubblewrapz [] has quit [Quit: CGI:IRC 0.5.6 (2005/02/09)]
11:53-!-Bdragon [] has quit [Quit: HydraIRC -> <- Now with extra fish!]
11:57<HoopyCat>mmm, brains.
11:58<Isvara>Lovely brains?
11:58<HoopyCat>i can smell your spicy brains.
11:59<Isvara>That would explain the heartburn I had this morning.
11:59<Isvara>Damned spicy brain.
11:59<scott>thats not brains that you smell...
11:59*caker discovers grep --color
11:59<@caker>grep --color root /etc/passwd
11:59<@caker>neato :)
11:59<scott>caker: also try agt-get moo
12:00<scott>i can play with that for hours
12:00-!-pbryan [~pbryan@] has joined #linode
12:00<@caker>just like your mom
12:00*scott goes for lunch
12:00-!-Bdragon [] has joined #linode
12:01<praetorian>(just for tj)
12:02<tjfontaine>where's mine?
12:02<praetorian>at&t are holding on to it for you
12:02<tjfontaine>for another 20 hours?
12:03<@caker>,2_2.0_5A347_Restore.ipsw <-- 2.0 firmware !
12:03<praetorian><3 .au
12:03<tjfontaine>caker: does your dad have his?
12:03<@caker>no. Apple is insane about when they arrive at the stores, and when they allow employees to get them
12:04<@caker>He said their training Sunday was nuts
12:04<Bdragon>Yay for easily replacable laptop keyboards...
12:04*Bdragon just put a new keyboard in his CF-51
12:05<Bdragon>My hyphen key has been missing for a long time XD
12:05<@tasaro>Bdragon: mikegrb's son is a certified keyboard remover, if you need help in the future
12:05<Talman>I have... 6 keys missing on my dell.
12:05<Bdragon>(only got around to ordering a replacement when the down arrow key bracket broke)
12:05-!-ondrej [] has quit [Remote host closed the connection]
12:05<Talman>Replaced a CF-29 for a friend, need to buy a dell keyboard.
12:05<@caker>Hunter: destroyer of laptops
12:05<Bdragon>Ha, it's not exactly a skill...
12:05<Bdragon>No tools required...
12:05<HoopyCat>removing a keyboard is easy
12:06<path->did mikegrb give him an old laptop as a toy?
12:06<HoopyCat>putting it back in, however, takes skill
12:06<@tasaro>HoopyCat: i didn't say he was certified at replacing
12:06<Talman>hmm ,test.
12:06<Bdragon>Fold screen back all the way, pop the things behind the hinges, lift up the keyboard, pop the ribbon connector, pull keyboard out, stick new ribbon in, push the connector, fold keyboard in, replace the top thingie
12:07<Bdragon>Took me all of what, five minutes? (most of which was fiddling with the ribbon connector
12:07<emag>no screws?
12:07<Bdragon>No screws
12:07<Bdragon>Keyboard sits on a couple of pieces of grippy rubber
12:07<Bdragon>and is held in place by the top front panel thingie
12:07<Bdragon>which can be removed without tools
12:08<Bdragon>Very nicely engineered for easy replacement...
12:08<@mikegrb>path-: no, he practiced on my laptop
12:08<Bdragon>Hell, I pretty much took this thing apart a week or two ago for cleaning...
12:08*HoopyCat upgrades mikegrb's battery compartment with a peanut butter and jelly sandwich
12:09-!-SelfishMan [] has quit [Ping timeout: 480 seconds]
12:09-!-Talman [~Talman@] has quit [Quit: Leaving.]
12:09<Isvara>Is it peanut butter jelly time already?
12:09<Bdragon>You could tell that they cared about the field servicability...
12:09<@mikegrb>HoopyCat: you can skip the corner of credit card in slot loading dvd burner
12:09<@mikegrb>HoopyCat: I already have one installed from when hunter tried to use my visa to buy something on
12:09<HoopyCat>mikegrb: heehee
12:10<HoopyCat>but it works at the stripper factory!
12:10<@mikegrb>everytime I buy something the cashier asks if I got hungry when they see my card
12:11-!-SelfishMan [~TheSelfis@] has joined #linode
12:13-!-abhijit [] has joined #linode
12:14<JasonF>mikegrb: caker: Did you see that "OMG HOW CAN I SEND TO AN OPTIN LIST" message on the forums?
12:14<JasonF>I don't agree with the idea that all "opt-in" communication is bollucks
12:14<Bdragon>Wait, stripper factory? Is that where they come from?
12:14<JasonF>but I do think that... yeah.
12:15<SelfishMan>I get too much mail as it is (no spam). I don't want theirs too.
12:17<HoopyCat>i paid a lot of money for this opt-in list!
12:17<HoopyCat>there are 100,000 e-mail addresses that are just begging to hear about my clock enhancement pills
12:18<Bdragon>You know, there probabaly IS a relevant discussion list for male enhancement somewhere...
12:19-!-Navi`_ [~Navi`] has joined #linode
12:19-!-Navi` [~Navi`] has quit [Read error: Connection reset by peer]
12:19-!-abhijit0 [~abhijit@] has joined #linode
12:20<HoopyCat>bdragon: Subject: Re: [short-wang-l] Has anyone tried the new Canadian cialis?
12:20<Bdragon>[short-wang-bulk] New product!
12:20<SelfishMan>Is it better than the Mexican cialis?
12:20<Isvara>Wasn't he the kid in Temple of Doom?
12:21<HoopyCat>i almost want to create that list so i can torment hotmail's anti-spam desk some more
12:21<mwalling>SelfishMan: OFFLIST!
12:21<mwalling>HoopyCat: i'll join
12:21<HoopyCat>mwalling: that's what she said
12:22-!-abhijit [] has quit [Remote host closed the connection]
12:23<HoopyCat>bbl, campaigning for lunch
12:30-!-dvgrhl [] has joined #linode
12:31<tjfontaine>HoopyCat: to the sea!
12:31-!-abhijit0 [~abhijit@] has quit [Ping timeout: 480 seconds]
12:34-!-abhijit [~abhijit@] has joined #linode
12:34-!-pbryan [~pbryan@] has quit [Quit: Leaving.]
12:34-!-bliblok [] has joined #linode
12:36<abhijit>I am sending mail from linode ... but its going to SPAM folder ... any clue
12:38<SelfishMan>Can anyone cite the exact RFC that requires that a HELO hostname be valid? I can't remember where it is or even if there is one.
12:39-!-agentbleubleu [] has quit [Ping timeout: 480 seconds]
12:40<Isvara>Have you looked in 2821?
12:41<Isvara>According to that, it doesn't have to be a valid hostname.
12:42<SelfishMan>It says it has to be a FQDN or send a domain literal
12:42<SelfishMan>oops, address literal
12:42<Isvara>It says SHOULD.
12:42<Isvara>Not MUST.
12:42<emag>i generally just require incoming HELO to not claim to be my systems (unless they really are), and reject out of hand if they're that blatant about lying
12:43<emag>some spam fighters have taken the stance that if you're not giving an actual hostname (and the one you're supposed to be), you're up to no good
12:43<SelfishMan>I have one person that keeps mailing using a technically valid fqdn but it doesn't actually exist. They refuse to do anything about it.
12:43<Isvara>Whereas in reality it's just as likely to be misconfiguration (typically giving an internal hostname).
12:43<linbot>SelfishMan: Host not found.
12:43-!-jimcooncat [] has quit [Quit: Ex-Chat]
12:44<Bdragon>Ha, how about godaddy's crap?
12:44<Bdragon> As if.
12:44<SelfishMan>Godaddy wants people to use google for email.
12:44<Isvara>Bdragon: They can have that soon ;-)
12:45<emag>Bdragon: they're in grenada?!
12:45-!-TofuMatt [] has quit [Ping timeout: 480 seconds]
12:45<Bdragon>emag: No, I'm guessing they're just careless with their split horizon dns
12:46-!-Schroeder2 [] has joined #linode
12:46<emag>maybe they should move there...
12:50-!-internat1 [] has joined #linode
12:50-!-Internat [] has quit [Read error: Connection reset by peer]
12:51-!-Schroeder [] has quit [Ping timeout: 480 seconds]
12:53-!-abhijit [~abhijit@] has quit [Ping timeout: 480 seconds]
13:19-!-Akazawa [] has joined #linode
13:27<SelfishMan>dammit I hate being wrong.
13:28<SelfishMan>RFC-2821 clearly states that you should not reject on bad HELO data
13:28<path->what is bad helo data?
13:28<tjfontaine>invalid non dns'able hostname
13:29<SelfishMan>sending "HELO dkdkdkdkddkdk" is invalid and rejectable but "HELO" is not even though it doesn't resolve
13:29<linbot>SelfishMan: Host not found.
13:30<path->i think i require a fully qualified hostname, but it doesn't need to resolve
13:30*path- goes to check
13:31<path->reject_invalid_hostname, reject_non_fqdn_hostname
13:31<path->for postfix
13:31<SelfishMan>didn't we just have almost the same discussion yesterday?!?
13:31<path->i dunno, i think there are repeats in this channel
13:33<Isvara>...cho ...cho ...o ...o
13:34<path->rm -rf $problem_user
13:35<SelfishMan>$problem_user_bill * 5 > $other_users_bills_total
13:35<SelfishMan>They pay a lot of money by a factor of 5 at least
13:36<lanstin``>you mean '/ 5' then
13:36<SelfishMan>sorry, my mind is still stuck on reject_unknown_hostname
13:36<path->$lawyer_billing_mode = true
13:37<SelfishMan>$is_client_my_lawyer = true
13:37<SelfishMan>I pay them too much, they pay me waaaaaaaaaaaaaaaay too much
13:37<path->i'm sure the accountants appreciate it
13:37<SelfishMan>But they are a good client with just one sender that won't fix their dns
13:38<path->what's wrong with their dns?
13:38<SelfishMan>I still maintain that the other people should just create an a record for
13:38<SelfishMan>their helo is
13:38<path->just put it in the /etc/hosts file
13:39<path->it probably isn't the proper thing
13:39<path->but whatever
13:40<SelfishMan>I'm a big fan of configuring a mail system correct so I believe other people should. The sending server rdns ( is generated data from cogent and their helo has no a record.
13:40<linbot>SelfishMan: Host not found.
13:41<SelfishMan>Well, at least they do *own*
13:41<path->no www either
13:42<Isvara>(Stupid, uninformed site.)
13:43<Isvara>I don't care whether you think you should or shouldn't use www, but I definitely maintain that you can't try to lead the debate when you haven't even heard of SRV records.
13:43<SelfishMan>what's a srv record?
13:45<Isvara>This one is just bizarre:
13:46<tjfontaine>yay for the intertubes
13:47<@caker>[ 3] 0.0-10.0 sec 1.07 GBytes 918 Mbits/sec
13:47*caker drools
13:48<Isvara>Lies! No one can live at that speed!
13:52<Isvara>I just read all the comments on There clearly needs to be a huge SRV campaign. I didn't realise it was so unknown.
13:57-!-Kassah [] has quit [Ping timeout: 480 seconds]
13:59<@caker>tjfontaine: what ever happened with the mysqlcc recompile?
13:59<tjfontaine>caker: it works kinda :) I kinda lost motivation for it
13:59<@caker>hmm .. ok
13:59<@caker>what didn't work?
14:00<tjfontaine>the widget/control that allows for table editing and browsing needs lots of help
14:01<tjfontaine>caker: so viewing queries didn't work :)
14:03-!-bssteph_ [] has joined #linode
14:03-!-SelfishMan [~TheSelfis@] has quit [Ping timeout: 480 seconds]
14:04-!-lolcatz [~lolcatz@] has joined #linode
14:04-!-bliblok_ [] has joined #linode
14:04<lolcatz>hey folks! does linode allow to create a 64Mb node for gateway and 476Mb node for an app server from one Linode540?
14:04<tjfontaine>hmm why do you need a 'gateway'
14:05-!-xitology [~xi@] has joined #linode
14:06-!-jetlagmk3 [~jeff@] has joined #linode
14:06-!-bd__ [] has joined #linode
14:06<tjfontaine>bd__ ...o0
14:06-!-Netsplit <-> quits: bssteph, euph, bliblok, jetlagmk2, bd_, mattbnz, lanstin``, xorl
14:07-!-Netsplit over, joins: xorl
14:07-!-Netsplit over, joins: euph
14:07-!-bd__ is now known as bd_
14:07<tjfontaine>nothing :)
14:07<lolcatz>all interestingness is going to happen between nodes in a private network. and i need ssh gateway to work through
14:08<@tasaro>lolcatz: you'd need two linodes
14:08-!-Kassah [] has joined #linode
14:10-!-mattbnz [~mattb^] has joined #linode
14:34-!-Isvara [] has quit [Quit: Leaving.]
14:35<lolcatz>hrm, guys does anyone know why it tells me "Please provide a valid American Express, Visa, Mastercard, or Discover number", even if i entered my CC number (16 chars, integer)?
14:36<lolcatz>should it be kinda devided by something, i.e. xxxx-xxxx or xxxx xxxx?
14:37<lolcatz>hm nevermind, i reloaded signup page and entered info again, everything works
14:37-!-ondrej [] has joined #linode
14:38-!-Navi`_ [~Navi`] has quit [Quit: Navi`_]
14:42-!-meff [~meff@] has quit [Quit: If you ask you know I don't mind kneeling, but when my knees hurt, I'd like to stand.]
14:43-!-lolcatz [~lolcatz@] has quit [Quit: Miranda IM! Smaller, Faster, Easier.]
14:43-!-meff [~meff@] has joined #linode
14:44*caker haz 2.0 firmware installed
14:46-!-meff [~meff@] has quit []
14:48<nickj_>oh I need to wait for unlock for my phone
14:48<nickj_>no valid tt card ;/
14:49-!-meff [~meff@] has joined #linode
14:56*scott haz birthday cake
14:57-!-jcn [] has quit [Quit: ircII EPIC4-2.6 -- Are we there yet?]
15:03-!-arooni_____ [] has quit [Ping timeout: 480 seconds]
15:10-!-SelfishMan [] has joined #linode
15:11<SelfishMan>I miss anything good?
15:11<tjfontaine>I've taken over the world
15:11<Dave>you have?
15:11<Dave>damn :(
15:12*SelfishMan rejects tjfontaine's reality and substitutes his own
15:13-!-arooni-mobile [] has joined #linode
15:14<scott>Dave: you get rockband yet?
15:14-!-privet [] has joined #linode
15:15<SelfishMan>A friend sent me the entire setup but I don't have a console to play it on
15:15-!-arooni-mobile [] has quit []
15:15-!-arooni_____ [] has joined #linode
15:15-!-arooni-mobile [] has joined #linode
15:15*SelfishMan plays with mikegrb's triggers
15:15<@mikegrb>mmm cake
15:15<SelfishMan>mmm cake
15:16<Dave>scott: nah
15:16<Dave>not going to get it
15:19<SelfishMan>Ever get the feeling you're being watched?
15:21<SelfishMan>Just reading my irc log for this room. Wow, where was I for some of this stuff?
15:21<linbot>New news from forums: CentOS 5 Can't Install Portsentry in Linux Networking <>
15:23-!-lolcatz [~lolcatz@] has joined #linode
15:25<lolcatz>if i have 2 linodes Linode540, is it possible to make one with i.e 64Mb ram and the other one with 1000Mb?
15:26<lolcatz>so why do they have ram limits?
15:26<SelfishMan>I think you could get a 1080 for less than the two 540's though
15:26<tjfontaine>you can limit one to a specific amount of memory, but you can't assign that memory to another linode
15:26<tjfontaine>SelfishMan: he has some usecase
15:26<SelfishMan>I figured that but thought I would mention it
15:27<SelfishMan>I use the memory limits for my sandbox linode. Helps me figure out how well a certain config will run with X amount of RAM
15:27<SelfishMan>But I don't really use my linodes for production as much as I do for testing
15:28-!-linville [] has quit [Quit: Leaving]
15:29<SelfishMan>Now, my question is why doesn't linode have 64 or 128MB linodes?
15:29-!-Ptahhotep [] has joined #linode
15:29<Dave>SelfishMan: demand?
15:29<tjfontaine>SelfishMan: deprecated, that's where they started
15:29<Battousai>$20 will be the lowest price
15:29<Battousai>and caker keeps giving out freebies so $20 gives you quite a lot nowadays
15:30<SelfishMan>Every day I've been in this channel I've seen someone wanting a 64 or 128MB linode.
15:30<SelfishMan>I think most people would take a high contention 128MB linode for $10
15:30<Battousai>so would a lot of script kiddies
15:30<tjfontaine>wouldn't be as profitable I'd imagine
15:30<SelfishMan>So the script kiddies can't afford $20?
15:31<Battousai>not four or five of them at once to start a one-day bot war
15:31<SelfishMan>I can see that
15:32<SelfishMan>On the other hand, as an established and (hopefully) in good standing customer, it would be nice to have the option.
15:33<SelfishMan>An extra $10 isn't that big of a deal but if I just want a minimal node in each DC that is $30 more for me per month
15:33<SelfishMan>But I'm probably just ranting.
15:33*SelfishMan climbs down from soap box and throws it at tjfontaine
15:33<Bdragon>lolcatz: Here's a better explanation: Different plans can't be on the same machine*. (* there are exceptions related to the beta xen boxes)
15:34<Bdragon>Heh, NUMAnode
15:34<SelfishMan>NUMAnode's are a myth
15:35<lolcatz>i can afford extra 10 bucks but, i'm not so sure i need 360 MiB of RAM for a ssh gateway that does only as a proxy between me and private network
15:35<Bdragon>Nematodes aren't though :P
15:35<lolcatz>what's NUMAnode?
15:35<lolcatz>and how to sign bor beta nodes?
15:35<Bdragon>It's a joke.
15:35<SelfishMan>Bdragon: That's freaky. I was just typing exactly that
15:36<SelfishMan>lolcatz: You still couldn't do what you want on the beta xen node. Besides, xen is out of beta
15:36<SelfishMan>At least it better be since they are putting everyone on it
15:36<SelfishMan>Of course, GMail is still beta
15:38-!-lolcatz [~lolcatz@] has quit [Quit: Miranda IM! Smaller, Faster, Easier.]
15:38<Battousai>you offended him
15:39<SelfishMan>I tend to do that
15:51-!-arooni-mobile [] has quit [Ping timeout: 480 seconds]
15:54<@tasaro>i'm sure some people in here could tell you stories about swap thrashing Linode 64s.. If you want a $10 system there are a million shared providers out there
15:55-!-jcn [] has joined #linode
15:55<bd_>you could also share a 360 with someone :)
15:56<bd_>o, he's gone anyway
15:56<SelfishMan>tasaro: But I want a *good* provider! I understand the reason for it.
15:56-!-lolcatz [~lolcatz@] has joined #linode
15:56<SelfishMan>I need to figure a bunch of stuff out so I can get more linodes setup in each dc.
15:57<SelfishMan>I thought linode was going to be the same as the rest so when I signed up a couple months ago but now I know better. Linode rocks and I want my nodes with them!
15:57<lolcatz>guys, /etc/hostname is the only place where i have to change hostname, isn't it?
15:57*scorche|sh 's grammar filter explodes
15:58<mendel>lolcatz: probably /etc/hosts too
15:58<mendel>i'd just 'grep -R oldhostname /etc'
15:58<SelfishMan>wow, random thoughts merged in that last message. Don't quote me on it!
15:58<mendel>or 'grep -R oldhostname /etc /usr/local/etc'
15:59<lolcatz>oh, right, have to sleep a bit
16:00<lolcatz>trying linode after slicehost :-/
16:03-!-pbryan [~pbryan@] has joined #linode
16:04<jcn>i had slicehost before linode.
16:06-!-kupesoft [] has quit [Quit: Ex-Chat]
16:06<lolcatz>jcn: so, why have you moved?
16:08<jcn>lolcatz: a number of reasons. one came down to flexibility - slicehost didn't let me add more hard drive space and linode did.
16:09<jcn>the second was price - for the space that I did need, i got more with linode for less money (i got my linode during the extra bandwidth when you prepay promo)
16:09<lolcatz>the main reason for me is a private network with free traffic, heh
16:09<jcn>i only have one box, so it's not really a huge deal for me.
16:09-!-bliblok_ is now known as bliblok
16:09<jcn>i also liked that i could slice up my allocated space into multiple partitions.
16:10<jcn>and there was much less latency to newark.
16:12-!-xitology [~xi@] has quit [Quit: Ex-Chat]
16:14-!-Schroeder2 [] has quit [Ping timeout: 480 seconds]
16:15<jcn>however, i am looking for some dedicated servers. anyone have any suggestions?
16:16-!-Schroeder2 [] has joined #linode
16:16<scorche|sh>what do you need a dedicated server for?
16:19<jcn>oh, work.
16:21<scorche|sh>as in, why do you need a dedicated server rather than just a big linode?
16:23<jcn>well, first of all, i'd like someone i can call up 24/7 in case something goes wrong.
16:24<jcn>and i need assloads of hard drive space.
16:25<HoopyCat>i'm sorry, jcn, we work in SI units here. you need kiloarseloads of space.
16:26<jcn>oh, my mistake.
16:27<bd_>well, the new NAS thing is coming soon
16:27<HoopyCat>jcn: i haven't shopped for a dedicated server in awhile, but most datacenters where linode has stuff offer servers, i believe.
16:28<lolcatz>HoopyCat, i guess, you're from linode team, aren't you?
16:28<jcn>the support thing is also a big deal for me. i don't mind it for my own box, but i don't want to be hanging out on IRC waiting for someone to show up if there's a problem.
16:28<HoopyCat>lolcatz: No.
16:30-!-lolcatz [~lolcatz@] has quit [Quit: Miranda IM! Smaller, Faster, Easier.]
16:30<HoopyCat>jcn: opening a ticket will page out; if it's urgent, it'll wake someone up
16:37<jcn>hoopycat: i suppose, but still, when our business depends on it, i'd like a phone number i can call.
16:39<bliblok>jcn: That's not a good solution. You only disturb the people who should be fixing your problem, which make the problem take longer to fix.
16:40<HoopyCat>jcn: i've redefined my business such that 3am is a non-critical time (o, retirement!), but yeah, i hear ya
16:41<jcn>bliblok: thanks for your input.
16:42<bliblok>Either that or you get to talk to someone who don't know anything about the system you're having problems with, wich is equally unhelpful.
16:43<jcn>again, thanks, but that is clearly untrue.
16:43-!-hfb [] has joined #linode
16:43-!-chovy [] has left #linode []
16:43<jcn>i have servers at both rackspace and servint right now, and i've appreciated the ability to pick up the phone and talk to either someone at the noc or someone who can prod someone at the noc.
16:46-!-Schroeder2 [] has quit [Ping timeout: 480 seconds]
16:46<bliblok>It depends on the nature of the problem, obviously. What i state is only true if it is a known problem for all of or many of the customers, that is just taking time to fix.
16:47<HoopyCat>bliblok: if they've got one guy sitting there at 3am, they've probably got two guys sitting there at 3am
16:48<jcn>yeah, i generally don't sit on the phone calling the noc if the network is down.
16:48<HoopyCat>back at $UsedToWorkThere, our after-hours dispatches were done by answering service; once i was on it, i'd tell them "yeah, if anyone else calls with that, it's a gronkfitz with the derflappenator, take their number and i'll call 'em back when it's fixed" and go back to fixing it
16:48<jcn>but if my server is doing weird things, i like having someone sitting there who can walk over to the box if needed.
16:50<HoopyCat>jcn: you might wanna consider a local shop; there're still some around. that's what $UsedToWorkThere mostly dones. 24x7 keycard access, free parking, right off the expressway.
16:52<jcn>what's an expressway?
16:53<HoopyCat>jcn: the place where the deer and the half-asleep tractor-trailer drivers like to play at 3am
16:53<HoopyCat>i only almost got killed once on the way to/from the datacenter!
16:55*emag builds a time machine to make sure the job's done right
16:57<HoopyCat>emag: i'm actually not kidding about being almost killed :-)
16:57-!-row [] has quit [Ping timeout: 480 seconds]
16:57<avongauss>jcn, not necessarily trying to sway you and I don't know your needs, but usually you can purchase multiple virtual hosts for the
16:58<avongauss>same price as a dedicated host. From a redundancy standpoint that would seem to be a good thing.
17:01<HoopyCat>avongauss: some problems are more difficult than others to reliably and efficiently spread across multiple geographically-diverse hosts, esp. ones that involve buttloads of data
17:01<HoopyCat>sorry, kiloarseloads
17:01<emag> virtue of the fact that dns went away for a bit, i'm gonna say someone's finally bothering to update bind
17:01<emag>here @ work, that is
17:02-!-neh [] has quit [Ping timeout: 480 seconds]
17:02<emag>HoopyCat: how many kiloarseloads are there in a metric fuckton?
17:02<avongauss>agreed HoopyCat, but even in the same data center if uptime is a concern I would rather have two than one.
17:03<avongauss>unless it's a hardware or network transit issue, whether it's a dedicated or virtual box you're still the one that has to fix it.
17:04-!-Schroeder2 [] has joined #linode
17:07<HoopyCat>avongauss: with a managed service, the datacenter will (officially) fix a lot more problems, for a price of course. :-) for the right price, you may not have to lift a finger. whether that is worth it or not is a business decision.
17:08<avongauss>managed is definitely different... ;) I was just reading the scroll back and didn't realize the discussion was about managed dedicated servers. My bad.
17:08<bliblok>emag: Try google for unit conversions.
17:09<HoopyCat>avongauss: it could be. there's a whole continuum of different services in the market, which is awesome.
17:10<avongauss>Agreed. I don't think its the cure for all problems, but I am a definite fan of virtualization - even for a single box with a dedicated purpose.
17:10<SelfishMan>Um...teh interwebs broked. i cant print
17:10<avongauss>Your router is blocking your IPP port... ;)
17:11<SelfishMan>I love those support calls. They called me up and told me that "the internet is broken" and after several minutes of trying to get more info from them I finally found out that they were trying to print a receipt from a web page.
17:11<HoopyCat>avongauss: me too... however, i don't see google running on a set of linodes. ('tho that might be what keeps filling up atlanta)
17:11<emag>try PoIP
17:12<SelfishMan>Told them to power cycle their printer and now magically teh interwebs work
17:12<emag>SelfishMan: so *that* was why my interweb wasn't working!
17:12<avongauss>Google doesn't run virtualization for their primary apps, but they do seem to use the same concept. More workers verses less heavy lifters.
17:13<avongauss>I'm not sure that was worded well, my point being they seem to have a preference for cheaper commodity hardware rather than larger specialized servers.
17:13<HoopyCat>i'm about to have to "Sir, did you read the part in bold about how everything you enter into that big box above will be released under a Creative Commons license?" someone, which i'm trying to avoid because he has no idea who i am yet. hmm...
17:14<HoopyCat>avongauss: yeah, google is perhaps a bad example, considering their architecture. how about... uhh... wikipedia!
17:14<HoopyCat>(the good ol' buncha-servers-in-one-datacenter theory)
17:15<avongauss>Wikipedia would be a good example, and a good showcase for virtualization in my opinion... ;)
17:16<avongauss>Probably 95% percent of their queries are to static content? The one thing that wouldn't work at the moment is their disk space requirements... :(
17:16<HoopyCat>wikipedia is based in one datacenter located right on the gulf coast of florida. what could POSSIBLY go wrong.
17:17<HoopyCat>avongauss: and memory... they memcached the living bejeezus out of things
17:17<avongauss>Caching is good, caching as a substitute for good architecture is not </rant>
17:18<pleia2>tasaro: thank you :)
17:18<SelfishMan>HoopyCat: Wikipedia was worldwide mirrors in all languages
17:18<avongauss>sounds like another satisfied ticket request - either that or they were writing in invisible characters...
17:18<HoopyCat>caching IS good architecture. :-)
17:19<avongauss>caching is not architecture.
17:19<SelfishMan>caching should increase page render times on stable architecture
17:19<avongauss>it might be a part of an architecture, but alone it is not.
17:20<HoopyCat>SelfishMan: unless something has changed drastically in the couple of months since i've looked at the page, there's a couple read-only servers in a couple different places but the database masters and everything-by-default is in tampa bay, incl. foreign language wikis.
17:20<SelfishMan>For example, twitter has bad architecture so their caching doesn't fix the problems with a slave mysql server going down. However, if their architecture was stable then good caching could make a page render in 0.100 seconds instead of 3.
17:21<avongauss>Twitter is a wonderful example. :)
17:21<HoopyCat>caching is an important part of a quality architecture, although when missing, nothing should catch fire. how's that? :-)
17:21<HoopyCat>twitter == catch fire
17:21<SelfishMan>HoopyCat: exactly
17:21<avongauss>We're getting closer... ;)
17:22<SelfishMan>if you turn caching off then things should still work and run in an acceptable amount of time. Caching on should make it all really really snappy
17:22<HoopyCat>i can kill memcached right now and nothing would break, because i architected this particular service to not have drastic growth ;-)
17:22<SelfishMan>what scares me is that twitter is working on more caching and hardware instead of optimizing sql queries and structures
17:22<SelfishMan>facebook is a good example of that
17:23<SelfishMan>I think.
17:23-!-pygmalion [] has joined #linode
17:23<avongauss>I agree SelfishMan.
17:24<SelfishMan>Hmm....I thought I heard from a major wikipedia person that it had several mirrors. Maybe it was something else or they are just trying to bring up mirrors
17:24<avongauss>I don't know how Twitter is configured, but it does seem they have a lot of problems. And this is before they've really "caught on"...
17:25<SelfishMan>twitter was bad architecture with a lot of laziness
17:25<HoopyCat>SelfishMan: twitter isn't a normal "web 2.0" problem... it's more of a real-time messaging system, in my eyes. route it to its destinations, *then* stick it in the database for web display...
17:25<HoopyCat>that said, i'm pulling all of this out of my ass.
17:26<SelfishMan>You're right, it is a real-time messaging system. Problem is they didn't engineer it that way.
17:26<avongauss>Real-time messaging system with real-time multi-point monitoring.
17:27<SelfishMan>They rely on too many complex sql queries after data is inserted into the db
17:27<SelfishMan>the various triggers and notices are pulled in an additional query to the database after the initial write instead of processing it all in the same loop
17:28<HoopyCat>i've got a couple ideas on how to make things more efficient, but unfortunately, so does everyone else :-)
17:28<HoopyCat>my solution will work, however, unlike everyone else's
17:29<SelfishMan>very true. I say gut the system and build it up from there knowing what it is.
17:29<HoopyCat>SelfishMan: agreed
17:29<HoopyCat>twitter, when it is 100%, is AWESOME
17:29<avongauss>Maybe that's exactly what they are doing? ... Nah...
17:29<SelfishMan>What scares me is most of the failures in the last few months have been because of "testing new code" on producton boxes
17:30<SelfishMan>Twitter can be great and people want to use it. That's why they keep coming back to it when it comes back up.
17:30<SelfishMan>That, and Twitter is where everyone is.
17:30<SelfishMan> <-- shameless plug
17:32<avongauss>more porn for everybody:
17:32<SelfishMan>Yep. Have four in my box now and have a few 2TB's coming next week
17:32<scorche|sh>avongauss: that url is awfully misleading
17:32<scorche|sh>SelfishMan: goddamnit
17:32<avongauss>the URL or tag line?
17:32*SelfishMan contracts with seagate
17:33<HoopyCat>i ran out of data
17:33<HoopyCat>i have no reason whatsoever for more disk space
17:33<SelfishMan>I only took this contract with them because I need the disk space
17:34<SelfishMan>I burn through teraarseloads every week
17:34<avongauss>Agreed, hence my tag line. Video libraries and Porn libraries are the primary reason an individual (remember, desktop drive) wants that much space.
17:35<HoopyCat>i don't care about movies or most television programmes and i like my porn fresh
17:35<avongauss>Although, the difference between consumer and server drives these days is debatable these days - especially with the different RAID options.
17:35<avongauss>minus one of those these days...
17:36<SelfishMan>massive amount of raw data here
17:36*scorche|sh slaps SelfishMan
17:36<SelfishMan>i have food under my apostrophe key
17:36<SelfishMan>and sometime''''''''''''''''''s i get lazy
17:37*scorche|sh smells BS
17:37-!-Schroeder2 [] has quit [Ping timeout: 480 seconds]
17:38<SelfishMan>teh keebord keyes are broked
17:40*SelfishMan glare's at scorche|sh
17:41<SelfishMan>I must admit my grammar used to be better before I started texting and chatting
17:42<SelfishMan>and stuff's
17:42*SelfishMan mocks scorche|sh
17:47-!-Ptahhotep [] has quit [Remote host closed the connection]
17:53-!-lakin [] has quit [Quit: Ex-Chat]
17:53-!-lakin [] has joined #linode
17:54-!-Schroeder2 [] has joined #linode
17:56<TJF>now we can kick it!
17:56*HoopyCat kicks it up old-school
17:58<HoopyCat>/mode #linode +b *!*@k?t?i?0?-?a?-?2?s?3?.?i?e?g?c?m?n?t
18:01*tasaro is now known as DevilDoll.
18:06-!-Bass10 [] has quit [Ping timeout: 480 seconds]
18:13-!-r3z` [] has joined #linode
18:13-!-r3z [] has quit [Read error: Connection reset by peer]
18:15<TJF>tasaro: ^5
18:16-!-TJF [] has left #linode []
18:17<HoopyCat>"I will work night and day until you are shut down!"
18:18<SelfishMan>What's that?
18:20<HoopyCat>a new user on a wiki who has, apparently, come up against some philosophical differences
18:20-!-ondrej [] has quit [Remote host closed the connection]
18:20<SelfishMan>That's awesome
18:24-!-lolo [] has joined #linode
18:26<HoopyCat>SelfishMan: yeah, it's probably going to easily be #3 if not #2 on the Battles of All Time list
18:27<HoopyCat>#1 made the newspaper, which sets a bit of a high bar
18:27<lolo>Hey Guys! I got this iptables script
18:28<lolo>I'm wondering where should I put it to make it run at startup?
18:30<HoopyCat>lolo: what distribution are you using?
18:39-!-bliblok [] has quit [Quit: Leaving]
18:42-!-jcn [] has quit [Ping timeout: 480 seconds]
18:45-!-pbryan [~pbryan@] has quit [Ping timeout: 480 seconds]
18:48-!-mendel [] has quit [Quit: mendel]
18:49<SelfishMan>OK, I feel like an idiot
18:50<SelfishMan>I've been fighting with a linode only to find out that a script was failing because I powered down the remote server last night. That was two hours well spent!
18:52-!-pbryan [~pbryan@] has joined #linode
18:53<JDLSpeedy>SelfishMan: o man, I hate when that happens
18:53<Peng>Heh, I just got an email from saying that my server has a 213-day uptime, but is going to be restarted for maintenance and a fsck in a few days. :)
18:54<JDLSpeedy>Peng: ouch
18:54<Dave>SelfishMan: I just spent 2 hours cursing an SQL join not working, only to reilise I was join'ing on the wrong fields :(
18:54<SelfishMan>What a productive group this is!
18:55<JDLSpeedy>Peng: why would you get a email from your hosting something?
18:55<Peng>JDLSpeedy: They're a file host thingy.
18:55<JDLSpeedy>Peng: ahh
18:55<SelfishMan>load average: 44.81, 27.69, 13.81
18:56<SelfishMan>and increasing
18:56<JDLSpeedy>SelfishMan: o man, lol, and your linode still responding?
18:56<Peng>SelfishMan: On your Linode?
18:56<SelfishMan>yep. Not even sure how. Rebooted it a few seconds ago.
18:56<SelfishMan>load average: 77.41, 40.59, 19.26
18:57<SelfishMan>up 5:25, 1 user, load average: 77.41, 40.59, 19.26
18:57<Dave>what the hell are you running on there?
18:57<SelfishMan>I have all services stopped right now
18:57<JDLSpeedy>what does your cpu% say on top?
18:58<SelfishMan>Cpu(s): 4.2%us, 5.2%sy, 0.0%ni, 8.2%id, 82.3%wa, 0.0%hi, 0.0%si, 0.1%st
18:58<JDLSpeedy>im surpised its still responding
18:58<SelfishMan>slow, but yes it is
18:58<SelfishMan>The OOM process nuked the kernel
18:59<Peng> was because you were swapping to death?
18:59<SelfishMan>Swapping what? I'm not even close to out of memory
18:59*Peng is confused now.
18:59<JDLSpeedy>running latest kernel?
18:59<SelfishMan>Mem: 553164k total, 5136k used, 548028k free, 480k buffers
19:00<SelfishMan>2.6 latest
19:00<SelfishMan>I'm not showing a single process but somehow I have 422 sleeping
19:00<SelfishMan>and the damn thing won't reboot. what is the sysrq to restart now?
19:00<Peng>SelfishMan: You can do it from lish.
19:01<Peng>I do K S U B (kill processes, sync disks, unmount disks, reboot).
19:01<SelfishMan>I did it from lish. stuck waiting. Just got the reboot notice.
19:01<SelfishMan>KSUB, thank you
19:01<JDLSpeedy>its going to take afew minutes
19:01<SelfishMan>I know, just trying to get it down now so I don't hurt other nodes
19:01<Peng>They'll probably be fine.
19:03<SelfishMan>OK, much better after reboot
19:03-!-FloodServ [] has quit [Service unloaded]
19:05<SelfishMan>Sorry to anyone else on Dallas71. Unless you are a jerk. In which case I meant to do that.
19:05<JDLSpeedy>LOL, your mean
19:06<SelfishMan>I prefer the term "arrogant american prick" thank you very much
19:06<SelfishMan>It only last 5 minutes. 7 tops. oops.
19:06-!-hfb [] has quit [Ping timeout: 480 seconds]
19:07<SelfishMan>Holy hell. just attempted 50,000 invalid recipients against my mail server causing it to freak out
19:07<purrdeta>nice lol
19:08<bd_>remarks: Any abuse reports, please send them to
19:08<SelfishMan>the one time I don't have rate limiting in place
19:08<SelfishMan>bd_: They don't respond. Never have, never will
19:08<booja>damn italians
19:08-!-Isvara [] has joined #linode
19:09*SelfishMan blinks with a blank expression
19:09-!-Isvara [] has quit []
19:09-!-TofuMatt [] has joined #linode
19:09-!-Kassah [] has quit [Ping timeout: 480 seconds]
19:10<SelfishMan>My CPU usage graph shows I Maxed out at 239.73%. I'm kinda proud of that!
19:11-!-Isvara [] has joined #linode
19:11<SelfishMan>TofuMatt: You missed my node running an insane load average from a mail attack
19:11-!-hfb [] has joined #linode
19:13<TofuMatt>ah, :(
19:13-!-FloodServ [] has joined #linode
19:16<SelfishMan>I will say that I'm impressed that the linode handled that as well as it did. This same system and config was on tektonic two months ago and could barely run under normal load.
19:20-!-Kassah [] has joined #linode
19:20*SelfishMan is a linode fanboy
19:21<bd_>only 239.73%?
19:21<bd_>surely you can do better than that :P
19:21<SelfishMan>I don't think the graph had enough data to show higher. 5-minute average, right?
19:22<SelfishMan>Again, sorry to anyone else on Dallas71 that was affected by that. Unless you are a jerk. In which case I meant to do that so screw you.
19:22<@mikegrb>mmm cake
19:23<SelfishMan>mikebot scares me
19:24<avongauss>not sure it would help, but you could always complain to's upstream providers.
19:25<SelfishMan>avongauss: It was a standard bot rapid mailing. Not worth the time for me to go after them. Rules are in place now so that won't happen again.
19:25<avongauss>50,000 is an excessive bot...
19:26<SelfishMan>I've had great luck submitting attacks to The Planet but most other providers don't care that much.
19:26-!-hfb [] has quit [Quit: Leaving]
19:26<SelfishMan>Yes it is. Probably a screwup in the config considering that the mail they were attempting to deliver had template variables in the cialis-type spam mail.
19:29-!-webPragmatist [~cleblanc@] has joined #linode
19:29-!-webPragmatis1 [~cleblanc@] has quit [Read error: Connection reset by peer]
19:33<SelfishMan>I'm out, have fun ppl
19:33-!-SelfishMan [] has quit [Quit: SelfishMan]
19:42-!-Kassah [] has quit [Ping timeout: 480 seconds]
19:42-!-scorche [] has quit [Read error: Connection reset by peer]
19:43-!-scorche [] has joined #linode
19:45-!-pdepartida [] has joined #linode
19:46-!-lakin [] has quit [Quit: Ex-Chat]
19:46-!-binel_ [] has joined #linode
19:47<pdepartida>hello everyone quick question if i can get help...
19:48<pdepartida>When using running a script in a ssh terminal screen with an &, such as "./ &"
19:48<pdepartida>will the script keep running once i close the terminal session?
19:48<straterra>should..but look into 'screen'
19:48-!-binel [] has quit [Ping timeout: 480 seconds]
19:49<pdepartida>straterra: thanks. will do.
19:49<Isvara>Look into 'nohup'
19:49<straterra>man screen
19:49<Peng>You could test it by running 'sleep 300 &' or something.
19:50<Isvara>You'd think so, but that doesn't actually appear to return.
19:50<Isvara>mac:~ dan$ time ssh projects 'sleep 5 &'
19:50<Isvara>real 0m6.617s
19:52<Peng>Maybe it waits for all jobs to complete.
19:52-!-Kassah [] has joined #linode
19:53<Isvara>If I ^C out of it, then ssh in and do a 'ps x', it's running.
19:53<Peng>If you ssh in (and get a shell), then type the command in, it'll keep running.
19:53<pdepartida>yep, just checked it, keeps running. "nohup" seems interesting also.
19:54<Isvara>ssh host 'sh sleep 1000 &' seems to work as expected.
19:55<Isvara>Oh, no it doesn't. That was a previous one.
19:57<pdepartida>it seem that 'nohup ./ &' does the trick.
19:57<Isvara>Oh, wait. You wanted to run something in an interactive session?
19:58*Isvara thought you wanted to execute it remotely
19:59<pdepartida>Isvara: i just wanted to ssh the server run a script and be able to close the session and be sure that the script finishes it's job.
19:59<Isvara>Yeah, I initially thought you wanted to do all that in one command.
19:59-!-daemonic [~Ryan0213@] has joined #linode
20:00<pdepartida>Isvara: so will 'nohup ./ &' do the trick?
20:01<Isvara>If it does happen to produce any output to the terminal, it will end up in nohup.out.
20:01<pdepartida>Isvara: thxs.
20:03<lolo>HoopyCat: sorry, I had to get away from my desk...i'm using ubuntu 8.04
20:04-!-pdepartida [] has quit [Quit: pdepartida]
20:06-!-Kassah [] has quit [Ping timeout: 480 seconds]
20:09-!-pbryan [~pbryan@] has quit [Quit: Leaving.]
20:09*TofuMatt has an SSL question
20:10<TofuMatt>If I have a mailserver,, and a vhost,, and want them both to have SSL, but don't wanna buy a wildcard, can they be on the same IP -- i.e. is it just https that needs different IPs?
20:11<Peng>As long as you use different ports for https and mail, you can use the same IP.
20:11<Isvara>They can be on the same IP address, but you'll need two certificates.
20:11<Peng>That too.
20:11<TofuMatt>yeah, I knew that much
20:11<TofuMatt>OK, excellent
20:12<Isvara>The only reason you need different IP addresses for HTTPS is that the server otherwise has no idea which server certificate to present, because the SSL handshake happens before the HTTP request (with Host field) is sent.
20:12-!-andaas [] has joined #linode
20:12*mwalling thinks someone should write http/1.2, which uses STARTTLS
20:13<Isvara>mwalling: HTTP/1.1 already has that mechanism, but no one uses it.
20:13<Isvara>RFC 2817
20:15<Isvara>Maybe I should try sending '426 Upgrade Required' to some web browsers to see what they do.
20:15-!-ryan8403_laptop [] has quit [Ping timeout: 480 seconds]
20:16-!-Kassah [] has joined #linode
20:17<Peng>Argh, I still use DreamHost for mail (yeah yeah), and their network is going to go down for maintenance tomorrow.
20:17<bd_>Peng: good time to migrate? :D
20:17<Peng>It was announced a few weeks ago, and I had been thinking about migrating, but there isn't really time now.
20:17<Isvara>What?? They just take their whole network down for maintenance?
20:17<Peng>Isvara: Well not often.
20:18<bd_>" On July 11, 2008 starting at 10pm we will have a brief network outage to restructure our routing tables. We are planning two hours to deal with any quirks that may come up, however our Cisco network engineer estimates the total downtime to be under 30 minutes.
20:18<bd_>During this downtime DreamHost will be completely inaccessible, however no data will be lost."
20:18<Peng>"This restructuring is to put the final fixes on various packet loss issues we have applied “band-aid” fixes to in the past, including installation of an extra core router at one of our facilities."
20:20-!-SelfishMan [] has joined #linode
20:20<Isvara>Says a lot about their network.
20:20<Isvara>To be fair, perhaps they mean "it should take less than five minutes, but we're saying 30 to be on the safe side."
20:20<Peng>They use NFS for everything, so even if they didn't take down the whole network, things would probably explode anyway.
20:22-!-andaas [] has quit []
20:25-!-lolo [] has quit [Remote host closed the connection]
20:25-!-andaas [~greg@] has joined #linode
20:25-!-mendel [] has joined #linode
20:34<SelfishMan>NFS sucks
20:36<SelfishMan>still what?
20:36<Isvara>In what way? We're trying to decide whether or not to use it.
20:36<SelfishMan>Depends on the situation. I run 20 PXE booting fat clients over NFS with no problem but it sucks for other stuff I do.
20:37<Peng>Linode might use NFS for the NAS thing.
20:38<SelfishMan>So what is the deal with the NAS thing? NFS makes sense for that. FTP would be nice but I don't know what the objective of it is.
20:38<daemonic>AFS is nice
20:38-!-Guest15 [] has quit []
20:38<Battousai>i heard they'll use SMB
20:38*Battousai ducks
20:38<@mikegrb>mmm cake
20:38<SelfishMan>cake is better than smb
20:38<SelfishMan>see, mikebot agrees
20:38<Isvara>SelfishMan: What's the "other stuff"? We're planning to use it for the file server for about four large Xen servers.
20:39<Isvara>Although iSCSI is still an option.
20:39<SelfishMan>Are you planning on running the xen nodes off of NFS or just mounting it and accessing data?
20:39<Isvara>The former.
20:39<SelfishMan>iSCSI is great when it works but many kernels are still very experimental
20:40<Isvara>It somehow seems a bit perverse having fake block devices as files on a NFS filesystem.
20:40<SelfishMan>I wouldn't run the xen nodes off of NFS unless they won't be accessing it much outside of boot.
20:41<SelfishMan>The 20 PXE nodes rarely read and never write to the NFS volume.
20:41<Isvara>Ah. These will be reading and writing a lot.
20:41<SelfishMan>I've tried using IPC::DirQueue on a NFS volume and that was terrible when there were 1000 items in the queue.
20:41<Isvara>iSCSI seems like a more natural fit, but no one on the team has used it in anger.
20:42<Isvara>Or we might just give up and use local LVM volumes and not migrate domains between servers very much.
20:43<SelfishMan>I would look at iSCSI but make sure the kernel support is solid. Honestly, in my experience, iSCSI targets aren't that stable yet. The initiator seems solid.
20:43<SelfishMan>I might have that backwards. Target = Server right?
20:43<SelfishMan>How large are the domains expected to be?
20:44<Isvara>By what measure?
20:44<SelfishMan>Raw bytes.
20:44<Isvara>Probably something like 10GiB each. They'll be things like ADS, XenApp server...
20:45<Isvara>A mix of Linux and Windows
20:46<Isvara>(Actually, knowing how Windows on XenServer works, it scares the hell out of me.)
20:47*path- 's had good experiences with iSCSI
20:47<Isvara>On Linux?
20:47<path->using hardware hbas are best, but $$
20:47<path->we use the software initiator on our current webserver
20:47<path->our new server will be a virtual machine on a host with an hba
20:48<path->host bus adapter
20:48<path->like a scsi card
20:48<path->qlogic is what we're using
20:49<path->they are like $500 or more for a dual port card
20:49<path->it turns tcp/ip iscsi and presents the host a scsi device
20:50<path->i think the software initiators are pretty stable
20:50<path->but i don't have much experience running the target (server) part in software
20:51<path->my coworker runs one at home
20:59-!-spasmface [] has joined #linode
21:01<Isvara>Man, how do you tell a company their recruiting sucks?
21:02<Isvara>Without sounding like an arse.
21:02<SelfishMan>10GB shouldn't be bad to shuffle between hosts
21:03<Isvara>No, not really, and it wouldn't even be often.
21:03<SelfishMan>Go with local drives then unless you have a month to fully burn in the hosts under heavy load
21:04<Isvara>I think some people are a bit blinded by the extra budget we were given if we promised to migrate from VMWare to XenServer.
21:04-!-wellhello [] has joined #linode
21:04<SelfishMan>I think the iSCSI target software is still considered experimental under several flavors of linux
21:05<SelfishMan>Dropping the licensing fees?
21:05-!-Kassah_ [] has joined #linode
21:05-!-Kassah [] has quit [Read error: Connection reset by peer]
21:06<avongauss>Imaging the extra budget if you just adopt Samba and Wine... ;)
21:08<Isvara>SelfishMan: We own Xen, so the edict came from upon high that we should be using it instead of a competitor's product
21:08-!-jcn [] has joined #linode
21:08<SelfishMan>The 20 fat client setup I put in at the library saved several hundred per year in AV licensing, cut support down from 30 hours/week to less then 2 and made the patrons happier.
21:08<avongauss>They might have a point...
21:08<Isvara>I think they're just trying to make that half a billion dollars they paid seem worthwhile
21:08<SelfishMan>Isvara: That's awesome.
21:09<SelfishMan>Xen is better than VMWare ESX
21:09<@mikegrb>mmm cake
21:09<SelfishMan>Cake is still better than Xen
21:09<path->you see the news about their ceo?
21:09<SelfishMan>mikebot agrees
21:09<Isvara>I've only used VMWare Server.
21:09<avongauss>I mean this as a totally serious question, but does Xen run Windows well at this point?
21:09<path->kinda scary
21:10<Isvara>path-: What news was that?
21:10<booja>I quite like vmware :(
21:10<Isvara>avongauss: Yeah, but how it does it is terrifying.
21:10<SelfishMan>avongauss: I have a few Xen Windows nodes and they run great
21:10<path->we're really invested in vmware now
21:11<path->i wouldn't mind going the xen route, but i don't think i'd get it past some of my coworkers
21:11<avongauss>Terrifying? Okay, you're definitely not associated with the sales department. ;)
21:11<avongauss>Xen Windows nodes?
21:11<path->one of them firmly believes that you get what you pay for
21:12<avongauss>I haven't used their ESX product in a while, so I'm mainly basing this off the Workstation and Free Server product, but I've become less of
21:12<Isvara>avongauss: Because Windows expects to start up in 16-bit real mode, Xen starts it on an emulated CPU in QEMU in userland. Then, when it switches to 32-bit protected mode, it moves the running domain from QEMU to the real, virtualised CPU, does a bit of binary kernel patching, and replaces the odd driver here and there.
21:12<SelfishMan>Each has a place. I use VMWare, Xen and a handful of others
21:12<path->so he's more involved in the windows end of the shop :P
21:12<avongauss>a fan of VMware in the last year or so - just a lot more bloat and less focus on core functionality in my opinion.
21:12<path->the virtual infrastructure stuff works well
21:13<path->things fail automagically
21:13<path->vmotion is cool
21:13<path->but it's all pretty pricey
21:13<avongauss>Terrifying was good word then... "Binary Kernel Patching"...
21:13<path->things fail over to other hosts automagically :)
21:13<booja>have any of you guys played with nomachine ?
21:13<Isvara>The Xen guys are ridiculously smart.
21:14<Isvara>Although so far they haven't been smart enough to move into an office with air conditioning.
21:14<SelfishMan>booja: Yes. I don't think it is worth it.
21:14<path->i wonder if there will be lots of people leaving vmware
21:14<path->maybe citrix will hire come of them
21:15<path->or will they start something else
21:15<SelfishMan>Being hot makes the blood flow faster which pumps more oxygen to the brain. Hence the intelligence.
21:15*avongauss confused goes to read the link path- posted.
21:16<Isvara>The link confused me. She was trying to get VMWare spun off from what?
21:16-!-robg [] has joined #linode
21:16<path->vmware is owned by emc, intel, and someone else
21:16<Isvara>And if they're so "pro-FOSS", why is it so bloody difficult to get VMWare working (and keep it working) on Linux?
21:16<path->mostly emc
21:16<path->they bascially control it
21:16<path->so the ceo was one of the founders
21:16-!-jcn [] has quit [Quit: ircII EPIC4-2.6 -- Are we there yet?]
21:17<mwalling>Isvara: it is?
21:17<path->vmware works on my linux machine
21:17<Isvara>mwalling: The stupid installation script that nearly always fails when you upgrade your kernel.
21:17<mwalling>my only issue was something screwy in the guest modules when you build them from source
21:17*Isvara keeps having to download some upgrade-150-150 script or something from some random site
21:18<Isvara>Otherwise it wouldn't build the kernel modules properly (not the guest kernel)
21:18<mwalling>21:17 < mwalling> my only issue was something screwy in the guest modules when you build them from source
21:19<Isvara>Yeah, this was not the guest modules.
21:19<Isvara>This is on the VMWare Server host
21:19<mwalling>thats what i ment
21:19<mwalling>hey, i'm tired
21:19<path->vmware server is pretty crappy compared to workstation or esx
21:19<mwalling>(and usually deploy vmware with linux guests onto windows hosts anyways)
21:20<path->i'm irritated that their vi3 client only runs on windows
21:20<straterra>thats why i have a win vm
21:20<path->i have a windows guest only to run the vi3 console to administer the esx servers
21:20<path->which is lame
21:20<Isvara>So anyway, VMWare has a new CEO from Microsoft, while Microsoft have built their own virtualisation technology, HyperV, using engineers from XenSource, and perhaps VMWare employees will quit and move to XenSource. It's all a bit mixed up.
21:22<Isvara>I'm sort of reminded of Stephen Colbert on AT&T
21:24<avongauss>I don't think you'll see that great of Exodus just yet, the clever ones will be look at it as an opportunity.
21:24<avongauss>The commentary about DEC is interesting though, I still have one of those Alpha processors.
21:24-!-jcn [] has joined #linode
21:25<path->a bunch of DEC engineers start equallogic
21:25-!-lolo [] has joined #linode
21:26<lolo>Hey all, i'm just trying out ubuntu 8.04 and it seems like wget is not installed....does that make sense?
21:26<HoopyCat>lolo: /etc/rc.d/rc.local would be a quick place to put something like that to occur on boot :-)
21:26<lolo>HoopyCat: thanks!
21:26<HoopyCat>lolo: apt-get install wget and you'll have it faster than... well, faster than i am. ;-)
21:27<path->linodes are a blank canvas
21:27<booja>mine still is
21:27*Isvara can't paint :-(
21:27<booja>the bottom corner has some scribbling
21:28*HoopyCat graffitis booja's linode
21:28<Isvara>I'm thinking about buying mode Linodes so I can make a flipbook.
21:28<HoopyCat>Isvara: that would actually be somewhat awesome
21:29<booja>lipbook ?
21:29<HoopyCat>Isvara: every time you install something, clone the image to another linode and do the next step there
21:29<Peng>Eek, I didn't think of that: Since I still use DreamHost for DNS too, their downtime will have impact on my Linode.
21:29<Isvara>booja: Lipbook? Like Facebook, but more focused?
21:29<Isvara>HoopyCat: LVM snapshots!
21:30<booja>paying lip service to the social networking phenomenon
21:31<Isvara>When I was young, I apparently hypothesised that it must be an ashtray.
21:31<HoopyCat>dude, that's... dangerous
21:33<HoopyCat>bbl, trying to make the third page of the metro section of the newspaper
21:33<Isvara>The nude section?
21:34<spasmface>... i gotta start reading the news paper
21:35<Isvara>Just subscribe to the Reddit RSS feed.
21:35-!-Schroeder2 [] has quit [Ping timeout: 480 seconds]
21:36<spasmface>nah i dont like news
21:36<Isvara>Then just subscribe to the RSS feed.
21:37<spasmface>its just like "this person got raped, this old guy got bashed then raped, a train derailed and raped a whole bunch of people, and a tsunami rape-enemaed a whole continent and heres a cute panda"
21:37-!-jcn [] has quit [Ping timeout: 480 seconds]
21:37<Isvara>Well there you go, then. There's a cute panda.
21:37<Isvara>Was the cute panda raping anyone?
21:38<booja>the cut panda raped an endangered cheetah
21:38<booja>cute even
21:38<Isvara>It was probably a cut panda after messing around with a cheetah.
21:39-!-Kassah [] has joined #linode
21:43-!-lakin [] has joined #linode
21:43-!-Kassah_ [] has quit [Ping timeout: 480 seconds]
21:46-!-Schroeder2 [] has joined #linode
21:53<SelfishMan>So there I am walking down the street...
21:53<Isvara>My street??
21:53*Isvara looks outside
21:53*Isvara closes the blinds
21:53<SelfishMan>When I realized that I wasn't wearing any pants
21:53<SelfishMan>So I did what any guy would do
21:54<SelfishMan>I kept on walking
21:58<JasonF>I just setup a network printer internet accessable.
21:58<Isvara>Excellent! To what IP address should I address my print job?
21:59-!-TheFirst [] has joined #linode
21:59-!-Kassah [] has quit [Ping timeout: 480 seconds]
22:00*Isvara connects
22:00<Isvara>Hey, you have the same printer as I do!
22:00<Isvara>What are the odds?
22:00*path- sets up his printer on
22:01<Isvara>path-: All...?
22:01<Isvara>... hosts. Awesome.
22:01<Isvara>There should be an 'all printers' address.
22:03<Isvara>Someone say
22:03<Isvara>Colloquy is weird.
22:04<Isvara>"JasonF: Isvara:" gave me a Growl popup saying "JasonF mentioned a highlight word," but I've never noticed it do that for "Isvara".
22:10-!-Kassah [] has joined #linode
22:12*SelfishMan launches DDoS against printer at
22:12<guinea-pig>Isvara: when you say it, or when others say it? :P
22:12<Isvara>Others, but it appears to be matching on 'Isvara'. I guess it's been doing it all along, but for some reason that particular one didn't disappear until I closed it manually.
22:13<guinea-pig>quite likely, it's matching on $NICK
22:13<Isvara>Yeah, but I dunno why one would stick.
22:14<TofuMatt>It matches "Matt" for me
22:14<Isvara>I'm both an IRC n00b and a Mac n00b, so neither Colloquy nor Growl are familiar to me ;-)
22:14*TofuMatt loves Colloquy, even though he's not an uber IRC or Colloquy user
22:14-!-scorche` [] has joined #linode
22:14*SelfishMan blames IRC for World War III
22:14<TofuMatt>Mac n00b = Mac user ;-)
22:15<path->Growl is the bomb
22:15<TofuMatt>At work, we don't say we're Mac users -- we're UNIX guys who think Mac has the best window manager ;-)
22:15<guinea-pig>i've never heard of either of those things
22:16<guinea-pig>i take it Colloquy is an IRC client... and Growl?
22:16<SelfishMan>guinea-pig: You've never heard of a Mac or UNIX?
22:16<guinea-pig>SelfishMan: no, i'm afraid not. I'm a linux guy...
22:16<TofuMatt>Growl is a notification thing for Macs
22:17<TofuMatt>It just displays little "growls" (pretty overlays on part of your screen) when stuff happens in various apps
22:17<TofuMatt>Firefox 3 will tell you there's a new version of an extension, or your name was mentioned in IRC, for instance
22:18-!-scorche [] has quit [Ping timeout: 480 seconds]
22:18-!-scorche` is now known as scorche
22:18*path- uses screen+irssi
22:18<path->but growl is nice.. it tells me the weather
22:18<TofuMatt>handy when you have a 1920 pixel wide monitor and aren't always paying attention to everything on your screen, haha
22:18<path->and when there is activity in iterm
22:19*TofuMatt should try iterm, he uses Leopard's Terminal
22:20<TofuMatt>anyone here have a "PossitveSSL" cert?
22:20<guinea-pig>how many of you own stock in Linux Gold?
22:20<TofuMatt>um, "positve"
22:20-!-TrevorP [] has quit [Read error: Connection reset by peer]
22:22<Isvara>TofuMatt: You should try GLTerm!
22:22-!-TrevorP [] has joined #linode
22:22<booja>linux gold ?
22:23<TofuMatt>Isvara: What will it give me that Leopard's doesn't?
22:23-!-Schroeder2 [] has quit [Ping timeout: 480 seconds]
22:23-!-Kassah [] has quit [Ping timeout: 480 seconds]
22:23<Isvara>TofuMatt: A convincing emulation of a 1970s green screen glass TTY!
22:24<TofuMatt>the screenshots on the site are from, like, Panther at least
22:24<TofuMatt>before my days as a Mac user
22:25<TofuMatt>"into Jaguar, the next major version of MacOS X" ... haha this is an old app :-)
22:27<Isvara>I don't think that's the one.
22:29-!-lakin [] has quit [Ping timeout: 480 seconds]
22:29-!-Chief [] has joined #linode
22:29<Chief>hey there :-)
22:29<Chief>Just looking in my mail log I notice this...
22:29<Chief>Jul 11 02:23:20 li32-138 postfix/error[31791]: 76CC410D92: to=<>, relay=none, delay=0.06, delays=0/0/0/0.05, dsn=5.1.3, status=bounced (bad address syntax)
22:29<TofuMatt>:O weird
22:30<Chief>which is odd because I have changed to by my hostname in /etc/hostname and also /etc/hosts
22:30<TofuMatt>this app look great :)
22:30<Isvara>Chief: What's that # doing there?
22:31<Chief>dunno thats the problem. I changed my postfix to show myhostname as something else
22:31<Chief>but no go...
22:32<TofuMatt>whoa, this must be a PowerPC app
22:35<Chief>ok I found that it was because the first line in the /etc/mailname had been commented out, which is why the # was there... but of course it picks the total first line of that file and dosn't look at comments
22:35-!-Kassah [] has joined #linode
22:36-!-Chief [] has quit [Remote host closed the connection]
22:37<Isvara>Hmm. Photos of a girl running fibre in a datacentre.
22:37*Isvara files under porn.
22:37-!-dvgrhl [] has quit [Remote host closed the connection]
22:37<spasmface>leopard's terminal is just as good as iterm these days
22:37<TofuMatt>yeah, that's what I've found
22:37<tjfontaine>spasmface: itym better
22:38<spasmface>negative capitan
22:38<TofuMatt>I have Leopard at home and work, but some guys at work still have Tiger, and one of them swear by iterm
22:38<spasmface>iterm is good or tabs, leo term already has it
22:39<JDLSpeedy>is dallas having network issues?
22:39<spasmface>and you can change background and keys n shit
22:39<spasmface>but i like it whiteb ecause when i use blacktree nocturne at night it goes black
22:39*TofuMatt is liking the full-screen GLTerminal ^_^
22:39<JDLSpeedy>caker: ?
22:39*TofuMatt uses the "novel" style in Leopard Term
22:40<spasmface>Have you used visor?
22:40<spasmface>quake-style terminal :D
22:40*Isvara has that installed
22:40<CaptObviousman>anyone feeling supa lag outta dallas?
22:41<JDLSpeedy>CaptObviousman: I am
22:41<JDLSpeedy>caker, mikegrb, tasaro: packet loss at dallas
22:41-!-Kassah_ [] has joined #linode
22:41<CaptObviousman>it's inside the DC. I'm connected to a mud who's also racked in there
22:42<CaptObviousman>it's lagging pretty badly
22:42<JDLSpeedy>224 packets transmitted, 143 received, 36% packet loss, time 223022ms
22:42<CaptObviousman>someone's leaning on the GENERATE 90% TRAFFIC button again
22:42<bd_>hmm, mtr NJ<->dallas has no problems
22:42<Isvara>Ping flood it. The packets should shake the button, waking up whoever is leaning on it.
22:42<JDLSpeedy>CaptObviousman: what host are you on?
22:43<bd_>or RR<->dallas
22:43<CaptObviousman>umm, I forget
22:43<JDLSpeedy>seams to be fine now....
22:43<straterra>i used to be on that
22:44*JDLSpeedy spoke to soon
22:44<tjfontaine>I've been seeing some lag out from my dallas node, but my connection to it was fine
22:44<CaptObviousman>other users are seeing it too, so it's not just us
22:44<JDLSpeedy>linode to linode fine, desktop to linode packet loss
22:45-!-bobby [] has joined #linode
22:45<bd_> dallas <-> NJ has been overall stable for a while :/
22:46<CaptObviousman>hmm, on the other server ( some users are fine, others are toasting
22:46<bobby>Does anyone know about the DNS manager in Linode?
22:46<CaptObviousman>what about it
22:46-!-neh [] has joined #linode
22:46<Isvara>You'll have to be more specific.
22:46<bobby>i want to host two domain on the same machine
22:46<bobby>i added two "master"
22:47<bobby>but only one is showing up
22:47<bobby>is there anything i need to do?
22:47-!-jcn [] has joined #linode
22:47<bd_>bobby: when did you add them?
22:47-!-Kassah [] has quit [Ping timeout: 480 seconds]
22:47<bobby>around 7pm
22:47<bd_>and how are you testing?
22:47<bobby>7PM EST
22:48<bd_>note that if you tried to access it too early, your ISP nameservers may have cached the negative result
22:48<bobby>i click on "check", and they both come back as "okay"
22:48<bd_>bobby: if you don't mind, what is the domain in question?
22:48<bobby>and also hitting the web address
22:49<bobby>thanks bd_, i can check again tmr morning i guess. but what is the difference between master and slave?
22:49<bd_>slave means it'll copy the entire domain from some other nameserver using AXFR
22:49<Isvara>bd_: The trouble is that it's hard to know what 'too early' is, because I don't think it says when the zone gets reloaded, or even whether it happens right away.
22:49<bd_>master means you edit it right there
22:50<bd_>Isvara: it's reloaded every 15 minutes - xx:00, xx:15, xx:30, xx:45
22:50<bd_>Check with nslookup domain
22:50<bd_>no caching there
22:50<Isvara>Oh, it is? I hadn't seen that.
22:50<linbot>New news from forums: Large Databases in Linux, Apache, Mysql and PHP (LAMP) Forum <>
22:50<bd_>bobby: have you checked that the top-level-domain server delegation is updated?
22:50<bd_>"Zones that require generation, activation/deactivation or deletion are done so every 15 minutes. "
22:53<bobby>bd_, how do i check if "top-level-domain server delegation is updated"
22:53<Isvara>bobby: What's the domain?
22:57<@mikegrb>should be better now
22:57<@mikegrb>(dallas packet loss)
23:00-!-euph [] has quit [Remote host closed the connection]
23:01<bd_>bobby: you'd host -v -t ns yourdomain, then host -t ns -v yourdomain one-of-the-servers-returned, etc, until you have entries of the form yourdomain. (number) IN NS (someserver)
23:01-!-Kassah_ [] has quit [Quit: Ex-Chat]
23:01<bd_>if those (someserver)s aren't ns[1234], you have a problem, which you need to sort out with your registrar :)
23:03<Isvara>dig ns
23:03<Isvara>We can assume com is correctly delegated ;-)
23:03<bd_>it might not be .com :)
23:03-!-Kassah [] has joined #linode
23:04<Peng>Dallas packet loss? What'd I miss?
23:04<Isvara>He doesn't seem to want to tell us.
23:04<Isvara>Same goes if it's a .net. .org has been all over the place!
23:05<Isvara>Or there's always Squish's DNS checker.
23:05<Peng>Ooh, my ssh connection timed out.
23:11-!-ElectricElf [] has joined #linode
23:13-!-meff [~meff@] has quit [Quit: If you ask you know I don't mind kneeling, but when my knees hurt, I'd like to stand.]
23:14-!-Kassah [] has quit [Quit: Ex-Chat]
23:16-!-Kassah [] has joined #linode
23:18-!-Electric1lf [] has quit [Ping timeout: 480 seconds]
23:19<CaptObviousman>oh lord, there it goes again. Yay packet loss!
23:19<Kassah>been having the same problem myself
23:19<CaptObviousman>Dallas is fucked up right now
23:19<CaptObviousman>someone needs to get null routed, and it ain't me!
23:20<Kassah>oh... that'
23:20<Kassah>that's the only location I don't have a linode at
23:20<Peng>I'm okay, I think.
23:21<Kassah>I'm having packet loss at home
23:21<Peng>SSH is laggy, but still connected.
23:23<Kassah>bah I need my new VMWare box so I can load up a router on it =)
23:25<TofuMatt> is slooooow too
23:26<TofuMatt>and by slow
23:26<TofuMatt>I mean not responding
23:26<bd_>mtr NJ <-> is getting a stable 38ms though
23:26<bd_>with no loss
23:26<bd_>same for
23:26<bd_>also from roadrunner maine :|
23:26<CaptObviousman>this os bloody awful
23:26<Kassah> seems speedy to me from CableOne Idaho
23:27<TofuMatt>which one?
23:27<Peng>Hmm, from DreamHost in California to Dallas, I got one packet lost total. From me in Florida, I got some bad loss in Level 3 and The Planet's network.
23:27<bd_>ah, my routes don't seem to be going through level3
23:27<CaptObviousman>always fucking level3
23:27<Peng>Or Savvis.
23:27<CaptObviousman>or Savvis, yes
23:28<CaptObviousman>what other backbone providers are there to blame? Cogent? SBC?
23:28<CaptObviousman>Worldcom used to be a backbone provider, weren't they?
23:28<Kassah>mine doesn't hit level3 or savvis either
23:28<Peng>Cogent are jerks, but I don't remember them being unreliable.
23:28*Kassah chuckles
23:29<Kassah>okay... brb... gonna try attaching the last device in the chain at my house and see if recreates my problem
23:30<Isvara>Kassah: Every time you do that my TV comes on.
23:30-!-nan [] has joined #linode
23:31<avongauss>I go away for a couple of hours and you all break Dallas. Sigh...
23:31<nan>hey everyone, I just signed up with linode but can't seem to login, it's telling me "Login incorrect or session timed-out". I do have cookies enabled. Is anyone else experiencing this?
23:31<Peng>nan: You do know your account has to be manually activated, right?
23:32<@tasaro>nan: i just activated your account
23:32<Kassah>Isvara: haha =) well glad I have a remote =)
23:32-!-Kassah [] has quit [Quit: Ex-Chat]
23:33<Isvara>Business at the Speed of Tasaro!
23:33<nan>oh ok thanks! I didn't know that
23:33*CaptObviousman keeps mtr going
23:34<CaptObviousman>hmm, a side effect of my iptables rules, I get large packet loss from my gateway
23:35-!-Kassah [] has joined #linode
23:36<Peng>I think Dallas is OK ATM.
23:38<avongauss>For a brief second when I sent out my last message, I was seeing something, but it's been fine since then.
23:40-!-nan [] has left #linode []
23:40-!-Kassah [] has quit []
23:44-!-Schroeder [] has joined #linode
23:45-!-Kassah [] has joined #linode
23:46-!-Dreamer3 [] has quit [Ping timeout: 480 seconds]
23:49-!-bobby [] has quit [Ping timeout: 480 seconds]
23:51<CaptObviousman>I was monitoring mtr, not a peep
23:54-!-Isvara [] has quit [Quit: Isvara]
23:54<bd_> nothing on my smokeping either :)
23:58-!-DennisTT [] has quit [Ping timeout: 480 seconds]
23:59-!-VS_ChanLog [] has left #linode [Rotating Logs]
23:59-!-VS_ChanLog [] has joined #linode
23:59-!-DennisTT [] has joined #linode
---Logclosed Fri Jul 11 00:00:40 2008