Back to Home / #linode / 2009 / 10 / Prev Day | Next Day
#linode IRC Logs for 2009-10-11

---Logopened Sun Oct 11 00:00:25 2009
---Daychanged Sun Oct 11 2009
00:00-!-mallory [~mallory@] has quit [Ping timeout: 480 seconds]
00:04<linbot>New news from forums: 2 tmpfs partitions on new Linode in General Discussion <>
00:13<TheFirst>i swear if there is anything more frustrating that getting xen installed I don't want to know what it is
00:14<Guspaz|m>That's why we let Linode do it :P
00:16<TheFirst>nah i want to set it up on a home machine for some things but ... argh!
00:16-!-MIRV- [] has joined #linode
00:16<MIRV->Hey guys how can I fix the hardware clock access error on my node?
00:17<MJCS>duct tape
00:17<MIRV->I wish
00:19<JoeK>submit a support ticket!
00:19<BarkerJr>MIRV-: where are you getting these errors?
00:20<MIRV->cron is kicking them out to my mail..
00:20-!-eTiger13 [] has joined #linode
00:20<BarkerJr>why is crond trying to access your clock?
00:20-!-bob2 [rob@2001:470:82de::1] has joined #linode
00:20<MIRV->because cron needs to know the time?
00:21<BarkerJr>what errors are you getting?
00:21<eTiger13>if i change the ip address on my a name records, how long does it take to show that?
00:23<BarkerJr>never seen it, sorry :/
00:23<HoopyCat>MIRV-: there is no hardware clock to set, so ignore the error and/or remove it from the startup/shutdown scripts
00:24<HoopyCat>eTiger13: less than 15 minutes for it to hit the servers, plus whatever the ttl was for it to clear everyone's caches
00:24<MIRV->but it loks like dcron itself is kicking out the error no?
00:25<HoopyCat>MIRV-: something in /etc/cron.hourly is trying to do it... maybe you have some cheap clock-setting script in there somewhere that does ntpdate then hwclock
00:26<MIRV->oh ya righty o
00:27<HoopyCat>MIRV-: you can safely zot it if you're using the default kernel (2.6.18-ish) since it slaves time by default; if you're running a newer experimental kernel, install ntpd and that'll take care of clock sync for you :-)
00:27<MIRV->ya I don't need that obviously
00:29<HoopyCat>anyway, i just spent three hours (!) poking at eagle trying to find a way to get no wires to cross in this amp design; i think i'm going to go to bed. i'm totally going to have nightmares
00:32-!-loxs [] has quit [Ping timeout: 480 seconds]
00:35<Smark>My laptop seems to have Interference issues with the Audio Bus... Theres always a buzzing/crackling in the background, it changes when I move my mouse. Makes me think theres crosstalk somewhere. Anything I can do about this?
00:36<BarkerJr>if you have PC Beep in your windows volume controls, try muting it
00:36<eTiger13>HoopyCat, hmm, any idea why it would take longer than 45 minutes?
00:37<Smark>BarkerJr, closest thing I have is System Sounds
00:38<@caker>eTiger13: ask our servers directly. Otherwise, you're probably seeing cached results
00:39-!-xQuasar [~xageha_@] has quit [Ping timeout: 480 seconds]
00:41-!-loxs [] has joined #linode
00:47<MJCS>get a line filter for your audio out
00:48<eTiger13>caker, ask the servers directly?
00:51<BarkerJr>host <hostname>
00:56<BarkerJr>that's how you ask the servers directly :)
01:00<MJCS>beans, beans the magical fruit...the more you eat the more you toot...the more you toot the better you eat your beans with every meal
01:07<eTiger13>I feel like I am missing something. I changed the a records in dns, updated apaches config to include my domain name and reloaded it. what else do i have to do?
01:07<MJCS>dns changes take time
01:10<eTiger13>isnt there a way to setup so the changes are instant? all im changing are the a name records
01:10<MJCS>24-28 hours
01:10<MJCS>depends on your ISP
01:10<MJCS>I've had domains take 2 weeks to update
01:12<Guspaz|m>I keep mine with a TTL of 30 minutes... It usually takes less than an hour for most people to get changes.
01:12<MJCS>yeah this si true but some people have shitty ISP's
01:13<MJCS>and eTiger13 has comcast so :(
01:13<Guspaz|m>ISPs usually obey TTLs.
01:15<bd_>keyword: usually
01:16<MJCS>comcast is known for bit shaping and other not so nice things
01:21-!-zed [] has joined #linode
01:21<eTiger13>can someone go to and tell me if goes to an actual site and not just a link
01:22<zed>Error: Server's host key did not match the signature supplied
01:22<zed>i am getting above error while setting FileZilla
01:23<zed>any one from linode team here??
01:23-!-Bass10 [] has quit [Ping timeout: 480 seconds]
01:25<MJCS>zeee I dont know
01:25<MJCS>probably sleeping
01:25<MJCS>Powered by Joomla!. valid XHTML and CSS.
01:25<zed>how to setup FTP for linode?
01:27-!-ravinder [LinodeJava@] has joined #linode
01:27-!-blognewb [~blognewb@] has quit [Quit: Client Quit]
01:27<zed>any moderator here?
01:30<zed>any support team memebr?
01:31-!-bogdanbiv [~quassel@] has joined #linode
01:32<BP{k}>they might be asleep
01:32<eTiger13>MJCS, thats what you see when you go there?
01:34-!-eTiger13 [] has quit [Quit: Leaving]
01:34-!-bogdanbiv [~quassel@] has quit [Remote host closed the connection]
01:35-!-bogdanbiv [~quassel@] has joined #linode
01:35<zed>anyone knows here how to setup FTP
01:37<CaptObviousman>Fart Transmission Protocol?
01:37<checkers>zed: what distro?
01:37<TheFirst> but if you're smart you'd require tls on it instead of plaintext
01:37*CaptObviousman demonstrates with a particularly wet one that might clear the channel
01:37<checkers>sudo apt-get install vsftpd
01:38<linbot>Please use SSH/SCP/SFTP/rsync-over-ssh instead of FTP:
01:38<TheFirst>no love for ftps?
01:38<bd_>Added bonus: SFTP and SCP are set up for you automatically.
01:39<bogdanbiv>Good morning Linoders!
01:40<zed>checkrs is it for FTP or SFTP...i need SFTP
01:41-!-Bohemian_ [] has quit [Quit: Bohemian_]
01:41<checkers>zed: sftp is not ftp
01:41<checkers>sftp runs over ssh and is already active
01:42-!-Bohemian [] has joined #linode
01:42-!-Bohemian [] has left #linode []
01:43<bogdanbiv>Yesterday, I found out from fellow Linoders Peng_ and Northwood that it's normal not to have kernel modules in the Ubuntu install.
01:44<zed>command for closing ubuntu
01:44<bogdanbiv>However, how can I install IPv6 support on it without the ipv6.ko?
01:44<packeteer>the xen kernel maybe
01:44<packeteer>zed: exit
01:44<packeteer>or Ctrl-D
01:45<TheFirst>bogdanbiv: just install the tools you need and away you go
01:46<bogdanbiv>Well, I configured /etc/network/interfaces to set up my ipv6 tunnel with hurricane's endpoint, but it's dead in the water.
01:47<TheFirst>if you did it by hand, with the commands he gives you it works?
01:48-!-ravinder [LinodeJava@] has quit [Quit: ravinder]
01:52<bogdanbiv>TheFirst: I tried the Hurricane's commands available on my tunnel's details page, but destination is still unreachable.
01:52<TheFirst>were you able to ping your side of the tunnel?
01:55-!-realityloop [] has joined #linode
01:55<bogdanbiv>TheFirst: Yeap, my side is ok.
01:56<TheFirst>just can't ping the ::1?
01:56<bogdanbiv>TheFirst: ::1 is fine too
01:57<TheFirst>ok so the tunnel is up...
01:57<zed>i want to use SFTP please guide me......
01:57<zed>using Ubuntu
01:57<bogdanbiv>Maybe I'm pinging the wrong address? I'm using 2001:xxx:2 for my endpoint and the server is 2001:same_XXX:1
01:58<bogdanbiv>TheFirst: is unkown.
01:59<bogdanbiv>I guess it resolves names over ipv6 too?!
01:59<TheFirst>using ping6 right?
01:59<zed>getting Error: Server's host key did not match the signature supplied
01:59<bogdanbiv>TheFirst: nope, I wasn't.
02:00<zed>while connecting thru filezilla
02:00<TheFirst>bogdanbiv: yeah...gotta use ping6
02:00<bogdanbiv>But ping6 hangs too...
02:01<bogdanbiv>zed: Could you check out that you have the correct ssh public key for yor server?
02:02<bogdanbiv>TheFirst: Should I file a issue report to get help from Linode support?
02:02<TheFirst>bogdanbiv: whats your ip -6 route?
02:03-!-JoeK [] has quit [Remote host closed the connection]
02:03<bd_>bogdanbiv: ipv6 isn't a linode service
02:03<bogdanbiv>Wait, I messed it up... I now have two tunnels
02:04<bd_>so they're not obligated to support you with it
02:04<TheFirst>bogdanbiv: so ip tunnel del them and start over
02:04<bogdanbiv>TheFirst: I'm too used with windows approach - I just restarted it
02:08<bogdanbiv>TheFirst: here's defa default dev he-ipv6 metric 1024 expires 21334357sec mtu 1480 advmss 1420 hoplimit 4294967295
02:08<TheFirst>bogdanbiv: so you've rebooted and rerun the commands from he?
02:09<bogdanbiv>TheFirst: Yes
02:10<TheFirst>bogdanbiv: and you can ping6 both sides of the tunnel?
02:10<bogdanbiv>TheFirst: 2001:xxxxxxxxxx::/64 via :: dev he-ipv6 metric 256 expires 21334357sec mtu 1480 advmss 1420 hoplimit 4294967295
02:10<bogdanbiv>TheFirst: The ssame, I can ping6 mine, but not servers...
02:10-!-JoeK [] has joined #linode
02:11<TheFirst>pastebin me your routes
02:13<zed>help me to setup SFTP
02:15<bogdanbiv>zed: What OS do you use as desktop?
02:16<zed>local Win 7 and in linode Ubuntu
02:16-!-purrdeta [] has quit [Ping timeout: 480 seconds]
02:17<TheFirst>bogdanbiv: probably a stupid question but you're sure you've got the ips right? I made that mistake .. the tunnel was 1f0f and the /64 was 1f0e in my case
02:17<bogdanbiv>Sftp works by default on your Ubuntu linode. From my Linux desktop I do: s
02:18<bogdanbiv>TheFirst: Sure i do, I copy paste from Hurricane's page.
02:19<bogdanbiv>They give commands on how to setup your tunnel with complete details.
02:19<bogdanbiv>zed: wait a sec.
02:21<bogdanbiv>TheFirst: I've updated the pastebin with my ip tunnel setup commands.
02:21<bogdanbiv>They give two choices and said both work fine.
02:22<TheFirst>bogdanbiv: what are you doing with sit0/sit1?
02:22<bd_>zed: install winscp locally, and you're done
02:22<bogdanbiv>TheFirst: Onestly, I'm not sure!
02:23<TheFirst>bogdanbiv: you don't need to be running the last set of commands ... 53-57 should do it
02:23-!-purrdeta [] has joined #linode
02:23<bogdanbiv>I tried both - deleting the first tunnels.
02:24<zed>i hav just insatlled winscp but getting error "Server's host key didnot match the signature "
02:24<TheFirst>set sit0/sit1 down, make sure they're not in your routes and rerun 53-57 ...
02:25<bogdanbiv>TheFirst: I haven't used the last command set since restarting the instance, so it's clean now.
02:26<bogdanbiv>zed: the problem is that I guess you have an old ssh key on your client.
02:26<bogdanbiv>zed: It fails at verifing the new key with the old key fingerprint.
02:27<TheFirst>bogdanbiv: this isn't a linode?
02:27<bogdanbiv>It sure is!
02:27<TheFirst> goes to slicehost so it would seem
02:28-!-thornheart [] has joined #linode
02:28<bogdanbiv>TheFirst: :D :)
02:29<bogdanbiv>TheFirst: OMG!
02:29<bogdanbiv>TheFirst: How Could It work?
02:32<bogdanbiv>TheFirst: Success: 88f::1: icmp_seq=1 ttl=64 time=92.4 ms
02:32<TheFirst>try google then
02:32<bogdanbiv>I had a slicehost account previously.
02:33<TheFirst>assumed that
02:33-!-nb_ [] has quit [Read error: Connection reset by peer]
02:33<bogdanbiv>I assumed I deleted the right tunnel too, it seems I haven't
02:33<bogdanbiv>And I was using the old one
02:34*TheFirst spent about 6 hours friday night getting his lan on ipv6 ... pain in the ass
02:35-!-nb_ [] has joined #linode
02:35<bogdanbiv>TheFirst: I have spent 3 months getting my Asus (wl500gPv2 loaded with DDWRT/OpenWRT) to work with IPv6 from hurricane... at no avail though.
02:36<TheFirst>bogdanbiv: i use tomato ... two iptables rules to forward proto 41 and worked like a charm
02:36<bogdanbiv>TheFirst: I had hit various walls such as DD-WRT or OpenWRT bugs
02:36-!-nb [] has quit [Ping timeout: 480 seconds]
02:36-!-nb_ is now known as nb
02:37<bogdanbiv>Does it work with pppoe too?
02:37<TheFirst>dd was nothing but a problems for me ... hence using tomato
02:39<bogdanbiv>I'll try because I'm not pleased with DDWRT anyway: from time to time it hangs, failing even to answer pings (local domain side)
02:39<TheFirst>yeah thats what it did to me
02:40<TheFirst>been running tomato for years without any problems
02:40<bogdanbiv>If I unplug it and replug it after 5-10min it works like a charm.
02:41<bogdanbiv>How is the support community there?
02:41<TheFirst>never needed to look for support
02:41<bogdanbiv>TheFirst: I have to try that!
02:42<bogdanbiv>zed is silent?
02:42<bogdanbiv>zed: Did you manage to solve your problem?
02:42<TheFirst>iptables -t nat -A PREROUTING -i vlan1 -p 41 -j DNAT --to
02:42<TheFirst>iptables -t filter -A FORWARD -i vlan1 -p 41 -d -j ACCEPT
02:43<TheFirst>that's what you need for ipv6 with tomato
02:44<bogdanbiv>But isn't that only for 1 host?
02:44<TheFirst>yeah, then you set up that host to route for the rest of the pc's on the lan
02:44<TheFirst>which really amounts to get the tunnel up and install radvd
02:44<linbot>New news from forums: General Confusion in General Discussion <>
02:46<bogdanbiv>TheFirst: Yeah, it could work since I keep a server up on my lan.
02:47<TheFirst>same ... tomato doesn't do native ipv6 ... only openwrt and dd do that
02:49<Guspaz|m>Does IPCP even support ipv6?
02:49<bogdanbiv>Guspaz|m: IPCP?
02:50<Guspaz|m>Internet Protocol Control Protocol. What PPP uses to tell the machine what IP settings to use.
02:50<Guspaz|m>And set up the tunnel
02:50-!-bogdanbivi [~quassel@] has joined #linode
02:54<zed>bogdanbiv thr?
02:55<bogdanbivi>Some where inside your client config you have an old ssh key which causes the error.
02:56<bogdanbivi>How do I see what I have missed while disconnected?
02:57<TheFirst>bogdanbiv: ?
02:57<bogdanbivi>chesty: I vim your /etc/dog
02:57-!-au [~sam@] has joined #linode
02:57-!-bogdanbiv [~quassel@] has quit [Ping timeout: 480 seconds]
02:58<chesty>hmm, there's logs somewhere on the net, I can't remember where
02:58<linbot>bogdanbivi: (help [<plugin>] [<command>]) -- This command gives a useful description of what <command> does. <plugin> is only necessary if the command is in more than one plugin.
02:58<au>How do I make Apache show CentOS in footer instead of RedHat? I don't have RedHat :p
02:58<au>it says: Apache/2.2.3 (Red Hat) Server at 74.207.---.-- Port 80
02:59<coax_>au you can change that with mod_security
02:59<coax_>i think theres a way to compile it also with a new message
02:59<coax_>but i dont remember the procedure
02:59<au>alright, i'll look at that
03:00<bogdanbivi>zed: Please give me your error message again, I fail to find it.
03:00-!-syntaxman [~wade@] has joined #linode
03:01-!-bogdanbivi is now known as bogdanbiv
03:02-!-wintermi [] has joined #linode
03:12<zed>ok thanks
03:13-!-au [~sam@] has quit [Quit: Goodbye.]
03:13-!-elhippo [] has quit [Quit: Leaving]
03:14<zed>ok let me giv it a try
03:17<bogdanbiv>zed: Could you try to connect to the server via Putty SSH?
03:19<bogdanbiv>Guys and gals, I still have a question: how come there are no kernel modules in Linode VPS? My ipv6 works now, with help from TheFirst, but I can't understand why.
03:19-!-JoeK [] has quit [Remote host closed the connection]
03:21-!-JoeK [] has joined #linode
03:23<bob2>because all the basic stuff is built into the kernel
03:23<bob2>ipv6 is a bad example anyway, since just enabling it as a module alters the kernel image itself
03:25-!-JoeK [] has quit [Remote host closed the connection]
03:25<bogdanbiv>But I thought Xen was a module too.
03:26<bogdanbiv>Ah, I get it: one can compile the kernel modularly or as a monolith, right?
03:26<bob2>well, it is a spectrum rather than a yes/no thing
03:29-!-newnick123 [] has joined #linode
03:30<newnick123>I have a question
03:30<newnick123>I did ps -aux|more and I have around 15% cpu but my linode's graphs say it is at 100% for the past 24 hours.
03:30<newnick123>Does anyone know the reason for this?
03:31-!-realityloop [] has quit [Quit: realityloop]
03:31<chesty>try running top
03:31<chesty>also look in your logs for OOM
03:32<newnick123>Sorry for my ignorance, but what is top?
03:33<bob2>top is the command to show reeal time cpu and ram use
03:34<newnick123>Everything is at 0.0
03:35<newnick123>Everything is at 0.0% in tops :P
03:38<newnick123>Rebooting solves everything, thanks for your help
03:38-!-newnick123 [] has quit [Remote host closed the connection]
03:39<chesty>till it OOMs again
03:45<bogdanbiv>This guy uses variables in his networking config:
03:46<bob2>this guy expects you do the variable substitution yourself ;)
03:46<bogdanbiv>Where Do I need to setup these variables? Can I do it in the script config file itself?
03:46<bob2>(i.e. if you have any $s in your /etc/network/interfaces, you screwed up)
03:46<bogdanbiv>bob2: Oh, well...
03:46<bob2>it is extremely common to write documentation in this manner
03:47<bob2>ignore the last step re: linode, btw, it's a lie
03:49-!-loxs [] has quit [Ping timeout: 480 seconds]
03:51<bob2>make sense?
03:54-!-JoeK [] has joined #linode
03:56-!-shakr [] has quit [Quit: Leaving]
04:15-!-blognewb [~blognewb@] has joined #linode
04:15-!-shakr [] has joined #linode
04:15-!-chesty [] has quit [Quit: Coyote finally caught me]
04:18-!-chesty [] has joined #linode
04:28-!-wintermi [] has left #linode []
04:50-!-blognewb [~blognewb@] has quit [Quit: Client Quit]
05:53<linbot>New news from forums: What do you use your Linode for? in General Discussion <>
06:02<JoeK>any reason why my atlanta linode isnt resolving the new ip of a domain?
06:02<JoeK>its been 2 hours since i updated it and all my other vpses are able to resolve it
06:11-!-Smark is now known as Smark[Gone]
06:16-!-loxs [] has joined #linode
06:18-!-laser` [] has joined #linode
06:21<bob2>JoeK: what's the domain?
06:21<JoeK>nevermind, it was just the app
06:29-!-vuf [~am@] has joined #linode
06:31<vuf>hi, I have a cron script pinging one linode from another, on the private network, every minute ... and it fails once every few days. anyone with similar experience?
06:32<encode>fails how?
06:32<encode>drops a single ping?
06:36<vuf>yes, and before you say that icmp is not reliable, I set this up because tcp connects to the database also fail every now and then
06:37-!-Edgeman [~edgeman@] has quit [Ping timeout: 480 seconds]
06:37<chesty>check ifconfig for errors at both ends
06:37<chesty>hmmm, that may not work
06:37<chesty>being a virtual interface an all
06:38<vuf>probably not. anyway, it's zero
06:39<vuf>i remember a discussion about packet loss at newark some weeks ago (i also had it back then), but I cannot find it in the irc log
06:39<chesty>put in a ticket
06:39<coax_>im at newark im not seeing any packet loss atm
06:40<vuf>this is veery rare, I would probably not notice it every week without the cron script
06:41<coax_>would such packet loss result in a disconnection from an irc server
06:41<coax_>my irc bot is stable for 3 weeks
06:41<vuf>i did put in a ticket, but before more people see the problem, what can they do ...
06:41<chesty>look at the physical interfaces for errors
06:42-!-redgore [] has joined #linode
06:42<chesty>do their own testing
06:43<vuf>well, right, they can do something, but I would not expect it for just a single user losing a packet every other day ... it would be fair to assume that it is me that's an idiot
06:45<coax_>vuf did you do a traceroute
06:45<coax_>cron a traceroute every now and then see what comes up
06:46<chesty>it's a flat network
06:49<vuf>thanks for your input, I will check here again at a busier time to maybe find someone with similar experience
06:51-!-Edgeman [~edgeman@] has joined #linode
06:53-!-Jonathan [] has quit [Ping timeout: 480 seconds]
06:57-!-realityloop [] has joined #linode
06:58-!-loxs [] has quit [Ping timeout: 480 seconds]
07:04-!-mcinerney [] has joined #linode
07:04-!-ph_ [~ph^] has joined #linode
07:05-!-Shinsaku [] has joined #linode
07:05-!-Big-Mama [] has joined #linode
07:07-!-loxs [] has joined #linode
07:08<Big-Mama>Gaah, any PHP experts here?
07:08<Big-Mama>Great! :)
07:08<Big-Mama>How can I shorten this varible: $name = "Gerald David McMillan"; | --> Gerald McMillan?
07:08<Big-Mama>I just want to have the first and last word in an variable.
07:09<laser`>$pieces = explode($name); $firstandlast = $pieces[0] . " " . $pieces[2];
07:09<laser`>Or something like that
07:09<laser`>$pieces = explode(" ", $name); $firstandlast = $pieces[0] . " " . $pieces[2];
07:10<Big-Mama>Thanks! Ill try it out.
07:11-!-peter [] has quit [Ping timeout: 480 seconds]
07:11<vuf>$name = preg_replace("/(\w+).*(\s\w+)/", "$1$2", $name);
07:30-!-Big-Mama [] has quit [Ping timeout: 480 seconds]
07:44-!-cpg [] has quit [Quit: cpg]
07:51-!-Shinsaku [] has quit []
07:59-!-realityloop [] has quit [Quit: realityloop]
08:05-!-tylerdu [] has quit [Quit: leaving]
08:13-!-tylerdu [] has joined #linode
08:14-!-ph_ [~ph^] has quit [Read error: Operation timed out]
08:15-!-ph_ [~ph^] has joined #linode
08:17*amitz just solved a non-existent bug! yay!
08:23-!-coax_ [] has left #linode []
08:31<amitz>say, is evdo reliable in your experience? I'm considering this new unlimited evdo based ISP.
08:33-!-zed [] has quit [Remote host closed the connection]
08:33-!-bogdanbiv [~quassel@] has quit [Read error: Operation timed out]
08:51-!-KingTarquin [] has joined #linode
08:51<KingTarquin>How do I go about obtaining an image of one of my linodes?
08:52<A-KO>you could tar up the file system and then dl it from your machine
09:05-!-bogdanbiv [~quassel@] has joined #linode
09:06-!-shakr [] has quit [Remote host closed the connection]
09:13-!-Bass10 [] has joined #linode
09:23-!-Shinsaku [] has joined #linode
09:41-!-alnr [] has left #linode [WeeChat 0.3.0]
09:57<HoopyCat>amitz: it's all in the implementation. my phone has an evdo backhaul which... well, i haven't had any problems fixed by temporarily disabling evdo, if that's what you're asking
09:58<HoopyCat>KingTarquin: boot up finnix (or another temporary distro) and use dd to copy the image elsewhere; like the custom distribution howto, but in reverse
10:03-!-loxs [] has quit [Ping timeout: 480 seconds]
10:07-!-silverblade [] has joined #linode
10:10<linbot>New news from wiki: Main Page <> || Talk:IPv6 <>
10:12-!-loxs [] has joined #linode
10:35-!-supine [] has quit [Ping timeout: 480 seconds]
10:38-!-packeteer [] has quit [Ping timeout: 480 seconds]
10:44-!-redgore [] has quit [Quit: Leaving]
10:45-!-packeteer [] has joined #linode
10:49-!-redgore [] has joined #linode
10:53-!-ericoc [] has joined #linode
10:56<linbot>New news from wiki: Main Page <>
10:59-!-bhankins [] has joined #linode
11:10-!-packeteer [] has quit [Remote host closed the connection]
11:10-!-idontworkhere [] has joined #linode
11:13-!-packeteer [] has joined #linode
11:14<amitz>HoopyCat: thanks. It's more like I'm wodnering the connection quality of EVDO using and EVDO modem. I'm looking at wikipedia right now.
11:17-!-aaronpk_ [] has joined #linode
11:19<HoopyCat>amitz: it can be really good. it can be really crappy. it depends entirely on the provider, how they run things, and what equipment they use, alas.
11:22-!-aaronpk [] has quit [Ping timeout: 480 seconds]
11:22<amitz>HoopyCat: yeah, I'm aware how different it can be according to circumstances. I just plan to average people's opinion :-p. Anyway, one vital question if you know the answer by any chance, is the signal strength of device that support CDMA 2000 1xRTT reflects the signal strength of EVDO in that particular place?
11:27-!-TheFirst [] has quit [Quit: You're a bloody puppet!]
11:28<linbot>New news from wiki: Internal Services <>
11:34-!-bogdanbiv [~quassel@] has quit [Ping timeout: 480 seconds]
11:34-!-Guspaz [] has quit [Read error: Connection reset by peer]
11:34-!-exor674 [] has quit [Ping timeout: 480 seconds]
11:34-!-Guspaz [] has joined #linode
11:36-!-redgore [] has quit [Remote host closed the connection]
11:42<quellhorst>what raid controller does linode use these days?
11:43<quellhorst>i need one for my home machine
11:44-!-Frools [] has quit [Ping timeout: 480 seconds]
11:46<HoopyCat>amitz: sometimes. in theory, the carrier will use the same towers for both 1x and EVDO, but i believe it uses different RF gear and antennas, may use an entirely different frequency, and isn't necessarily installed at *all* the carrier's sites
11:47<HoopyCat>amitz: that said, using the "bars of signal" test (which is about one notch above a random number generator), i usually have slightly better EVDO signal than 1x, especially interior to buildings. right now, f'instance, my cellphone reports 4/4 EVDO and 3/4 1x from my holster
11:49-!-ericoc [] has quit [Quit: leaving]
11:49<HoopyCat>amitz: (this, of course, applies only to verizon wireless in the three places i've noticed it: home, a hallway in building 9, and the middle of a forest in the middle of nowhere where i had 1 bar of EVDO and no 1x, despite somehow not being able to reach town with 50 watts of VHF into a high-gain antenna)
11:51<HoopyCat>speaking of which, those bastards want me to find the derivative of f(x)=x(x^2+1)... i'll show them. afk
11:51-!-Frools [] has joined #linode
11:53<amitz>yeah, the bars of signal never make sense. Never comparable to other handphones. I believe some cellphone manufactures skew the signals to make it as if their phones have good reception. anyway, for voice purposes, signal strength in the city is never a problem.
11:53-!-TheJoe|ZzZz is now known as TheJoe
11:56-!-adnc [] has joined #linode
12:00<HoopyCat>amitz: given the nature of CDMA, compression, and microwave RF, there's no particularly good way to figure out "how good the signal is" without two human brains actually evaluating the intelligibility of a two-way phone call. so, it's bollocks, but they can make some good guesses based on control channel error rates, signal/noise ratio, etc.
12:12-!-Jonathan [] has joined #linode
12:19<amitz>HoopyCat: enjoying school so far? Won't you learn something you already learned? Or your current knowledge doesn't really intersect with school?
12:19-!-nb_ [] has joined #linode
12:20<amitz>I wish I can go back to school again.
12:23-!-idontworkhere [] has quit [Remote host closed the connection]
12:25<Jonathan>Hi Guys - would the cidr cover and
12:25<tylerdu>no and yes
12:26<HoopyCat>Jonathan: no (second octet is outside of the mask) and yes...
12:26-!-loxs [] has quit [Quit: Leaving]
12:26<Jonathan>thanks, trying the url
12:27<HoopyCat>... which doesn't show host ranges for /8, /16, and /24 networks, of course, but those ought to be memorized as 1.*.*.*/8, 1.2.*.*/16, and 1.2.3.*/24 i suppose :-)
12:28<HoopyCat>amitz: so far, there's little intersection, which is nice. even when there has been, it's been not a bad thing.
12:28<Jonathan>I think need a network to english dictionary
12:29<BarkerJr>and :)
12:30<HoopyCat>back when i did networking every day, i had something like pinned on my wall so i could translate CIDR notation to netmasks in a hurry
12:30<Jonathan>plugging in a cable is about as far as my networking knowledge extends
12:31<BarkerJr>well, CIDR confuses a lot of people who know networking
12:32<Jonathan>oh, you know its not that, im just stupid, I didn't notice the different numbers
12:32<BarkerJr>you'd think they'd have been able to make it easier to understance, rather than using bits
12:32<Jonathan>I thought they were both on 192.168 and they are not
12:32<BarkerJr>yeah, I'm honestly amazed tylerdu caught that... I didn't timm my third read :)
12:32<Jonathan>nope, they are, i just typo'd it
12:32<HoopyCat>aside from that, it's just knowing that a dotted-quad IP address is merely a convenient representation of a 32-bit binary number and then knowing that /24 says the first 24 bits are "fixed" (leaving the least significant 8 as host addresses, aside from the all-zeros and all-ones), and then knowing that a netmask shows 1 for the "fixed" addresses and 0 for the "host" addresses
12:33<HoopyCat>and then... say, do you need a pencil?
12:33<Jonathan> and
12:33<BarkerJr>what's really fun is when you start dealing with /20 and such
12:35-!-redgore [] has joined #linode
12:35<BarkerJr>for instance, is 72.14.176.* through 72.14.191.*
12:35<Jonathan>whats 192.168.134.*? :D
12:36<Jonathan>thats what I put, but doesn't seem to like it
12:36<BarkerJr>well, you have to replace the * with 0
12:37<Jonathan>yeah I did
12:37<HoopyCat>Jonathan: yup, and are within
12:37<Jonathan>great, thanks a lot, probably messed something else up then
12:37<BarkerJr>maybe the software you're using doesn't support cidr :)
12:38<Jonathan>it asked for in cidr specifically (I never heard of cidr before)
12:38<Jonathan>I'm trying to setup iptables
12:39<BarkerJr>well, I don't run firewalls, so I can't help with that :)
12:40<Jonathan>thanks for the subnet help :-)
12:40-!-mysty [] has joined #linode
12:40-!-syntaxman [~wade@] has quit [Ping timeout: 480 seconds]
12:40<HoopyCat>Jonathan: yup, iptables likes CIDR. (which is nice, because i like CIDR too)
12:41<amitz>say, how to populate /sys when installing using debootstrap? I forget..
12:41<HoopyCat>Jonathan: if you remember old-timers (and/or people brainwashed by the Church of Ci$co's auditing^Wcertification training) mentioning "class C", for instance, that's /24.
12:41<bd_>manually mount -t sysfs none /sys ?
12:41<Jonathan>yeah I don't know what i've done wrong, the machine is accessible from the public IP, but the private one is blocked... and I'm using the setup script for dummies and still got it wrong
12:41<Jonathan>ahhh, yes I've heard people say class C
12:42<HoopyCat>Jonathan: CIDR stands for "classless inter-domain routing"... the lack of "class C", etc, is where the "classless" comes from :-)
12:42<HoopyCat>that, and the fact that it belches at the dinner table and passes great wind
12:43<BarkerJr>I remember Quake 1 rejecting connections to my LAN server cause the player wasn't on my Class C
12:43<Jonathan>haproxy is requesting a file, and the log keeps saying: abcmurders kernel: Spoofed packet: IN=eth0 OUT= MAC=fe:fd:61:6b:81:7c:fe:fd:61:6b:80:ed:08:00 SRC= DST=
12:46<amitz>bd_: ah, that sounds right, thanks!
12:46<mysty>does anyone know if fail2ban will work on top of a CentOS with Bastille & psad going?
12:50<vuf>hi, I have a cron script pinging one linode from another, on the private network, every minute ... and it fails once every few days. anyone with similar experience?
12:54-!-silverblade [] has quit [Remote host closed the connection]
12:55<HoopyCat>Jonathan: there's probably a rule hitting it (whichever one is the log statement with "Spoofed packet" in it)... a lot of "prefab" iptables packages do weird stuff on linodes, especially if they consider "public" vs. "private" interfaces
12:55<HoopyCat>mysty: dunno. it should.
12:56<HoopyCat>vuf: what are the conditions that would cause it to report a failure? a single packet dropped, an ICMP error, a...?
12:56<Jonathan>HoopyCat - I think, just reading the output when I tried to reconfigure, it doesn't like aliased interfaces - Warning: weird character in interface `eth0:0' (No aliases, :, ! or *).
12:56<laser`>Isn't a spoofed packet one where the origin is set to something not belonging to an interface
12:56<Jonathan>Which would explain the behaviour
12:56<HoopyCat>Jonathan: yup, that would do it
12:57<mysty>HoopyCat: thanks - seemed to install OK, not sure whether shorewall (fail2ban dependency) and iptables play nice together, guess I will find out :)
12:57<vuf>HoopyCat: ping -c1 -w1 address
12:59<HoopyCat>vuf: on a real network, that will fail once in awhile. try -c5 or so, with a pass if 3 or more are received
12:59<HoopyCat>assuming a failure every 3 days, you're seeing 0.023% packet loss :-)
13:00<HoopyCat>ugh, i should have cleaned this television before turning it on
13:02<vuf>HoopyCat: I know. I did not ask for advice, but similar experience. I set up this cron script because my database connections fail every now and then
13:02<Jonathan>I just rolled the rules onto eth0 and got rid of the aliases, it all seems good, thanks all
13:03<HoopyCat>vuf: on other real-world networks, yes.
13:04<mysty>this cant be good. bastille-firewall & ip6tables are 2345 on, iptables and shorewall are off
13:05<vuf>HoopyCat: do you see any reason that database connects between two linodes should fail?
13:05-!-Keith-BlindUser [] has quit [Quit: Freedom Chat - Your Home Away From Home | | tIRC script by the Freedom Chat Leets]
13:06<mysty>I would have thought starting fail2ban would have started shorewall if it was a dependecy. Maybe it works with ip6tables
13:09<HoopyCat>vuf: not generally, no. it should be handled pretty transparently by the client if it does happen, i figure.
13:11<HoopyCat>vuf: if it is maintaining a long TCP connection that sometimes goes idle for awhile, sending keepalives or noop queries every few minutes might help. it's a trick that normally works with stateful firewalls/NATs
13:12<HoopyCat>less than 5 minutes into the first quarter and the fans are already booing
13:15-!-Turl [] has joined #linode
13:22<innociv>linode's are 32bit right?
13:22<@caker>they're either
13:23<chesty>either 16 bit or 32 bit
13:23<innociv>How do I check which I'm using?
13:23<chesty>uname -a
13:24-!-elfgoh [] has quit [Quit: Leaving]
13:24<innociv>i686. So 32bit right?
13:26<Jonathan>I've managed to stop ssh access to a linode somehow. I've disabled the firewall but still no access. Lish can connect. sshd is running, anyone have any ideas what I might have messed up?
13:28<chesty>iptables -nvL ;netstat -nap | grep 22;cat /etc/hosts.deny
13:29<Jonathan>yup, I'm in there
13:34-!-Shinsaku [] has quit []
13:41-!-memenode [] has joined #linode
13:42-!-lesouvage [~chatzilla@] has joined #linode
13:45-!-internat [] has quit [Read error: Connection reset by peer]
13:45-!-internat [] has joined #linode
13:46-!-derek [] has joined #linode
13:46-!-nb [] has quit [Quit: nb]
13:46-!-nb_ is now known as nb
13:47-!-derek [] has quit []
13:47-!-Big-Mama [] has joined #linode
13:48<linbot>Big-Mama: Now 66% full. Last emptied Saturday at 10:30 EDT, last full on Saturday at 09:30 EDT after running for 39.0 hours.
13:48<Big-Mama>Haha HoopyCat :D
13:48-!-liberfiasco [] has quit [Ping timeout: 480 seconds]
13:52<bhankins>hehe, who's dehumidifier is that coming from?
13:52<purrdeta>HoopyCat I think
13:53<Big-Mama>what is a dehumidifier?
13:53<purrdeta>It sucks water out of the air.
13:53-!-redgore [] has quit [Remote host closed the connection]
13:53-!-BarkerJr [~BarkerJr@] has quit [Read error: Connection reset by peer]
13:53-!-TheFirst [] has joined #linode
13:53<purrdeta>that makes it sound a lot cooler than it really is
13:54-!-Turl [] has quit [Quit: Leaving.]
13:55-!-NotInternat [] has joined #linode
13:59-!-z8000 [~z@] has joined #linode
13:59-!-z8000 [~z@] has quit []
14:00-!-internat [] has quit [Ping timeout: 480 seconds]
14:02-!-Turl [] has joined #linode
14:03-!-gmcharlt [] has joined #linode
14:08<bhankins>If it was emptied two Saturdays ago I would it give a date?
14:10<bhankins>Never mind that.
14:11<HoopyCat>bhankins: it only scans back 5 days, so it'd give "unknown"
14:11-!-quellhorst [] has left #linode []
14:12<bhankins>That's pretty cool, does the dehumidifier itself have some kind of interface? or is that something you put in to the tank to measure it?
14:14<bhankins>I am making some reporting for my roomba so I can monitor it from work.
14:16<nb>is it possible for staff to see if they can give me another IP in the same subnet so I don't have to have 2 gateways?
14:16<HoopyCat>bhankins: mwalling's been working on some stuff for a roomba
14:17-!-TheFirst [] has quit [Quit: You're a bloody puppet!]
14:17<HoopyCat>nb: you don't need to set two; they're on the same physical subnet, so just pick one
14:17<nb>so even though the network tab lists 2 gateways i can just use one?
14:18<bhankins>Thanks a bunch.
14:19<HoopyCat>nb: yup... they're just both listed in case you ditch the other IP at some point
14:19<nb>HoopyCat, ok thanks, that makes it simpler :)
14:20<bhankins>Oh yeah, I see his twitter account too about the dehumidifier. That's awesome.
14:26-!-adnc_ [] has joined #linode
14:27<amitz>say, how do you guys manage your scripts? By functions, by dates, longnames,shortnames, multiple criteria by softlink,etc? Any tool to manage scripts?
14:28<HoopyCat>amitz: i manage my scripts the same way i manage my women: dark roasted and in the freezer
14:28<amitz>scripts for admin works.
14:28<HoopyCat>... wait
14:28<HoopyCat>amitz: git
14:29<amitz>SpaceHobo: HoopyCat: what do you mean keeping and managing it using version control?
14:29<amitz>or do I miss something?
14:30-!-elhippo [] has joined #linode
14:31<amitz>well, yeah, since you develop the script and changing it you want to version control it. But how do you manage the scripts such that you can easily search a script you want to use?
14:33-!-adnc [] has quit [Ping timeout: 480 seconds]
14:33<amitz>well, over time I may create script to mirror something, or script to clean up printing jobs and cache. One day I may forget that I had created such script thus I need to be able to find out if I already made one.
14:34-!-nb [] has quit [Read error: Connection reset by peer]
14:35<tkoskine>Put them all under one directory and use meaningful names instead of and ;)
14:37<amitz>tkoskine: long descriptive names? Reading them will be cumbersome and auto complete doesn't help if I don't know the first word. So maybe there is like a good thumb rule for the first word? Something like that.
14:38-!-tsp [] has joined #linode
14:39<amitz>or maybe shorter names but put under directory? Maybe there is a directory consisting of all scripts that touch a certain application? or a certain function, like printing, installing, etc?
14:39<HoopyCat>amitz: i write blog posts about the good ones, so when i need to do it in the future, i google for it and find that i already wrote it
14:40<tkoskine>Version control can solve that problem also. Put the description of the scripts in the log messages and later just grep the log.
14:41<amitz>HoopyCat: oh, interesting. tkoskine: oh, that's how to use the version control!
14:41<HoopyCat>amitz: i also keep all my trees in /home/rtucker/dev on my workstation, push them to github, then pull them elsewhere as i use them and make symlinks from /usr/local/bin or /usr/local/sbin, etc
14:41-!-nb [] has joined #linode
14:42<HoopyCat>lrwxrwxrwx 1 root root 52 2009-08-23 14:51 /etc/munin/plugins/dehumid_hoopydehumid -> /home/rtucker/dev/munin-dehumid-status/dehumidstatus
14:42<amitz>HoopyCat: ooh, I see what you mean!
14:43<amitz>thank you all :-)
14:43-!-ondrej [] has joined #linode
14:44<Jonathan>Does anyone know if rsync would be able to detect a file thats being written to and know not to try and copy it while open? or is that filesystem specific?
14:49-!-Twayne [] has quit [Ping timeout: 480 seconds]
14:50<HoopyCat>Jonathan: it'll probably copy it, as it exists at the time it starts copying it. this is good behavior for log files, but probably bad for other types of files.
14:50<rainman`>hi all
14:53-!-cpg [] has joined #linode
14:53<HoopyCat>Jonathan: workarounds: 1) if it's a file being created/"uploaded", give it a temporary filename and then rename it when it's "done", or 2) run rsync regularly, and it'll get fixed eventually :-)
14:54<Jonathan>Thanks, I was going to write it in a non-targeted dir and then copy it, but the rename idea is probably more efficient than a copy
14:57<HoopyCat>Jonathan: nod... use rename instead of copy, and even between directories on the same filesystem, it'll rock efficiently. depending on what's creating the file, you might be able to have it write to a temp file that "doesn't really exist" and then create the directory entry when it's done...
14:57-!-DephNet[Paul] [] has joined #linode
14:58<Jonathan>hmmm, not sure how to do that, it's my own code thats writing the file
14:59<HoopyCat>Jonathan: depending on the language, there's probably a library that will make a "good" temporary file
14:59<Jonathan>I'll take a look, the rename if nothing else should work nicely
15:00-!-Musicmasteria [] has joined #linode
15:01<HoopyCat>Jonathan: indeedity
15:05-!-Paul_ [] has joined #linode
15:05-!-DephNet[Paul] [] has quit [Read error: Connection reset by peer]
15:08-!-Paul_ [] has left #linode []
15:08-!-DephNet[Paul] [] has joined #linode
15:11-!-Big-Mama [] has quit [Ping timeout: 480 seconds]
15:12<TheJoe>Of all the things that could go wrong, I have no idea where a self-compiled .so should go so it can be used by MySQL
15:12<Musicmasteria>is caker here?
15:12<Musicmasteria>hello there
15:12<TheJoe>Caker is like the matrix, he's always here, he's always there.
15:13<Musicmasteria>that's good to know
15:17-!-Musicmasteria [] has left #linode []
15:24<tylerdu>!xkcd network
15:24<linbot>tylerdu: Network:
15:27<bhankins>!xkcd exploits of a mom
15:27<linbot>bhankins: Exploits of a Mom:
15:28<tylerdu>!xkcd ballmer peak
15:28<linbot>tylerdu: Ballmer Peak:
15:34<bhankins>That's a good one.
15:35<TheJoe>Seriously - where the hell is this thing supposed to go?
15:36-!-techman224 [] has joined #linode
15:36<HoopyCat>TheJoe: i'm assuming that, whatever it is, it doesn't have documentation saying where to put it
15:37-!-bogdanbiv [~quassel@] has joined #linode
15:37<TheJoe>HoopyCat: It does, but it's BSD specific. And that's just not doing Debian any good.
15:37<HoopyCat>TheJoe: i don't seem to have an obvious mysql-specific place in which to put shared libs on my system
15:37<TheJoe>Give me one of them
15:38<Jonathan>I think theres a udf folder by default
15:38-!-schmidp [] has joined #linode
15:38<TheJoe>Yeah I have no idea *where*
15:40<TheJoe>So yeah
15:40<TheJoe>A clear, solid solution would be helpful.
15:40<Jonathan>no default, you define your own using LD_LIBRARY_PATH= I assume in my.cnf
15:40<schmidp>how long does it usually take to verify an account after signup?
15:41<chesty><-- this long -->
15:41<schmidp>I'm having problems with my slicehost vm, so I thought I quickly signup for a linode
15:42<jtsage>schmidp- anywhere between instant and maybe a few hours.
15:42<praetorian>chesty: that's what she said.
15:42<schmidp>a few hours, uh thats bad
15:42<jtsage>often much, much closer to the former
15:42<TheJoe>schmidp: It's 9 times out of 10 instant
15:42<TheJoe>So you'll get it instantly.
15:43<@caker>plus I'm around (and others, too) if it falls into the manual review pile
15:43<jtsage>few hours was accounting for the possiblilty of things like hurricanes, power outages, and the rapture
15:43<praetorian>I'm asquare
15:43<schmidp>because I currently cannot upload our data, to the us (we have vm at slicehost to running a openvpn)
15:44<schmidp>but the i only get 20kb/s
15:44<schmidp>I hope linode is faster
15:45<HoopyCat>schmidp: it ultimately depends on where and what the problem is
15:45<TheJoe>Jonathan: Whereabouts does said LD_LIBRARY go?
15:46<Jonathan>I think I messed that up TheJoe - you can pass --plugin_dir to mysql_safe
15:46<Jonathan>Theres some details here -
15:46-!-adnc_ [] has quit [Ping timeout: 480 seconds]
15:46<schmidp>HoopyCat: I guess there is a network bottleneck somewhere on my way from europe to the us
15:47<schmidp>atlanta gives me 900kb/s
15:47<schmidp>all other 3 datacenters between 20-60kb/s
15:47<schmidp>(tried the files from this link:
15:52<HoopyCat>schmidp: generally speaking, newark is "best" from eu (being geographically closer), so if that's slow too...
15:53<schmidp>atlanta seems to be by far the fastet for me
15:53<chesty>try mtr
15:53<schmidp>fremont, atlanta and newark all use different routes over the atlantic
15:53-!-Bohemian [] has joined #linode
15:55-!-lesouvage [~chatzilla@] has quit [Ping timeout: 480 seconds]
15:55<schmidp>fremont: austria -> germany -> (netherlands) -> new york city -> ....
15:56<HoopyCat>schmidp: is the view from newark and atlanta
15:56<schmidp>atlanta goes: austria _> germany -> london -> atlanta
15:59-!-mysty [] has quit [Quit: HydraIRC -> <- Nine out of ten l33t h4x0rz prefer it]
15:59-!-adnc_ [] has joined #linode
16:02<schmidp>any idea why i'm getting so inconsistant transferspeeds when doing: wget
16:02<schmidp>once I get 30kb/s stable, after I restart wget I get 800kb/s stable
16:02<schmidp>after another restart 80kb/s stable
16:02<schmidp>then 1mb/s stable
16:02<schmidp>why would that be?
16:06<bhankins>Did you pause all your torrents?
16:06<HoopyCat>schmidp: lay down a ping for a minute or two and see how the packet loss and jitter (variance in round trip time) look. if there's congestion or packet loss at the beginning of a TCP session, it will err to the side of caution and take a long time to pick speed up
16:08-!-lesouvage [~chatzilla@] has joined #linode
16:12-!-shakr [] has joined #linode
16:18-!-schmidp [] has quit [Ping timeout: 480 seconds]
16:18-!-schmidp [~Adium@] has joined #linode
16:27-!-TheFirst [] has joined #linode
16:30<linbot>New news from forums: Automatic backup for CA datacenter? in Feature Request/Bug Report <>
16:30-!-syntaxman [~wade@] has joined #linode
16:35-!-moonaysl [] has joined #linode
16:39-!-Jonathan2 [] has joined #linode
16:39-!-Jonathan [] has quit [Ping timeout: 480 seconds]
16:41-!-moonaysl [] has quit [Remote host closed the connection]
16:42<TheJoe>Don't suppose anybody knows a good registrar?
16:43<TheJoe>Supposed to be taking over this website but the previous owner has been spending £30 a year on the domain. £30 is insane.
16:44<MJCS>anyone familiar with webdav?
16:45<bhankins>I have some domains there and I really like them.
16:45<bhankins>And it seems they can do
16:45<metap|pe>MJCS: somewhat. set up a few locations to know it's a flakey ass protocol
16:46<laser`>TheJoe: can
16:46<laser`>It's who I'm with
16:46<Jonathan2> is the cheapest I found for uk domains
16:46<TheJoe>laser`: I'll take a look, thanks
16:47-!-bogdanbiv [~quassel@] has quit [Remote host closed the connection]
16:50-!-aaronpk_ [] has quit [Quit: aaronpk_]
17:06-!-Turl [] has quit [Quit: Leaving.]
17:09-!-Smark[Gone] is now known as Smark
17:09-!-dajhorn [] has joined #linode
17:12-!-lesouvage [~chatzilla@] has quit [Ping timeout: 480 seconds]
17:21-!-elky [] has quit [Ping timeout: 480 seconds]
17:21-!-Bohemian [] has quit [Ping timeout: 480 seconds]
17:27-!-Musicmasteria [] has joined #linode
17:27-!-irgeek [] has quit [Remote host closed the connection]
17:27-!-irgeek [] has joined #linode
17:27-!-mode/#linode [+o irgeek] by ChanServ
17:27-!-schmidp [~Adium@] has left #linode []
17:28<Musicmasteria>where can I find someone to setup my server for me (my way is really messed up atm)
17:28<straterr1>err...maybe craigslist
17:28<straterr1>or put an ad in your local paper
17:30<HoopyCat>or, if it's a linode, post in the linode forums with what you're after and include some contact information so folks can contact you
17:31<TheJoe>laser`: YEah 123-reg certainly seem perfect... Too bad the current registrar has locked access to the current control panel. :\
17:32<laser`>Ah :/
17:32-!-irgeek is now known as Guest871
17:32-!-irgeek [] has joined #linode
17:32-!-mode/#linode [+o irgeek] by ChanServ
17:32<TheJoe>The previous owner (who's tried giving me the website) let his billing details expire. Perfect.
17:35-!-elky [] has joined #linode
17:35-!-BarkerJr [~BarkerJr@] has joined #linode
17:38<TheJoe>Guess I'll just have to change it to my card and hope I don't get billed
17:40<jtsage>PHP Warning: Module 'mcrypt' already loaded in Unknown on line 0 <- anyone seen this one before? not on a linode, on an ubuntu desktop. doesn't seem to hurt anything, just makes my IDE beep needlessly
17:41<bob2>jtsage: first hit on google
17:42-!-BarkJr [] has joined #linode
17:42<jtsage>huh. that was easy. thanks. :)
17:42<TheJoe>Ok I think I just messed up the transfer.. fantastic
17:42-!-BarkJr [] has quit []
17:44-!-Musicmasteria [] has left #linode []
17:57-!-Keith-BlindUser [] has joined #linode
17:57<Keith-BlindUser>Is there anyy way to erase the job history for a Linode without removing that particular Linode? Or what. I was curious if that job history could be erased? What if you got tired of seeing old jobs.
17:58<Keith-BlindUser>in the Linode platform control pannel.
17:58<TheJoe>Covering your tracks, eh?
18:02<Keith-BlindUser>I am 22-years old, Male from Littleton, Colorado.
18:02<Keith-BlindUser>Stupid script
18:03<TheJoe>I don't know whether I should feel bad for laughing at that.
18:09<MJCS>you just have to not do the last thing I need you to do before deployment to the masses
18:09<amitz>ah, I remember the day when my nickname reflects my asl..
18:10<amitz>MJCS: it wants to make your life uh..colorful.
18:11<MJCS>language wise it is working
18:12<MJCS>all i need to do is programatically set the group permissions of 360ish folders
18:13-!-fwermus [] has joined #linode
18:14<fwermus>I need to disable remote terminal in my account
18:14<MJCS>you mean ssh?
18:14<Keith-BlindUser>Why would a user wish to redeploy a Linux distro becides the obvious compromize?
18:14<MJCS>you know hwo to use lish?
18:15<fwermus>I dont
18:15<MJCS>well if you disable ssh you will only be able to log in via lish
18:15<linbot>LISH allows you to perform certain actions without having to log into the LPM. Lish's primary function is to allow you access to your server's console, even if networking is disabled.
18:15<MJCS>you could just restrict ssh to certain IP's
18:15<vuf>fwermus: why do you need to turn it off?
18:16<fwermus>I havent read yet, but what I like to do is to disable ssh from and enable when I need it
18:16<Keith-BlindUser>I'm trying to decide on something. I can't decide between Arch/Ubuntu/Debian. I've used all three. But I can't make up my mind for a server deployment. While one might argue that Debian stable is fine, I don't think so. What is the use when you aren't up-to-date with what's out there in upstream? I was on a Gentoo Linode for a while until about 5 minutes ago.
18:16<fwermus>when I am not working with the server, I dont want ssh be on
18:16<Peng_>fwermus: Not possible.
18:17<MJCS> (Keith-BlindUser) : - Use what you are familiar with. I personally like debian as it is stable for me. What are you going to use your server for?
18:17<vuf>Keith-BlindUser: noone can help you with that
18:17<bd_>Keith-BlindUser: The advantage is you don't have to redo config files/workaround new bugs for one daemon or another every week when one upstream or another releases a new major version. Yes, it means you wait longer for features, but it also means that very little changes over a long period, which is an advantage in production environments.
18:17<BarkerJr>hmm, can you import your linode whitelist into ssh?
18:18<Keith-BlindUser>PHP, Apache, MySQL, and a few other services.
18:18<vuf>BarkerJr: I guess the linode api can be used for that
18:18<Keith-BlindUser>I've found Debian to be okay in some cases. But for particualr packages, it doesn't have them available in the repo. And I don't want to track those packages by hand.
18:18<fwermus>I cannot use whilelist because I have a dynamic ip at home
18:19<Keith-BlindUser>I.e. Latest version of UnrealIRCd, Anope services, and such are not available through the Debian/Ubuntu repos.
18:19<MJCS>fwermus: user dydns or another service like that
18:19<Peng_>fwermus: It'll still be limited to certain netblocks.
18:19<Peng_>Most of the time.
18:19-!-vuf [~am@] has quit [Quit: Ex-Chat]
18:19<Keith-BlindUser>I can track those by hand. But that adds an extra thing I have to do when a new version is released, going to each site, for Unreal, for Anope, etc. When I can just handel that by the distro packagem anager.
18:20<bd_>Keith-BlindUser: you can use debian testing. Or stick with the stable versions for a while :)
18:21<Keith-BlindUser>Yes. But taht doesn't point out what I just said. Package X you have to compile by hand. I'd rather not deal with that.
18:21<bd_>it's all a matter of if you value things not breaking on their own, or if you value latest versions of everything
18:21<bd_>The choice is yours - debian stable is for those in the former category
18:21<bd_>but you can't have both :)
18:21<Keith-BlindUser>If the package is available in the repo itself I.e. Apt-get UnrealIRCd wich does not work, etc. Or apt-get anope (likewise) not available, that's usually easier than me dealing ith the ./configure scripts and such.
18:22<bd_>oh, you mean it's not there at all?
18:22<bd_>*shrug* someone would need to add it I suppose
18:23<Keith-BlindUser>besides. It's easier handeling security updates in packages installed via the package manager of a system than dealing with software that you installed manually. An administrator of a server might forget to update an isntalled package by hand, but have other packages up-to-date, while still having a secury hole in other packages. I.e. I know of a server admin who is running UnrealIRCd 3.2.5, when 3.2.8 soon to be 3.2.9 is current.
18:24<Keith-BlindUser>Thus, by running an old version of a given software version, you have a possibility of not only having bugs for that given version, but users that wish to attack you might be able to (crash) the given service.
18:24<Keith-BlindUser>I desided to go with Arch for now. Much better use of CPU utilization under my Linode360 than the Gentoo 2008.0 installation instance I was using.
18:24-!-fwermus [] has left #linode []
18:27<@caker>just about anything that's NOT Gentoo would utilize CPU better
18:27<BarkerJr>I've been thinking about that with the forum posts about 360's not being able to handle more than 15 apache children
18:28<linbot>--zomg-funroll-loops --ricer-cflags --flail-wildly --moar-disk-thrashing
18:28-!-TheFirst [] has quit [Quit: You're a bloody puppet!]
18:28<BarkerJr>I can handle 50 apache children on my centos 360 fine
18:33<laser`>Heh, I can't on my 540
18:33<laser`>I guess it's what you're serving
18:33<@caker>depends on what you've got enabled, I'd think
18:35<chesty>!newcalc 360 / 50
18:36<linbot>chesty: 360 / 50 = 7.2
18:37-!-adnc_ [] has quit [Quit: bye]
18:38<BarkerJr>that's probably about right... my apache processes claim 11 - 4 (shared), which is 7
18:38<BarkerJr>since not all requests require everything, I guess I could swap some of that, too
18:41-!-Turl [] has joined #linode
18:45-!-fwermus [] has joined #linode
18:46<fwermus>I am trying to connect to mysql through mysql browser, but I failed. I am using port 3306 and mysql is installed through app-get. I have my pass, etc. What things could I watch first to realize what is going on?
18:46<bob2>connecting how?
18:46<bob2>if you mean "over the internet", mysql doesn't listen on external interfaces by default
18:46<bob2>(in debuntu)
18:47<fwermus>in debian?
18:48<bob2>one half of teen pop sensation 'debuntu'
18:48<@caker>netstat -pln <-- will tell you what's listening on what interface and what port
18:48<fwermus>connecting using mysql query browser from my laptop remotely to linode
18:48<bob2>you can either make it listen on eth0 (in /etc/mysql/my.cnf, bad idea) or ssh tunnel (ssh -L3306:localhost:3306, then connect to localhost:3306 in 'mysql explorer')
18:48<fwermus>if it doesnt listen to internet by default. Do you know what I have to instruct to mysql to listen it?
18:49<bob2>don't do that
18:49<bob2>it's a terrible idea
18:49<bob2>because if you ever have a weak password, your db gets owned
18:50<bob2>using ssh is less effort, but if you insist, set up a firewall, allow access to port 3306 from only IP (yours), then edit bind-interfaces in /etc/mysql/my.cnf
18:50<fwermus>this is the first time I would like to deploy. It isnt yet production
18:50<HoopyCat>security is a habit
18:50<fwermus>can I use ssh with my mysql?
18:51<fwermus>I mean, can I instruct mysql to use ssh to connect?
18:51<bob2>ssh tunnel (ssh -L3306:localhost:3306, then connect to localhost:3306 in 'mysql explorer')
18:51<fwermus>great! I will test it
18:53<TheJoe>Hm don't suppose anyone's heard what the results of the LCROSS impact were
18:53<HoopyCat>and gosh darnit, if it weren't for the code teal, zunzun wouldn'ta made the useless forum post i was going to make
18:54<HoopyCat>TheJoe: i believe the moon is still intact (which is good). haven't heard official word from NASA, but i think we'd have noticed if it were damaged significantly.
18:55<bob2>reminds me of a geeks in space episode
18:55<TheJoe>Why the hell has the English channel reached my doorstep?!
18:55<HoopyCat>TheJoe: NASA-TV's ratings for the coveted 4am-7am daypart were off the chart, too
18:55-!-litwol|mac [] has quit [Remote host closed the connection]
18:57<HoopyCat>TheJoe: actually, in all seriousness, it's one of those things where they'll be crunching data for awhile before they announce any Findings
18:57<TheJoe>HoopyCat: Oh right, just wasn't sure if there was anything released yet or not
18:58<BarkerJr>so in all thoughts here, people should be setting the max requests per child smaller
18:59<BarkerJr>that will decrease memory usage per child
18:59<BarkerJr>4000 may be too high
18:59<HoopyCat>TheJoe: it's a long weekend, too, so if they do know anything, they probably won't do much with it until tuesday when they have all hands on deck
18:59<HoopyCat>18:58 [@DandC:21] Buffalo Bills report card #ROC
18:59<BarkerJr>also, people should comment out the LoadModule commands for modules they don't use
19:00<BarkerJr>by default, every module is loaded
19:00*HoopyCat chews a handful of Tums and readies his adjective notebook
19:02<linbot>New news from forums: Use Registrar Name Server Entries in Linux Networking <>
19:05<BarkerJr>is the blog down?
19:05<bd_>looks like it
19:09<linbot>New news from forums: Settings for high performance Apache, MySQL and PHP in Linux, Apache, Mysql and PHP (LAMP) Forum <>
19:09<linbot>SelfishMan: Atlanta360 - 68, Atlanta540 - 27, Atlanta720 - 23, Atlanta1080 - 30, Atlanta1440 - 22, Atlanta2880 - 2, Atlanta5760 - 2, Atlanta8640 - 1, Atlanta11520 - 1, Atlanta14400 - 1
19:09*SelfishMan adds a node
19:10<BarkerJr>a 14400?
19:10<TheJoe>Oh how I long for the day when I can willy-nilly say "I'm adding a new node".
19:10<HoopyCat>already talked my wife into upgrading to ubuntu 9.10; i'm now trying to talk her into trying out with the bills for middle linebacker
19:10<TheJoe>I wish I had money
19:11<TheJoe>Isn't 9.10 only in beta?
19:12<TheJoe>HoopyCat: I always imagined you as a Debian uber Ubuntu guy.
19:12<BarkerJr>linode has a hundred dollars of mine I gotta burn at some point
19:12<HoopyCat>she's smart, has some muscles, and looks convincing in drag. i bet she could make it at least a game and a half before a season-ending injury.
19:12<bob2>BarkerJr: linode 2880
19:12<HoopyCat>(which is about par for a buffalo linebacker)
19:13<HoopyCat>TheJoe: it is indeed only in beta. that's why i had to talk her into it, and why i haven't upgraded just yet
19:13<TheJoe>Oh right
19:13<TheJoe>Damn I don't think I've actually used Linux as a desktop at all since 9.04
19:14<bob2>so, what do people use for keeping track of (server) bandwidth use these days?
19:14<TheJoe>bob2: My 'node control panel.
19:14<TheJoe>I have some really weird bandwidth usage right now... incoming usage that shouldn't be there.
19:14<HoopyCat>TheJoe: i like the ubuntu release schedule. it's nice to have a twice-yearly refresh of my workstation, while still having my servers on a two-year cycle and using the same distro on both
19:15<HoopyCat>bob2: i generally don't. my backups make graphing pretty much useless.
19:15<bob2>TheJoe: * outside of linode
19:15-!-j2d2 [] has joined #linode
19:15<HoopyCat>bob2: no, really:
19:15-!-CWii_ [] has joined #linode
19:15<BarkerJr>TheJoe: give me your IP and I'll show you incoming traffic :)
19:16<TheJoe>BarkerJr: No thanks. I officially have paying customers now!
19:16<j2d2>If I have a 540 account and I'd like to setup a second OS using smoe of the resources I can put aside, how do I do it? Can I have a second IP?
19:16<HoopyCat>bob2: bloody useless. :-)
19:16<bob2>j2d2: you can't split it into two and boot both of them
19:16<bob2>(at the same time)
19:16<j2d2>oh i see
19:16<HoopyCat>j2d2: you can deploy another distro using unused disk space, and get a second IP address, but it's like a dual-boot PC: you can only run one at a time
19:17<TheJoe>j2d2: You can have a second disk image for another OS, and a new IP is an extra dollar a month. You can't have have both OS's running at the same time.
19:17<bob2>j2d2: you can install another OS in another partition, and you can get a second IP, but you can't get two running linodes for the price of one
19:17<BarkerJr>talk about support!
19:17<j2d2>dang. y'all are good.
19:17<j2d2>so is the right approach to just get another system then?
19:17<bob2>HoopyCat: non-incremtnal, every day?
19:18<BarkerJr>you can downdrage your existing node to 360 if you don't need 540
19:18<TheJoe>Also downgrade
19:18<j2d2>what's the best way to compare resource use compared to alloted?
19:18<HoopyCat>j2d2: yup. adds/deletes/changes are pro-rated to the day, so if you just need a second one occasionally, you can stash its stuff on your full-time linode, add a second one, clone the images over, do what you need to do, then clone the images back and delete the linode when you're done
19:19<BarkerJr>well, bandwidth and disk space you can tell in the manager... ram is best determined by the 'free' command
19:19<TheJoe>Specifically - 'free -m'
19:20<TheJoe>Oh boy. Only 201MB free here
19:20<j2d2>ram can be tricky cuz it'll use what's there, most of the time.
19:20<HoopyCat>bob2: incremental twice daily (three levels), full weekly. there's a decent amount of database confrappuchino that i haven't cared about enough to optimize quite yet
19:20<TheJoe>There's definitely a memory leak in lighttpd...
19:20<TheJoe>Just restarted it and was greeted with an extra 50MB free :\
19:20-!-JDLSpeedy [] has quit [Quit: Ex-Chat]
19:20<BarkerJr>with 'free', you want to read the second line, not the first
19:21<HoopyCat>TheJoe: what did ps auxwww say before you restarted lighttpd, and are you doing anything involving spawning php processes or anything?
19:21-!-CWii [] has quit [Read error: Operation timed out]
19:21-!-Guspaz|m [] has quit [Quit: ajax IRC Client]
19:21<j2d2>BarkerJr: can you elaborate?
19:21<TheJoe>Err.... 2 wordpress sites....
19:21<BarkerJr>the cache and buffers are expendable
19:21<chesty>TheJoe: that's normal, it happens with all software
19:21<BarkerJr>so you want to add them to the free ram
19:21<TheJoe>But this memory usage goes up quite a bit over time
19:22<TheJoe>It's not much of a concern, yet.
19:22<HoopyCat>TheJoe: are you running spawn-fcgi on its own, or is lighttpd spawning the php handlers for you? if the latter, it probably just dumped a shitload of caches within php
19:22<j2d2>BarkerJr: So, first line - second line = actual free requirements?
19:22<TheJoe>HoopyCat: I believe spawn-fcgi is doing it
19:22<TheJoe>But I set it up so long ago I can't remember heh
19:22<BarkerJr>no, second line is the actual free requirements
19:22<HoopyCat>TheJoe: ps auxwww | grep php
19:23<BarkerJr>second line = free(3) + buffers(5) + cache(6)
19:23<HoopyCat>TheJoe: if the time in the START column is from about when you restarted lighttpd, there you go :-)
19:23<j2d2>BarkerJr: oh i see. second line is doing the math to handle expendibles. ok
19:23<TheJoe>And 'supervise'
19:23<BarkerJr>yep :)
19:23<TheJoe>What the hell is supervise
19:24<BarkerJr>j2d2: the way linux works is it will fill all free memory with file caching
19:24<BarkerJr>so, the more free ram you have, the faster your system will be
19:24<TheJoe>The default output of 'whatis' is amazing
19:24<j2d2>BarkerJr: right
19:24<HoopyCat>rtucker@framboise:~$ whatis love
19:24<j2d2>BarkerJr: i understand the way, but not so much the tools. been like 8 years since i was admining in any seriousness.
19:24<HoopyCat>love: baby don't hurt me, don't hurt me, no moar
19:24<BarkerJr>k :)
19:25<j2d2>anyway, i just resize my linode 540 to a 360 and start a new 360 account and i'm basically all good for $10 more, right?
19:25<TheJoe>joe@ranger:~$ whatis love
19:25<TheJoe>love: nothing appropriate.
19:25<TheJoe>aww :(
19:26<HoopyCat>j2d2: yup... the actual amount charged to your card right now will not be exactly $10, but in the long run, you'll be paying ~$40/mo instead of ~$30/mo
19:26<HoopyCat>there's calculus involved
19:28<j2d2>just to be sure, setting the plan to 360 requires no setup on my end, right? it just does it's thing. 12 minutes later it's back up exactly as it was, just with less resources?
19:28<j2d2>no ip's change or anything?
19:29<HoopyCat>j2d2: assuming you have less than 16GB of images, yep. (if you have more than 16GB, i believe it will make you resize down before doing anything)
19:29<BarkerJr>images of what are we talking about here? :)
19:30<j2d2>HoopyCat: 15360mb :)
19:30<j2d2>i shrunk it to that size no problem
19:30<HoopyCat>j2d2: and yup, IPs change the same. (the hostname to which you ssh for the emergency console (lish) changes too, so if you have it bookmarked or written down somewhere (which you should!), you'll want to update that afterwards too)
19:31<j2d2>HoopyCat: say that again..? "change the same"?
19:31<HoopyCat>j2d2: remain the same
19:31<HoopyCat>i write good
19:31<j2d2>i ain't need no english
19:32<HoopyCat>you shoulda been here the day i said "all things being equal"... i've got a post-it note on my monitor saying "all -->OTHER<-- things being equal" now :-/
19:32<j2d2>is it hard to upgrade a box remotely? i've never done it from one os release to a newer one
19:33<HoopyCat>j2d2: depends on the distro. with ubuntu, it's pretty easy... i think it's one command and a reboot. (note: you can rent some extra disk space for awhile and duplicate your image into it before doing major changes. very good idea.)
19:33-!-TheJoe is now known as TheJoe|ZzZz
19:36<j2d2>cool. good idea
19:36<j2d2>once i hear back from my users i'll consider these things
19:36<j2d2>thanks for all the help. to anyone who works for linode, i've been a happy customer for three years.
19:37<j2d2>i'm out for a while
19:37-!-j2d2 [] has quit [Quit: later guys!]
19:38<HoopyCat>bob2: just told the backup server to skip /var/lib/postgresql/8.3/main/base ... i dump the databases out nightly, so backing up the binary db files on a running database engine just makes rsync waste time
19:39<HoopyCat>that's 1.3GB or something right there
19:40-!-dc0e [] has joined #linode
19:40-!-dc0e [] has quit []
19:43-!-j2d2 [] has joined #linode
19:43<j2d2>back again!
19:43<j2d2>what method do you guys like best for hosting django projects?
19:44<j2d2>i've recently been using apache + mod_wsgi and like it. are there better methods?
19:44<bob2>all the cool kids use nginx
19:44<j2d2>i wanna be a cool kid
19:45<j2d2>nginx + wsgi + memcached
19:45<HoopyCat>that wasn't a very long while ;-)
19:46<HoopyCat>j2d2: i don't use django, but i do have a python-based application of considerable frameworkness that is running lighttpd and wsgi, with memcached of course
19:46<j2d2>why lighttpd?
19:46<j2d2>also, which framework?
19:47<HoopyCat>j2d2: because apache was murder and nginx wasn't around at the time and/or required too much russian
19:47<HoopyCat>j2d2: a wiki called sycamore
19:48<j2d2>does it handle mediawiki's markup?
19:49<j2d2>oh neat. it's hosted on github
19:51<j2d2>thoughts on cherrypy?
19:51-!-Guspaz|m [] has joined #linode
19:51<@caker>put a smile on your face 10 miles wide
19:53<j2d2>why would people use nginx and cherrpy together? seems like they're botg standalones
19:53<BarkerJr>cherry pie? :)
19:54<bob2>because nginx is almost certainly faster for static content
19:54<metap|pe>mmmm, pie
19:54<j2d2>oh i see
19:55<j2d2>is it correct to say you'd use cherrypy or wsgi then?
19:55<j2d2>both with nginx?
19:58<HoopyCat>j2d2: it uses different markup, kinda moinmoin-style.
20:01<BarkerJr>dis cherry py is moin-moin!
20:01-!-supine [] has joined #linode
20:01-!-TheFirst [] has joined #linode
20:01<BarkerJr>allz moin-moin!
20:02<j2d2>thanks again fro the help guys.
20:02-!-fwermus [] has quit [Remote host closed the connection]
20:03<HoopyCat>j2d2: np
20:03<HoopyCat>linode needs product names more along the lines of automotive tires
20:04<HoopyCat>Mastercraft GLACIER GRIP II
20:04<chesty>what does COA mean when talking about XP license
20:05<HoopyCat>chesty: certificate of authenticity, if i recall correctly
20:05<chesty>ah thanks
20:08-!-tomaw is now known as Guest879
20:08-!-Guest879 [] has quit [Remote host closed the connection]
20:08-!-ph_ [~ph^] has quit [Remote host closed the connection]
20:09-!-tomaw [] has joined #linode
20:10-!-Keith-BlindUser [] has quit [Quit: Freedom Chat - Your Home Away From Home | | tIRC script by the Freedom Chat Leets]
20:10-!-j2d2 [] has quit [Ping timeout: 480 seconds]
20:11<Smark>how does one go about configuring MySQL to store its databases in memory, and access them from there?
20:12<bob2>use the memory engine type
20:12<bob2>(in create table)
20:12<BarkerJr>well, you can configure query cache
20:13<laser`>I think the query cache can probably accomplish what you want
20:13<HoopyCat>Smark: create table blah (poop int, flollop bool) engine=memory;
20:13<bob2>lots of things are better than mysql for some in memoery stuff thought
20:13<bob2>er though
20:14<BarkerJr>you have to keep in mind that linix will cache the table files in memory automatically
20:14<HoopyCat>you can also "create table blahtemp engine=memory select * from blah" or something like that to copy the entirety of blah into memory
20:14<Smark>im looking to all but eliminate disk IO for MySQL accesses
20:14<bob2>and to lose all your data on reboot?
20:15<Smark>is it possible to have MySQL store all its data in memory, then send syncs down to the disk every x minutes?
20:15<Smark>or atleast eliminate them during accesses
20:15<Smark>writes to the db wont happen often
20:15<BarkerJr>use INSERT...DELAYED instead of just INSERT to minimize disk access
20:16<BarkerJr>REPLACE...DELAYED instead of UPDATE
20:16<Smark>whats the mysql query cache do?
20:16<HoopyCat>Smark: "create table blah engine=not_memory select * from blahtemp" :-)
20:16<BarkerJr>stores the results of the query in memory then returns that next identical query
20:16<laser`>Smark: caches the result of recent queries
20:17<laser`>If you've got an app that does the same sort of queries over and over again it's good
20:17<BarkerJr>then you can hint at it on specific queries with SQL_CACHE or SQL_NOCACHE
20:18<Smark>how does one set that up? is it as simple as apt-getting a package then enabling it for a table/db?
20:18<BarkerJr>it's just a few settings in /etc/my.cnf... I forget what they are, though :)
20:19<Smark>i can google it. Thanks for all your help guys
20:19<BarkerJr>well, let me know... I want to enable it :)
20:23<BarkerJr>I love how tony soprano is so screwed up that even his shrink has a shrink
20:24<Smark>(what im going to be trying)
20:24-!-laser` [] has quit [Read error: Connection reset by peer]
20:25<BarkerJr>thx, looks right
20:28-!-JDLSpeedy [] has joined #linode
20:29-!-jtaji [] has quit [Remote host closed the connection]
20:30<BarkerJr>I'm thinking we want to set query_cache_size a little lower :)
20:30<Smark>what do you mean?
20:30<linbot>encode: Fremont360 - 24, Fremont540 - 46, Fremont720 - 23, Fremont1080 - 19, Fremont1440 - 14, Fremont2880 - 2, Fremont5760 - 2, Fremont8640 - 1, Fremont11520 - 1, Fremont14400 - 1
20:30<Smark>lower than 16mb?
20:31<BarkerJr>oh, the setting down below is 256MB :)
20:31<Smark>ah, yes
20:34-!-ondrej [] has quit [Read error: Operation timed out]
20:34<BarkerJr>then it says it's in KB :)
20:57-!-JoeK [] has quit [Remote host closed the connection]
20:58-!-JoeK [] has joined #linode
21:12-!-inc_ [] has joined #linode
21:13<inc_>any server admins active?
21:16<linbot>Don't ask to ask; just ASK
21:16<linbot>The staff may or may not be around but if you tell us your problem then someone in here may be able to help
21:16<inc_>i cant find anywhere how many ip's you get with the cheapest vps package
21:17<bd_>!skynet ips
21:17<linbot>bd_: Can I purchase additional IPs? Yes. All new accounts include one IP and are permitted to add an additional IP via the Linode Manager for $1.00 per month. If you need more than two IPs, you can open a support ticket and provide proper justification (we abide by ARIN's regulations). (70.711%)
21:17<inc_>do they support bsd?
21:20<BarkerJr>Smark: I got Qcache_hits 402
21:20<bd_>note: read the warning at the top of that howto. twice.
21:22<Alan>gah, i hate being up late.... everybody in my country is asleep :(
21:23<inc_>how long till my account is activated/do they take google checkout
21:23<bob2>not long unless your signup gets flagged, no
21:24<inc_>how much ram on the $20 a mo one
21:24<bd_>!skynet forms of payment
21:24<linbot>bd_: What forms of payment do you accept? We accept Visa, MasterCard, American Express, and Discover. We have month-to-month, 12-month, and 24-month terms available. Paypal transfers are not supported but Paypal debit/credit cards should work.
21:25<bd_>inc_: the RAM's right there on ...
21:35-!-walterheck [~walterhec@] has quit [Ping timeout: 480 seconds]
21:38-!-inc_ [] has quit [Quit: leaving]
21:38<Alan>bd_: good warning!
21:49<encode>where are the linode .config files found these days?
21:50-!-Guspaz|m [] has quit [Quit: ajax IRC Client]
21:50<encode>oh nevermind
21:50<encode>brain fart
21:52<HoopyCat>how much memory do i get with the $20 linode 360 plan, and how much does it cost?
21:52<HoopyCat>oh wait, *checks scrollback*
21:52<encode>HoopyCat: you making fun of me?
21:53<HoopyCat>encode: i'm serious as cancer when i say rhythm is a dancer
21:54<Peng_>OK, so there's 360 MB of memory, but how much does it cost? :D
21:55<HoopyCat>i can't find the button to pay with beenz
22:00<bhankins>What year did Linode start its hosting business?
22:00<bd_> Record created on 26-Jan-2003.
22:00<bd_>^^^ whois
22:01<encode>HoopyCat: but, you've been around long enough to know the plan name = amount of memory
22:01<HoopyCat>bhankins: since you specifically worded it like that, i'm going to say 1998.
22:02<HoopyCat>encode: i wasn't mocking you, i was mocking someone a half-hour before you. :-)
22:02<bhankins>ahh ok. I had an account previously back in 2005. Just wondering if it was near the beginning.
22:05-!-Bohemian [] has joined #linode
22:05-!-Bohemian [] has left #linode []
22:05-!-walterheck [~walterhec@] has joined #linode
22:07<HoopyCat>The InterNIC charges $70 to register a new domain name. This fee covers the first two years. They currently bill at a rate of $35 per year. The InterNIC will send you an invoice via e-mail shortly after the domain is registered.
22:07<HoopyCat>ah, the olden days, when domains cost about as much as a used car
22:12-!-ondrej [] has joined #linode
22:12<A-KO>any reason why apache would try and download a php file rather than execute it? I've got libapache2-mod-php5 loaded...
22:13<A-KO>php5 is installed..
22:15<metap|pe>that usually happens when apache doesn't get the message to actually enable mod-php for php files.
22:15<metap|pe>older debian and ubuntu used to do this a lot. very annoying.
22:15<A-KO>any way to fix without blowing it away?
22:16<metap|pe>yeah, you need some config directives for apache
22:16<A-KO>well it's including the files...
22:16<metap|pe>of course, it could be any one of a number of things
22:16<MJCS>Its fun when it takes 24hours to write 30 lines of code
22:17<MJCS>ah fuck it didnt work
22:18<Alan>MJCS: even better!
22:19<Alan>MJCS: I personally like days where the total line count goes down but functionality is added
22:20<metap|pe>A-KO: could also be from mime non having php and/or apache not using the mime db
22:26-!-zack [~zack@] has joined #linode
22:27-!-TheFirst [] has quit [Ping timeout: 480 seconds]
22:28-!-ondrej [] has quit [Ping timeout: 480 seconds]
22:28-!-zack_ [] has joined #linode
22:32<A-KO>that is annoying
22:32<A-KO>dropping ubuntu for a hosting server now
22:33<A-KO>not sure wtf exactly happened but it's being retarded so :P
22:33<Alan>Ugh, sleep time
22:33<Alan>Mon Oct 12 03:34:57 BST 2009
22:33<Alan>got work ... well, today
22:33<metap|pe>try: grep php /etc/mime.types
22:33<Alan>Luckily, I've got work at whatever time i choose to turn up
22:34<Smark>having some issues with php-cgi eating up huge amounts of CPU. Anyone have any ideas? Using lighttpd with php5-cgi running Wordpress.
22:35<checkers>stupid PHP code? just guessing here
22:35-!-zack_ [] has quit [Read error: Connection reset by peer]
22:35<Smark>most likley
22:36<A-KO>it's there metap|pe
22:36*checkers suggests disabling plugins until the stupid PHP code stops running then
22:36-!-zack [~zack@] has quit [Ping timeout: 480 seconds]
22:37-!-ondrej [] has joined #linode
22:39-!-bhankins [] has quit [Quit: Leaving]
22:40-!-bhankins [] has joined #linode
22:49<walterheck>A-KO: check for an AddHandler yadda php in your apache configs
22:50<walterheck>e.g. if you use mod_fcgid it'd be: AddHandler fcgid-script php php5 fcg fcgi fpl
22:50<Smark>I'm really at a loss here, I'm setting up a site that gets a LOT of traffic. Running a Linode 1080, I've set up MySQL to cache queries. php-cgi just hogs all the CPU. Its a wordpress site running quite a few plugins, but i cant see hows its eating up this much CPU
22:52<HoopyCat>Smark: if php-cgi is hogging the CPU, it's probably not related to mysql queries. are you using some sort of opcode cacher for php (i believe i use xcache)? how about something that caches outgoing pages (my blog software has this built-in; got me from ~4 pages/second to ~80 pages/second)?
22:53-!-dajhorn [] has quit [Quit: ChatZilla 0.9.85 [Firefox 3.5.4/20091006224018]]
22:53<Smark>HoopyCat, i was using apc, then install xcache and it got worse, was in the process of moving back to apc
22:54<Smark>there a special reason you use xcache?
22:54-!-ondrej [] has quit [Ping timeout: 480 seconds]
22:55<HoopyCat>Smark: i could install it with "apt-get install php5-xcache" :-)
22:56<Smark>i installed apc with "apt-get install php-apc"
22:57-!-Berto [] has joined #linode
22:58<HoopyCat>Smark: ubuntu 8.04 doesn't seem to have it, but that's neither here nor there. does your application support any sort of caching of output? it sounds like it's just sitting there generating the same pages over and over again for visitors
23:02-!-Berto [] has quit [Quit: Leaving]
23:02<HoopyCat>Smark: (i'm also assuming that, by "a lot of traffic", you mean "more than a few pageviews per second"... if it's less than a dozen a minute, there's probably just something sitting and spinning for no good reason)
23:03<Smark>looking at one view every 1.3 seconds or so
23:04-!-walterheck [~walterhec@] has quit [Ping timeout: 480 seconds]
23:04<HoopyCat>Smark: yeah, that's a lot of traffic
23:05-!-Shinsaku [] has joined #linode
23:07<HoopyCat>Smark: either use less CPU per page generation (fewer plugins) or generate fewer pages (output caching)
23:08-!-TheFirst [] has joined #linode
23:08-!-walterheck [~walterhec@] has joined #linode
23:08<HoopyCat>(or both)
23:09<HoopyCat>i'm heading off to bed... if anyone needs me, and click on "random card"
23:15*amitz should take the time to install an LCD monitor above his bed..
23:19-!-jcn [] has joined #linode
23:20<jtsage>amitz- i'd consider killing a guy for that rig denzel washington had in the bone collector.... lots of computer, no need to get up :)
23:32-!-_banana [] has joined #linode
23:34<JoeK>anybody good with php who would help me make something? D:
23:34<jtsage>depends on what you're trying to do :)
23:38-!-arooni [] has quit [Remote host closed the connection]
23:41-!-arooni [] has joined #linode
23:42<JoeK>jtatum: ping servers and return if they are online or not
23:43-!-techman224 [] has quit [Read error: Connection reset by peer]
23:43-!-techman224 [] has joined #linode
23:43<jtsage>ah. hmm. that's not all that hard to do really. command line or web based?
23:44<JoeK>web based
23:45<jtsage>if it were me, i'd just exec the system ping command and check the exit code. it's probably the simplest
23:46<metap|pe>icmp also requires root (suid). so something in php to do actual pings is probably right out
23:49-!-jtaji [] has joined #linode
23:50<amitz>jtsage: or if taken to the extreme, matrix style connectivity. I don't even have to eat/drink/poop/pee :-p
23:51<jtsage>ya know, they never do bother to cover how that all works once they hook up in the ship. wallowing in your own filth and reddish goo, sure - but that ship has got to smell horrendous.
23:51-!-orudie [] has joined #linode
23:51<orudie>is there such thing as clock background ?
23:52<orudie>for windows 7
23:52<reillyeon>orudie: offtopic?
23:53<orudie>reillyeon, its not freenode shush
23:53<jtsage>orudie- does that mean they shitcanned activedesktop completely finally?
23:54<orudie>no i just want a nick clock
23:54<orudie>on the 52" LCD
23:55<encode>yeah, but coming into #Linode and asking that sort of question straight up is a bit odd
23:55<encode>especially given this is clearly a linux related channel
23:56<orudie>encode, did I hurt your feelings ?
23:56<orudie>it was dead in here anyway
23:57<encode>dead? you asked less than a minute after joining
23:57<orudie>its not like I was jumping in while someone was being helped
23:57<amitz>jtsage: I'd argue that they feed you with 100% digestible nutrition. No feces and close to pure urine.
23:57<encode>i'm just saying, don't expect a reasonable answer to your question in here
23:57<jtsage>amitz- fair enough. heh.
23:58-!-supine [] has quit [Quit: Leaving.]
23:58<metap|pe>so they feed you twinkies?
23:58<jtsage>gah. no food talk. i need to go to bed soon, not make a snack :P
23:59-!-VS_ChanLog [] has left #linode [Rotating Logs]
23:59-!-VS_ChanLog [] has joined #linode
---Logclosed Mon Oct 12 00:00:01 2009