--- Day changed --- Log opened Mon Sep 28 23:59:02 2009 00:03 < mike__> whats wrong with my syntax? "rsync -aHSKDvz /mnt/xvda1 -e ssh / root@ip.here:/mnt/xvdb/" 00:04 < mike__> it tries to copy the entire box over instead of just /mnt/xvda1 00:05 < Pryon> is that a stray '/' in there? 00:05 < Pryon> Also, why not stick the -e switch in the front. It's a little weird there 00:05 < Pryon> (IMHO) 00:08 < mike__> Pryon: thanks, it was the stray / 00:08 -!- Bass10 [~Bass10@c-76-113-194-7.hsd1.mn.comcast.net] has quit [Ping timeout: 480 seconds] 00:10 -!- dassouki [~dassouki@fctnnbsc15w-156034092086.pppoe-dynamic.nb.aliant.net] has joined #linode 00:10 < dassouki> i'm starting up a server for opensource development running linux 00:10 < dassouki> i need to setup an admin guy other than me, but i don't wnat to give him the same privelage 00:10 < dassouki> however i'd like him to be able to do stuff like setup an irc bot 00:11 < dassouki> or setup user accounts 00:11 < dassouki> db accounts/connection and stuff like that 00:11 < dassouki> i'd also like to monitor hi activities since i'm not sure i trust the guy 00:12 < bd_> sudo can be configured to restrict the commands he runs, and to log them, but in general, if you don't trust the person _don't give them root_ 00:12 < bd_> too many ways to escape a not-fully-thought-out sudo config :) 00:12 < metap|pe> if you don't trust them, don't give them local access 00:12 < bd_> dassouki: is this a box on Linode? or a physical box? 00:12 < dassouki> box on linoe 00:12 < bd_> ah okay 00:15 < dassouki> intereting. 00:16 < dassouki> so how can i setup my own ircbot i guess? any ideas on where i cna start reading 00:16 -!- pror [~4a48e4bb@webuser.linode.com] has joined #linode 00:16 < bob2> which ircbot? 00:16 < bob2> alternateively, what is your goal 00:17 < dassouki> i wnat to setup an ircboat for a channel to log messages and to setup commands such as ??function 00:17 < dassouki> but the ultimate goal is to setup a server admin on my server that will not use the server to his personal needs 00:17 < Pryon> supybot's popular and probably comes with a logger out of the box 00:17 -!- pror [~4a48e4bb@webuser.linode.com] has left #linode [] 00:18 -!- ondrej [~ondrej@24-176-186-44.static.reno.nv.charter.com] has joined #linode 00:21 < Pryon> Rather than try to cage somebody you can't trust, why not work with somebody you *can* trust. 00:21 -!- blognewb [~User82934@70.134.68.122] has joined #linode 00:23 < dassouki> indeed it's a start up project and a few ppl have asked to volunteer 00:25 -!- panpainter [~panpainte@c-66-235-62-16.sea.wa.customer.broadstripe.net] has quit [Quit: panpainter] 00:25 < dassouki> any good reads about setting up users on an ubuntu linode 00:25 -!- panpainter [~panpainte@c-66-235-62-16.sea.wa.customer.broadstripe.net] has joined #linode 00:35 -!- User23423 [~User82934@70.134.68.122] has joined #linode 00:40 -!- User23423 [~User82934@70.134.68.122] has quit [] 00:40 -!- User12482 [~User82934@70.134.68.122] has joined #linode 00:42 -!- blognewb [~User82934@70.134.68.122] has quit [Ping timeout: 480 seconds] 00:50 < amitz> dassouki: library.linode.com ? 00:52 < dassouki> cool 00:52 < dassouki> i'll check it out 00:54 < amitz> dassouki: The more sophisticate she is, the more difficult to do it properly. In general, just give her a normal account, no sudo. You do all sudo stuff. 00:54 < dassouki> coool 00:55 -!- gicque [~gicque@S01060014d1426beb.wp.shawcable.net] has joined #linode 00:55 < dassouki> umm that' probably gonna be the best solution 00:55 < dassouki> give them admin support for svn and postgis 00:55 < gicque> hi 00:55 < dassouki> and have a way to autobackup 00:56 < gicque> does anyone know which package contains "which" in centos? and "vi" as well. i have never been on a wunix system without these commands so no clue how to install them 00:57 < amitz> in what way you don't trust this guy? 00:57 < metap|pe> i've run into that too (centos which) 00:57 < gicque> also, my previous image was also centos and i dont recall install those. why are they not in this new image i deployed today? - very frustrating and time consuming for simple things 00:57 < dassouki> amitz: what if they upload illegal material and distribute it ? 00:57 < metap|pe> which is officially deprecated 00:58 < amitz> gicque: try searching in google, there should be a centralized info on this, maybe a direction to do it locally? 00:58 < Smark> Ok, I have no idea how you get https working. I assume with https you need a ssl cert right? 00:59 -!- Andrew_ [~Andrew@c122-107-193-19.eburwd5.vic.optusnet.com.au] has quit [Read error: Operation timed out] 01:00 < gicque> try searching google for "which"??? 01:00 < metap|pe> i show a which package for centos5 01:00 < dassouki> that's like trying to search google for R tutorials 01:00 < gicque> dont you think i tried that. which is a common word 01:00 < amitz> gicque: "how to find out which package to install if you want file X". 01:01 < amitz> that's the rough google phrase I meant. 01:01 < metap|pe> rpm -qf /usr/bin/which 01:01 < metap|pe> which-2.16-7 01:02 < amitz> dassouki: usually you can only deal with it after the fact. You can't monitor him 24 hours and immediately block his illegal material within seconds. 01:02 < dassouki> cool 01:03 < dassouki> at the end of the day i'm a transportation engineer. I've been running linux since redhat 7, but haven't ever had the need to do any server stuff 01:03 < metap|pe> dassouki: can this person afford $20 a month for a 360 node? 01:03 < dassouki> thi to me is a whole new ball field 01:03 < amitz> dassouki: anyway, that will be a hosting problem.Concerning whether you can be blamed or not, that will also be the responsibility related to hosting problem. 01:03 < dassouki> metap|pe: probably can but if i own the hosting he can always blame it on me 01:04 < dassouki> i'm just being hypothetical, since this is a new thing for me 01:04 < amitz> dassouki: that's why if you're not ready for this stuff, find someone you can trust first. 01:04 < amitz> if you're unsure, just make sure you know all the risk. 01:05 < dassouki> indeed, so the suggestion is then to give them non sudo access 01:06 < dassouki> so i'll just do sudo addusr user_name 01:06 < dassouki> and then add him to a usergroup 01:06 < amitz> yeah in general. But it's not going to be 100% bullet proof for sure. 01:06 < dassouki> no worries 01:07 < amitz> just keep that in mind. Without even sudo access, he can make your life a living hell :-p 01:07 < gicque> there was a page that contained centos set up guide which i can no longer find. is it gone now? 01:08 < metap|pe> if you wanted to go all out, you could use some kind of filesystem auditor with the database on a remote machine 01:08 < metap|pe> that way you can run an audit and it will detect changed files 01:09 < dassouki> well all he' gonna be taking care of are: gedjango, svn, wiki, potgres and perhaps be involved in regular cleanup/backup activities 01:13 < gicque> ciao 01:13 -!- gicque [~gicque@S01060014d1426beb.wp.shawcable.net] has quit [Read error: Connection reset by peer] 01:13 < mike__> is there a way to exclude certain files from an rsync transfer? i've booted into finnix and mounted my partitions, yet some files keep locking up :-/ 01:14 < mike__> really makes no sense; they arn't possibly in use 01:15 < Peng_> Yes. 01:15 < Peng_> Check rsync's manpage, it has a few exclude features. 01:15 -!- Musicmasteria [~musicmast@c-76-120-44-192.hsd1.co.comcast.net] has joined #linode 01:16 < Musicmasteria> how do I ban someone's ip address using shorewall? 01:16 < metap|pe> would tcpwrappers work? 01:17 < Musicmasteria> never heard of them 01:17 < metap|pe> most tcp-using daemons are compiled with support for it. 01:18 < metap|pe> you can just add a rule to a text file, like /etc/hosts.deny iirc 01:18 < Musicmasteria> i looked for that file, it doesn't exist on mine 01:20 < metap|pe> which distro? 01:20 < Musicmasteria> ubuntu 01:21 < metap|pe> it should definitely have tcpwrapper support 01:22 < Musicmasteria> so how do I set that up? 01:22 < metap|pe> what does this say: dpkg --get-selections |grep libwrap 01:23 < Musicmasteria> libwrap0 install 01:24 -!- panpainter [~panpainte@c-66-235-62-16.sea.wa.customer.broadstripe.net] has quit [Quit: panpainter] 01:25 < metap|pe> http://ubuntu-tutorials.com/2007/09/02/network-security-with-tcpwrappers-hostsallow-and-hostsdeny/ 01:26 < mike__> Peng: I've read over the manpage, but im still unsure of how to exclude specific files 01:27 < metap|pe> --exclude "*~" --exclude "/some/path/*" 01:28 < mike__> metap|pe: thank you very much :-) 01:28 < metap|pe> np 01:30 -!- chesty [~chesty@chesterton.id.au] has quit [Quit: Coyote finally caught me] 01:32 -!- chesty [~chesty@chesterton.id.au] has joined #linode 01:32 < JoeK> When a user logs onto ssh, they get "login as: ", is there a way i can add an ascii image onto that screen? 01:33 < metap|pe> configure a motd 01:33 < JoeK> not AFTER they login 01:33 < metap|pe> ohhhhh. yes, that would be handy to do. 01:33 -!- nachtkriecher [~robot@li72-172.members.linode.com] has joined #linode 01:33 < nachtkriecher> hi 01:33 < nachtkriecher> anyone here ever used flex/bison? 01:34 < SelfishMan> JoeK: You can with a custom pam 01:34 < JoeK> o_o? 01:34 < SelfishMan> hell, look for pam_captcha or something like that 01:34 -!- nachtkriecher [~robot@li72-172.members.linode.com] has quit [] 01:34 < JoeK> hmm, ive seen it done and it looks cool xP 01:34 < SelfishMan> It's pretty easy with PAM but overkill typically 01:35 < SelfishMan> You might be able to do it with an ssh banner file 01:35 -!- litwol|mac [~litwol@ool-182f9dc3.dyn.optonline.net] has quit [Remote host closed the connection] 01:36 -!- nachtkriecher [~robot@li72-172.members.linode.com] has joined #linode 01:37 < nachtkriecher> stupid wireless 01:37 < nachtkriecher> anyone here have experience with flex/bison? 01:39 < bob2> !ask 01:39 < linbot> Don't ask to ask; just ASK 01:39 < nachtkriecher> im not sure what my question is yet 01:39 < nachtkriecher> well.... 01:39 < nachtkriecher> i think my flex is fine 01:40 < nachtkriecher> so i have some tokens that dont appear in my grammar 01:40 < nachtkriecher> i call them BAD 01:40 < nachtkriecher> and they throw a syntax error 01:40 < nachtkriecher> which is good 01:40 < nachtkriecher> but if there are any characters not BAD present, it also shows the good message 01:43 -!- Musicmasteria [~musicmast@c-76-120-44-192.hsd1.co.comcast.net] has left #linode [] 01:44 < nachtkriecher> anyone familiar with flex/bison at all? 01:55 < linbot> New news from forums: Understanding configuration profiles in Sales Questions and Answers 01:57 < JoeK> you should integrate forums with regular accounts 01:57 < JoeK> that would be greatly appreciated 01:57 < metap|pe> forums do tend to get hacked a lot, however. 01:59 < JoeK> make custom forums xP 01:59 < nachtkriecher> depends on the forum software 01:59 < nachtkriecher> and the people running it 01:59 < nachtkriecher> in my experience phpbb3 doesnt get hacked 02:00 < nachtkriecher> as long as it's installed right 02:00 -!- ph_ [~ph^@cm-84.211.180.8.getinternet.no] has quit [Remote host closed the connection] 02:16 -!- r3z [~r3z@r3zurector.net] has quit [Remote host closed the connection] 02:30 < mike__> heh, does anybody have any clue why rsync might freeze on certain files? i've booted into an alternate environment and mounted the partition, so they definitly arn't in use. and the files read fine earlier today, so it doesn't seem like a failing drive. 02:31 < bob2> freeze = ? 02:31 < mike__> I tried excluding a few of them, but the rest I need. I'm really at a loss here :-/ 02:31 -!- NotInternat [~nf@123-243-184-161.static.tpgi.com.au] has joined #linode 02:31 < bob2> ie strace rsync and look ath the output of 'dmesg' 02:31 < mike__> yes, as in im running it in verbose mode and the file just stays transfering..I've let it run in a screen overnight to confirm that it had definitly frozen 02:33 -!- kelvinq [~kelvinq@bb220-255-112-2.singnet.com.sg] has joined #linode 02:33 < bob2> ^^ 02:33 -!- elhippo [~elhippo@cpe-70-112-188-172.austin.res.rr.com] has quit [Quit: Leaving] 02:35 < mike__> so, any ideas? :-/ 02:35 < bob2> 16:31:38 < bob2> ie strace rsync and look ath the output of 'dmesg' 02:37 -!- internat [~nf@123-243-184-161.static.tpgi.com.au] has quit [Ping timeout: 480 seconds] 02:48 -!- TheJoe|ZzZz is now known as TheJoe 02:56 -!- Talman|Ghosting [~Talman|Aw@97-127-12-112.mpls.qwest.net] has joined #linode 03:00 -!- TheJoe is now known as TheJoe|ZzZz 03:03 < Talman|Ghosting> ... What do we know about the Hurricane Electric DC? I have to, it seems, write ad copy to counter ... this. 03:03 < Talman|Ghosting> http://www.servingdot.com/minnesota-semidedicated-hosting.htm 03:03 < Talman|Ghosting> OMG THEY USE A "more premium, exclusive network." 03:03 -!- Talman [~Talman|Aw@97-127-12-112.mpls.qwest.net] has quit [Ping timeout: 480 seconds] 03:10 < Peng_> Um. We know that it doesn't have enough power and/or cooling for modern servers. 03:11 -!- ph^ [~ph^@81.175.5.65] has joined #linode 03:11 < Peng_> (Definitely power; not sure about cooling.) 03:18 -!- darkbeholder [~darkbehol@124-168-246-42.dyn.iinet.net.au] has quit [Read error: Operation timed out] 03:20 -!- darkbeholder [~darkbehol@124-168-246-42.dyn.iinet.net.au] has joined #linode 03:24 -!- queso [~queso@cpe-71-72-215-28.cinci.res.rr.com] has quit [Remote host closed the connection] 03:25 -!- nachtkriecher [~robot@li72-172.members.linode.com] has left #linode [] 03:33 -!- jaiZ [~jaiZ@59.95.169.52] has joined #linode 03:33 -!- darkbeholder [~darkbehol@124-168-246-42.dyn.iinet.net.au] has quit [Remote host closed the connection] 03:34 -!- darkbeholder [~darkbehol@124-168-246-42.dyn.iinet.net.au] has joined #linode 03:41 -!- tiny [~ivob@tm.213.143.88.152.dc.telemach.net] has joined #linode 03:50 -!- darkbeholder [~darkbehol@124-168-246-42.dyn.iinet.net.au] has quit [Remote host closed the connection] 03:52 -!- darkbeholder [~darkbehol@124-168-246-42.dyn.iinet.net.au] has joined #linode 03:55 -!- dassouki [~dassouki@fctnnbsc15w-156034092086.pppoe-dynamic.nb.aliant.net] has quit [Quit: Lost terminal] 03:57 -!- internat [~nf@123-243-184-161.static.tpgi.com.au] has joined #linode 04:00 -!- NotInternat [~nf@123-243-184-161.static.tpgi.com.au] has quit [Ping timeout: 480 seconds] 04:10 -!- darkbeholder [~darkbehol@124-168-246-42.dyn.iinet.net.au] has quit [Remote host closed the connection] 04:27 -!- jimcooncat [~jim@pool-72-65-97-56.ptldme.east.myfairpoint.net] has joined #linode 04:29 < jimcooncat> morning! How do I recurse through a website I transferred to my linode, and substitute all instances of "efficiencypros.org" with "mabep.org"? find and sed? 04:30 < pwnguin> sounds like a reason to use relative URLs 04:32 < array> jimcooncat: sed -i 's/efficiencypros/mabep/g' *.html 04:32 -!- User12482 [~User82934@70.134.68.122] has quit [Quit: Client Quit] 04:32 < jimcooncat> pwnguin: you're most correct. I'm just temp-hosting for a client who doesn't understand them. I might break something though if I automatically convert, I'm afraid. 04:33 < jimcooncat> thanks array, I didn't know sed by itself had that ability. 04:33 < array> np :) 04:33 < array> dinner time! bbiab 04:45 -!- darkbeholder [~darkbehol@124-168-246-42.dyn.iinet.net.au] has joined #linode 04:50 -!- jaiZ [~jaiZ@59.95.169.52] has quit [Read error: Operation timed out] 04:51 -!- tiny [~ivob@tm.213.143.88.152.dc.telemach.net] has quit [Ping timeout: 480 seconds] 04:51 -!- tiny [~ivob@89-212-253-180.static.t-2.net] has joined #linode 04:52 -!- Andrew_ [~Andrew@c122-107-193-19.eburwd5.vic.optusnet.com.au] has joined #linode 04:59 < linbot> New news from forums: Think this would work? - Multiple paying users on one Linode in Feature Request/Bug Report 05:00 -!- darkbeholder [~darkbehol@124-168-246-42.dyn.iinet.net.au] has quit [Remote host closed the connection] 05:05 -!- jaiZ [~jaiZ@59.95.188.143] has joined #linode 05:05 < linbot> New news from forums: http://www.linode.com/forums redirection in Feature Request/Bug Report 05:05 -!- Smark is now known as Smark[Gone] 05:08 -!- nb [~nb@delta.bebout.net] has quit [Quit: nb] 05:09 -!- nb [~nb@delta.bebout.net] has joined #linode 05:16 -!- DephNet[Paul] [~Paul@93-97-225-151.zone5.bethere.co.uk] has joined #linode 05:16 -!- DephNet[Paul] [~Paul@93-97-225-151.zone5.bethere.co.uk] has quit [] 05:19 -!- DephNet[Paul] [~Paul@94-194-180-156.zone8.bethere.co.uk] has joined #linode 05:23 -!- cpg [~cpg@c-76-126-208-117.hsd1.ca.comcast.net] has quit [Quit: cpg] 05:24 -!- Duke [~snubby@snubby.user.oftc.net] has quit [Quit: G-Lined: Excessive clones] 05:59 -!- Harry_Mudd [~jon@c-71-235-73-29.hsd1.ct.comcast.net] has quit [Quit: WeeChat 0.3.0] 06:00 -!- ondrej [~ondrej@24-176-186-44.static.reno.nv.charter.com] has quit [Quit: Leaving] 06:04 < encode> completely offtopic, i'm loving my new toy --> http://www.ladyada.net/make/icetube/ 06:25 -!- laser` [~laser@dyn244136.shef.ac.uk] has joined #linode 06:27 -!- gotgeek [~cb7ac26b@webuser.linode.com] has joined #linode 06:27 < gotgeek> hello 06:27 < gotgeek> quick question 06:27 -!- praetorian [praetorian@124-170-80-19.dyn.iinet.net.au] has quit [Read error: Operation timed out] 06:28 < SpaceHobo> !ask 06:28 < linbot> Don't ask to ask; just ASK 06:28 < gotgeek> i was wondering if somone could provide me with an IP for each of the datacentres so i can test latency between them. As i am located in asutralia 06:28 < SpaceHobo> !download 06:28 < linbot> http://www.linode.com/forums/viewtopic.php?t=2636 06:28 < SpaceHobo> you can use the hostnames from that first post 06:29 < gotgeek> thanks 06:30 < gotgeek> woo dallas :p 06:30 < SpaceHobo> heh 06:30 < gotgeek> probs cause internode have a peer point near there.. 06:30 < SpaceHobo> being in yurp, I get best times to newark 06:30 < SpaceHobo> you may find fremont closest 06:30 -!- praetorian [praetorian@124-168-156-183.dyn.iinet.net.au] has joined #linode 06:30 < SpaceHobo> but I know someone in NZ who has a node not in fremont 06:31 < SpaceHobo> he doesn't seem to mind too much 06:31 < gotgeek> RUGO 06:31 < gotgeek> gah 06:31 < gotgeek> praetorian- your in aus.. wats the best datacentre for you? 06:31 < gotgeek> in regards to latency/round trip times 06:31 < SpaceHobo> praetorian: in your anus. What's the best datacentre for kegels? 06:34 * jdub is an internode user with dallas linode. it's fine. 06:35 < jdub> but i rarely feel the need to play inverse ping time penis length competitions. 06:36 < jdub> (plus, i read my mail in mutt on the linode. it's fine.) 06:37 -!- gotgeek [~cb7ac26b@webuser.linode.com] has quit [Remote host closed the connection] 06:37 < jdub> internode. linode. wonder how many other *node services i use. 06:37 < praetorian> SpaceHobo: I want you in my anus <3 06:38 < praetorian> i have a dallas linode and an atlanta linode, i noticed no difference much between the two 06:38 < praetorian> wasn't HE always the good datacentre latency wise for .au? 06:38 * praetorian looks at jeff 06:52 < Andrew_> was it? 06:52 < Andrew_> if it is, I'm changing.. which oe is HE? 06:59 -!- K-Zodron [~K-Zodron@irc.zezeniaonline.com] has quit [Read error: Connection reset by peer] 06:59 -!- K-Zodron [~K-Zodron@irc.zezeniaonline.com] has joined #linode 07:01 -!- chris9 [~chris@C-59-101-189-125.mel.connect.net.au] has joined #linode 07:07 -!- chris9 [~chris@C-59-101-189-125.mel.connect.net.au] has quit [Quit: chris9] 07:17 -!- jforman [~jforman@c-98-217-11-231.hsd1.ma.comcast.net] has quit [Quit: leaving] 07:25 < tiny> Hello! I'm faced with a task of hosting our domain(with authoritative DNS, email and other services) over 2 DSL lines from different ISPs. I know it's not Linode related. I have an account on linode and planing to do all that for private purpose too. 07:26 < tiny> This is a big job. Integrating everything so it works. 07:26 < tiny> Dealing with issues such as DNS server availability, SPAM, ... 07:26 < tiny> Any detailed guides ? 07:28 -!- jforman [~jforman@grenache.jeffreyforman.net] has joined #linode 07:28 < tiny> I've started with setting up dedicated FW box that will handle NAT, load balancing, VPN gateway, ... 07:31 < tiny> What are the cons if I decide not to use my own nameservers? 07:31 -!- darkbeholder [~darkbehol@124-168-246-42.dyn.iinet.net.au] has joined #linode 07:35 < praetorian> Andrew_: it's what i heard. cannot confirm if thats true tho 07:36 < Andrew_> hmm 07:36 < Andrew_> I'd have to ping them all 07:36 < Andrew_> but with Datacentre is HE? 07:36 < Andrew_> with = which 07:37 < praetorian> not atlanta, not newark, not dallas 07:37 < praetorian> the other one 07:37 < praetorian> !avail 07:37 < linbot> praetorian: Linode360 - 215, Linode540 - 132, Linode720 - 143, Linode1080 - 115, Linode1440 - 86, Linode2880 - 9 07:37 < praetorian> !avail-he 07:37 < linbot> praetorian: Fremont360 - 55, Fremont540 - 21, Fremont720 - 27, Fremont1080 - 9, Fremont1440 - 6, Fremont2880 - 2, Fremont5760 - 1, Fremont8640 - 1, Fremont11520 - 0 07:38 < praetorian> fremont 07:38 < Andrew_> fremont.. ahh already on that 07:41 < thornheart> Hello 07:42 -!- jtaji [~jtaji@c-76-124-234-212.hsd1.pa.comcast.net] has quit [Remote host closed the connection] 07:49 < Andrew_> ey 07:56 -!- Edgeman [~edgeman@dyn216-8-129-3.ADSL.mnsi.net] has quit [Ping timeout: 480 seconds] 07:57 -!- Edgeman [~edgeman@dyn216-8-173-8.ADSL.mnsi.net] has joined #linode 08:03 -!- irgeek [~jcsincla@mt-elbert.irgeek.com] has quit [Remote host closed the connection] 08:03 -!- Guest70 [~jcsincla@mt-elbert.irgeek.com] has quit [Remote host closed the connection] 08:03 -!- irgeek [~jcsincla@mt-elbert.irgeek.com] has joined #linode 08:03 -!- mode/#linode [+o irgeek] by ChanServ 08:05 < rainman`> thornheart, all you can do to avoid single point of failure is change DNS automatically :( 08:05 < rainman`> er, tiny: 08:07 < tiny> rainman`: Can I have someone run DNS servers for me while I still have full control over subdomain names, adding email accounts and such? 08:07 < tiny> I'm a bit rusty. Will have to read a lot. 08:07 < rainman`> yes, of course 08:08 < rainman`> well, with proper DNS hosting then 08:08 < irgeek> DNS doesn't have anything to do with the accounts on an email system. 08:09 < tiny> irgeek: yes of course (what was I thinking) 08:10 < irgeek> Also, hosting anything over two DSL lines from different ISPs is just asking for problems. You can't control routing so you can't fail-over between them. 08:10 < irgeek> You can try to do fancy NAT to try and balance their use, but it won't ever be perfect. 08:11 < tiny> irgeek: having 2 DSL lines is mostly for redundancy 08:11 < tiny> will not do much hosting (www, ftp). I just want email control over here. 08:12 < rainman`> it depends on what you balance 08:12 < irgeek> You can make them redundant for outbound communications, but for inbound communications you need really short TTLs and DNS that automatically updates. 08:12 < rainman`> just what i was about to say 08:12 * tiny tries to digest that 08:13 < irgeek> In Egypt I had to do a similar task. I ended up hosting the receiving MX on a Linode and having it know about the two routes into the network. If one failed, it would just try the second one. 08:13 < tiny> Ok. I think I got the part on DNS auto update. What's the story with short TTLs irgeek ? 08:14 < irgeek> TTL is how long DNS entries are cached. 08:14 < tiny> I see... 08:14 < tiny> This is more complicated then I thought. 08:14 < tiny> So .. 08:14 < irgeek> If you TTL is long and one of the DSL lines goes down, anyone who has that address cached won't be able to find your mail server until the DNS changes *and* the TTL expires. 08:15 < tiny> yep 08:15 < tiny> I figured all kind of problems will arrive. 08:18 < irgeek> For email, it's a solvable problem because the system is designed to fail-safe. Mail will be delayed but will almost always get through. Other types of hosting, however, aren't so simple. 08:19 < irgeek> Definitely don't host DNS on the DSL lines. 08:19 < irgeek> You want DNS off-site so it can monitor and switch over when there's a problem. 08:20 -!- v0lksman [~shayne@ottawa-hs-64-26-169-151.s-ip.magma.ca] has quit [Quit: Ex-Chat] 08:20 < tiny> Thanks for these tips irgeek 08:20 < tiny> irgeek: so you had receiving mail server on a linode account which basically forwared mail to your server behind working DSL line? 08:20 < irgeek> Yup. 08:21 < tiny> irgeek: sounds like a plan 08:21 < irgeek> Outbound went through the Linode as well because the Egyptian IP space is blacklisted in a lot of lists. 08:21 < tiny> I see. 08:22 < irgeek> Also because nobody at the ISPs in Egypt understood the concept of reverse DNS. It took me about six months to get them to fix it. 08:22 < tiny> irgeek: I could just use our ISPs SMTP servers for outbound right? 08:23 < irgeek> You could. If you trust your ISP to keep them functioning. 08:23 < irgeek> In Egypt, that's a gamble. 08:23 < tiny> irgeek: gotcha! 08:24 < irgeek> Not to mention, there are weird requirements in some places (like keeping copies of every email) which are much easier if you have full control of the inbound/outbound mail server. 08:29 < rainman`> indeed, NL has that for example 08:29 < rainman`> if you send mail through an ISP SMTP, they must store sender,recipient,subject 08:29 < tiny> irgeek: Yep. One more if you have time. :) ... How does DNS "monitor and switch over" and how does that relate to receiving SMTP on linode which switches between two routes? 08:29 < tiny> rainman`: NL? 08:29 < rainman`> netherlands 08:30 < tiny> ahh 08:30 < tiny> irgeek: I mean how does SMPT server on linode know when to switch? You wrote some scripts? 08:31 < tiny> irgeek: sorry if questions are a bit "cloudy" 08:31 < tiny> I'm guessing you wrote some simple script that just monitors IP availability 08:32 < irgeek> It doesn't need to know when to switch. It just knows there are two routes in. When the first route it tries doesn't work, it tries the other one. If that doesn't work either it queues the mail and tries again later. 08:32 < irgeek> That's normal behavior for SMTP. 08:32 < tiny> irgeek: I see. Thank a lot! 08:33 < tiny> Thanks* 08:33 < irgeek> You can do the same thing with two MX entries for the domain, but then you're relying on the rest of the world to deliver the mail correctly. I prefer to have a little more control than that. 08:35 < tiny> irgeek: And I suppose I don't even have to run my own nameserver. I can use the linode's nameserves and their DNS manager? 08:35 < irgeek> Yup. 08:35 < irgeek> I do. 08:36 < tiny> Well. I think I know which way I'll go. I simply don't have the time to do all this over DSL and system would be unreliable anyway. 08:37 < tiny> irgeek: thanks again! 08:37 < irgeek> No problem. 08:37 < straterra> is today tuesday? 08:37 < tiny> depends on your TZ :) 08:37 < irgeek> In many places. 08:37 < straterra> damn 08:41 -!- Bass10 [~Bass10@c-76-113-194-7.hsd1.mn.comcast.net] has joined #linode 08:45 < tiny> irgeek: what did you mean by " then you're relying on the rest of the world to deliver the mail correctly". How do you have more control with only one MX record? 08:45 < tiny> MX entrie* 08:50 < irgeek> With two MX records that point at two DSL lines you're assuming the rest of the Internet has their systems configured correctly to try the second MX if the first one fails - and most of the Internet does. That little bit that doesn't, however, always ends up causing problems though. 08:51 < irgeek> Also, when both DSL lines are down, you're still receiving mail and queuing it. When you get one (or both) back up you can flush the queue and have all the mail delivered. 08:51 < G> straterra: actually it's Wednesday in NZ, and about to be in Australia 08:52 < irgeek> Without the Linode in the middle, you end up waiting for everyone else's mail server to retry before mail gets delivered. 08:54 -!- supine [~marty@merboo.mamista.net] has quit [Ping timeout: 480 seconds] 08:57 -!- Mathew [~Elvis@cpc4-flit1-0-0-cust346.lutn.cable.ntl.com] has joined #linode 08:58 -!- jforman [~jforman@grenache.jeffreyforman.net] has quit [Remote host closed the connection] 09:00 -!- HalJordan_ [~HalJordan@host-69-144-128-127.bln-mt.client.bresnan.net] has quit [Remote host closed the connection] 09:00 -!- HalJordan_ [~HalJordan@host-69-144-128-127.bln-mt.client.bresnan.net] has joined #linode 09:05 < tiny> irgeek: ty 09:05 < tiny> understand now 09:06 -!- jforman [~jforman@grenache.jeffreyforman.net] has joined #linode 09:07 -!- supine [~marty@merboo.mamista.net] has joined #linode 09:08 -!- kelvinq [~kelvinq@bb220-255-112-2.singnet.com.sg] has quit [Quit: kelvinq] 09:17 -!- v0lksman [~shayne@ottawa-hs-64-26-169-151.s-ip.magma.ca] has joined #linode 09:23 -!- borris84 [~Lars@rrcs-96-10-93-114.se.biz.rr.com] has quit [Read error: Connection reset by peer] 09:23 -!- borris [~Lars@rrcs-96-10-93-114.se.biz.rr.com] has joined #linode 09:32 -!- r3z [~r3z@r3zurector.net] has joined #linode 09:49 -!- kelvinq [~kelvinq@bb121-7-88-165.singnet.com.sg] has joined #linode 09:50 -!- queso [~queso@cpe-71-72-215-28.cinci.res.rr.com] has joined #linode 09:56 < Kerem> linode.resize returns linodeid on success? 09:56 < linbot> New news from forums: I've lost my swap space in General Discussion 09:57 < Kerem> oops wrong window:-) 09:58 -!- Kerem [~kerem@keremdurmus.com] has quit [Quit: leaving] 09:58 -!- Kerem [~kerem@keremdurmus.com] has joined #linode 09:58 -!- linville [~linville@sapphire.tuxdriver.com] has joined #linode 10:03 -!- Edgeman [~edgeman@dyn216-8-173-8.ADSL.mnsi.net] has quit [Remote host closed the connection] 10:04 -!- ph^ [~ph^@81.175.5.65] has quit [Remote host closed the connection] 10:06 -!- TheJoe|ZzZz is now known as TheJoe 10:06 -!- atula [~neobreed@c-71-232-0-65.hsd1.ma.comcast.net] has joined #linode 10:09 -!- Edgeman [~edgeman@dyn216-8-173-8.ADSL.mnsi.net] has joined #linode 10:10 -!- Redgore [~redgore@94-193-27-221.zone7.bethere.co.uk] has joined #linode 10:12 -!- tiny [~ivob@89-212-253-180.static.t-2.net] has quit [Remote host closed the connection] 10:13 -!- Guspaz|m [cffdca03@webchat.mibbit.com] has joined #linode 10:25 -!- litwol|mac [~litwol@ool-182f9dc3.dyn.optonline.net] has joined #linode 10:26 -!- SpaceHobo [~spacehobo@spacehobo.user.oftc.net] has quit [Ping timeout: 480 seconds] 10:28 -!- supine [~marty@merboo.mamista.net] has quit [Ping timeout: 480 seconds] 10:30 -!- hfb [~hfb@pool-96-247-114-183.lsanca.dsl-w.verizon.net] has joined #linode 10:31 -!- Twayne [~waynemilt@cpe-071-070-201-028.nc.res.rr.com] has joined #linode 10:32 -!- PHPdiddy [~johns@comp.stedwards.edu] has joined #linode 10:40 -!- supine [~marty@merboo.mamista.net] has joined #linode 10:43 -!- Dreamr_3 [~Dreamer3@96-28-100-3.dhcp.insightbb.com] has joined #linode 10:44 -!- johndbritton [~john@ool-4574433a.dyn.optonline.net] has joined #linode 10:45 -!- hcl2 [~hardcorel@75.41.110.112] has joined #linode 10:46 < amitz> some codecs providers are asshole. I can understand the part where you have to install their toolbar. But now they don't give the uninstall method as promised. 10:46 < Guspaz|m> Which codecs in particular? 10:46 < Karrde> why are you installing codecs? 10:46 < Andrew_> Windows? 10:46 < Karrde> either use VLC (not recommended) or install CCCP 10:46 < Karrde> done 10:47 < Guspaz|m> ffdshow and Media Player Classic Homecinema combined can decode virtually everything, and neither have any toolbars. 10:47 < amitz> media.player.codec.pack.v3.8.0.setup.exe . yeah windows. 10:47 < Andrew_> in windows 7,there don't seem to be many codecs left that need to be installed 10:47 < Andrew_> why not use K-lite amitz? 10:47 < Guspaz|m> Don't ever touch those codec packs, just install ffdshow and be done with it. 10:47 < Karrde> because K-Lite is redundant and unintelligenmt 10:47 < Guspaz|m> None of this k-lite or codec pack bull. 10:47 < Andrew_> works fine here.. 10:48 < Guspaz|m> CCCP is probably the only tolerable codec pack, although I still don't see the need. 10:48 -!- Dreamer3 [~Dreamer3@96-28-100-3.dhcp.insightbb.com] has quit [Ping timeout: 480 seconds] 10:48 < Andrew_> Although, I no longer need it in windows 7... 10:48 < amitz> Andrew_: oh, I'm relatively not-knowledgeable in windows. 10:48 < Andrew_> In OSX you cant uninstall any codecs though 10:48 < amitz> heh, never believe I said that and it's relatively true :-D 10:48 < Guspaz|m> Andrew_: Windows 7's built-in decoders aren't exactly good. As in, I pop in a DVD and it produces a wonderful bunch of static with hints of the underlying image. 10:48 < Andrew_> You got an ATI Guspaz? 10:49 < Guspaz|m> No, GeForce GTX 285s in SLI. 10:49 < Andrew_> Because I know the ATI drivers had SERIOUS problems in windows 7.. 10:49 < amitz> I remembered back in my windows day, I can install this one codec and everything is playable.. 10:49 < Andrew_> at least in Windows media centre for TV 10:49 < Guspaz|m> Admittedly I haven't tested any other DVDs. The only reason I had that one using the Windows 7 codec is because I had accidentally disabled both MPCHC's internal codec and FFDSHOW's MPEG-2 codec. 10:50 < amitz> damn, can't remember the codec name. 10:50 < Andrew_> On an 8800 though, all my DVD's play perfectly though 10:50 < Guspaz|m> amitz: ffdshow. You still can. 10:50 < Guspaz|m> http://ffdshow-tryout.sourceforge.net/ 10:50 < Andrew_> amitz, What windows are you using 10:50 < Andrew_> thats the most important thing.. 10:50 < amitz> Guspaz|m: not that one but I would try that. XP. 10:50 < Guspaz|m> Alternatively, Media Player Classic Homecinema has internal decoders for almost everything these days (http://mpc-hc.sourceforge.net/) 10:50 < Andrew_> Codec pack usefulness differs based on the version of the OS 10:50 < Andrew_> and if its 64bit or 32bit 10:51 < Guspaz|m> It shouldn't matter; 64-bit stuff isn't really stable, so most media players and codecs you'll get are 32-bit. And they can't mix. 10:51 < Guspaz|m> I mean, you can get MPC, ffdshow, Haali's Splitter, all those in 64-bit, but it's not necessarily recommende. 10:51 < Andrew_> yeah.. thats true.. 10:52 < Andrew_> been there done that 10:52 < Andrew_> I actually reverted to 32bit Vista from 64bit 10:52 < Andrew_> But knowing me, I'll probably get 64bit Windows 7 again.. 10:53 < Guspaz|m> I'm not talking about 64-bit in general. I'm happy running Windows 7 x64, and I have to to use my RAM (12GB). But for decoding stuff, the 64-bit software provides no real speed benefit, and has poor stability. 10:53 < Guspaz|m> So you don't really get anything out of it, but you pay in buggyness. 10:53 < Andrew_> yeah.. thats true.. I know what you mean.. been there done that 10:54 < Andrew_> but even 64bit Windows in general, I decided to downgrade.. But I guess if I had a Core I7, 64bit Windows would be pure win 10:54 < Guspaz|m> But I do :P 10:54 < Guspaz|m> I wouldn't fancy cutting my available RAM by 75%. 10:55 < Guspaz|m> Although I could probably get more than 4GB working using PAE under 32-bit. 10:55 < Andrew_> I have 4GB, so I had to make that choice.. 10:56 < Andrew_> But I figure, I don't even use 3GB generally 10:56 -!- r3z [~r3z@r3zurector.net] has quit [Read error: Connection reset by peer] 10:57 < amitz> heh, different providers but exactly equal installation display 10:58 -!- SpaceHobo [~spacehobo@78-105-8-188.zone3.bethere.co.uk] has joined #linode 10:58 < Guspaz|m> Going from 4 to 3 though is a bit different from going from 12 to 3 10:58 < Guspaz|m> I do tend to use it though. I don't necessarily need 12, probably I wouldn't notice much difference dropping to 6. But I do tend to have lots of stuff running, and so having a bunch available for caching is nice. 10:59 < Andrew_> yeah.. lol 10:59 < mikegrb> lolz 10:59 < Andrew_> its the geek factor.. 10:59 < Andrew_> either way.. for you, not worth it.. 11:00 < Guspaz|m> Also I spent the last several years on 2GB of RAM, which was painfully insufficient :P 11:00 < Guspaz|m> Always swapping 11:00 < Andrew_> thats 2GB though.. Wtf are you running? 11:00 < Andrew_> 3D studio max, maya, and after effects at the same time? 11:00 < Andrew_> or just eclipse/Openoffice (that woud probably do it) 11:01 < Guspaz|m> Vista, Chrome, HydraIRC, random background apps, media players, etc. It adds up fast. 11:01 < Guspaz|m> Especially browsers, they're a big hog. 11:01 < Guspaz|m> Good for half a gig usage at least, or more. 11:01 < rainman`> what happened to the days where we could run a browser in 10MB 11:01 < Guspaz|m> rainman`: links still works :P 11:02 < Andrew_> but most of Songbird is probably caching 11:02 < rainman`> not for most sites :( 11:02 < Andrew_> err.. sorry.. Firefox 11:02 * Andrew_ is a Songbird Champion, and they are both XUL 11:02 -!- daMaestro [~jon@content.beatport.com] has joined #linode 11:02 < Guspaz|m> I'm very happy with Chrome, although it's not necessarilly memory efficient. It doesn't tend to bloat like firefox since closing a tab kills the process, but if you throw enough RAM at it, it's really fast. 11:03 < Guspaz|m> Although Firefox was horrible, it'd get into the gigabytes of RAM usage and then you'd have to kill it. 11:03 < Andrew_> not for me.. 11:03 < Andrew_> Ever tried closing tabs? 11:04 < Guspaz|m> Yeah, but Firefox didn't tend to really free the RAM when I closed tabs, and I tended to have a decent number of tabs open usually. 20+ wasn't uncommon. 11:04 < Andrew_> remember though, a lot is caching.. Sngbird does that too, it checks your available memory, and it uses a certain fraction of it for the database 11:04 < Andrew_> ahh ok 11:04 < Andrew_> I still think it may be caching related anyway.. but Chrome is faster.. 11:04 < Guspaz|m> Caching or not, it leads to memory fragmentation. You free the RAM, but it's fragmented and not very usable. 11:04 < Andrew_> Anyway, use what you want 11:04 < Guspaz|m> Chrome only gets around it because when you close the tab the entire process is killed. 11:04 < Guspaz|m> Although that doesn't help with Chrome's shared memory usage, which still does tend to slowly balloon over time. 11:05 < Guspaz|m> Just not as badly as Firefox did. 11:05 < Andrew_> I shut down my computer when not using it, so I never notice 11:05 < Andrew_> and not really a background app person.. 11:05 < Guspaz|m> Ah, see, before I was a laptop user so I put it to sleep (never really not running), and my desktop and work computers stay on. 11:07 < Andrew_> I hibernate too these days.. but I close stuff a lot.. 11:07 < Andrew_> its just me.. 11:08 < Andrew_> like Beyonce is on TV at the moment, I was considering closing it, but now Rihanna and Kanye west.. Which means definitely.. 11:08 -!- r3z [~61591a6f@webuser.linode.com] has joined #linode 11:09 < r3z> Bleh 11:09 < r3z> Server keeps oom'ing 11:09 < r3z> or something.. 11:09 < r3z> Locks up hard. 11:09 < Andrew_> When you go to lish, you get a whole bunch of weird errors? 11:09 < SpaceHobo> SERVER KEEPS NOMING 11:09 < SpaceHobo> NOM NOM NOM NOM NOM 11:09 < Guspaz|m> OM NOM NOM NOM 11:09 < r3z> NOM NOM NOM NOM NOM CRUNCH NOM NOM NOM NOM NOM 11:09 < Andrew_> NOM NOM NOM NOM NOM NOM NOM NOM ... NOM NOM 11:10 < Andrew_> just kill bind.. You'll save 80mb easy.. 11:10 < Guspaz|m> LOLBIND 11:10 < r3z> No bind. 11:10 < r3z> I use linode for dns ;) 11:10 < Guspaz|m> Swap out Apache for lighttpd/nginx? 11:10 < r3z> No sense in reinventing the wheel 11:11 < r3z> Ya I need to do that. 11:11 < Andrew_> actually, lighttp uses more thn I thought 11:11 < r3z> Never used lightttpd or nginx though 11:11 < Andrew_> reduce the fastcgi processes? 11:11 < Andrew_> Actually, I found in Mysql, killing the other database types saved me 40mb 11:11 < Guspaz|m> Andrew_: More how so? It just has the one process, which isn't usually too heavy on RAM. 11:12 < Andrew_> hmm on mine it has a few.. maybe its the difference between Lighttpd, and Apache 11:13 < Andrew_> On lighttpd, you specify the max children and processors.. 11:13 < Andrew_> but meh.. dunno 11:13 < r3z> I have ClamAV and amavis. 11:13 < r3z> they eat a shitton of ram 11:13 < Andrew_> you on a 360 r3z? 11:13 < r3z> ya :( 11:14 < r3z> I think I might get rid of clamav and amavis. 11:14 < Andrew_> You are probably best off posting a ps aux to paste bin.. 11:14 < Andrew_> so we can laugh.. 11:14 < r3z> hah ya for real 11:14 < r3z> Killed both of those and now only using 100mb ram lol 11:14 < mikegrb> lolz 11:14 < Andrew_> I had this prob a week ago though.. 11:14 -!- zefster [~zef@c-76-25-216-75.hsd1.co.comcast.net] has joined #linode 11:15 < Andrew_> You might want to check email still works though.. 11:15 < Andrew_> wait.. you using virtualmin or something? 11:16 < Guspaz|m> clamav is a huge memory hog. 11:16 < Guspaz|m> Also spamassassin. 11:16 < Andrew_> did you install them though r3z? 11:16 < r3z> Ya I will have to route around clamav 11:17 -!- zefster [~zef@c-76-25-216-75.hsd1.co.comcast.net] has quit [] 11:17 < Andrew_> Another I had was FTP.. I actually dumped FTP completely, and use SFTP now (which puggybacks off OpenSSH) 11:18 < Andrew_> egh.. mediawiki.. what a nightmare.. 11:20 < Andrew_> seriously though, if you post your ps aux, we could give you suggestions 11:20 < Andrew_> anyone here who has ever owned a 360 knows what an OOM is by now.. 11:21 * path knocks on wood 11:23 * Guspaz|m owns a 540, and has never had OOMing issues ^_^ 11:23 < adj> i have a 360 and havent had any ooms 11:23 < Andrew_> Ok.. maybe not :P 11:23 < Guspaz|m> I'm pretty low load though. Only have to put out about a million pageviews a month, which is pretty tiny. 11:24 < rainman`> my 360 never OOMd 11:25 -!- Intensity [1g1gsHJ6JA@panix5.panix.com] has quit [Ping timeout: 480 seconds] 11:25 < Twayne> . 11:28 < r3z> Crashed tables in mysql are bad MMkkk... 11:29 < rainman`> well 11:29 < rainman`> at least you have backups 11:29 < Andrew_> Hmm.. not as bad as when your "indexed URL counter" in Google webmaster tools actually starts going backwards.. 11:29 < r3z> No need for the backup. Just repair table tablename and boom done ;) 11:30 < r3z> Most of the time 11:30 < r3z> lol 11:30 < mikegrb> lolz 11:30 < r3z> I guess thats what you get for doing a wordpress update... 11:30 < Huitzilopochtli> rad 11:31 < adj> a poorly configured 360 will oom ;) 11:31 < rainman`> just give it 5G swap 11:31 < r3z> Wasnt ooming.. 11:31 < Andrew_> Good idea rainman`.. I wonder what the other users on the server would say about that.. 11:31 < r3z> I thought it was at first until I got on the console. 11:32 < Guspaz|m> I wonder what anybody trying to use services hosted by that linode would say :P 11:32 < Guspaz|m> It wasn't OOMing, it has actually been posessed by LAGLAR, the evil demon who invades your computer to cause lag? 11:33 < Andrew_> speaking of being possessed.. My node is running NSD, which I uninstalled 12 hours ago.. 11:33 -!- praetorian [praetorian@124-168-156-183.dyn.iinet.net.au] has quit [Remote host closed the connection] 11:33 -!- praetorian [praetorian@124-170-95-23.dyn.iinet.net.au] has joined #linode 11:33 < rainman`> i've actually chased lpd on a few servers of mine for a year 11:33 < rainman`> it just came back every few months 11:33 < Andrew_> gee.. I guess I'm major fail 11:34 -!- loxs [~loxs@85-130-31-93.2073154325.ddns.cablebg.net] has joined #linode 11:34 < Guspaz|m> Just because you've deleted the executable doesn't mean it isn't still running. You delete the copy on disk, but it's still in RAM. 11:35 < Andrew_> yeah.. 11:35 < Andrew_> i know 11:36 < SpaceHobo> or you can unlink a file and a running process will still have a handle open, keeping the blocks allocated 11:36 < SpaceHobo> woo anaonymous files! 11:36 < Guspaz|m> True, but that is also handy. 11:36 -!- bluetegu [~bluetegu@93-172-195-184.bb.netvision.net.il] has quit [Ping timeout: 480 seconds] 11:36 < Guspaz|m> It means that you can unlink or move a file that's being used, unlike in Windows where it will just refuse. 11:36 < Guspaz|m> It means that you can move a file without worrying about killing current downloads of it. 11:37 < SpaceHobo> sure 11:37 < SpaceHobo> yep 11:37 < Guspaz|m> Or delete it. 11:37 -!- elfgoh [~dingding@adsl30.dyn212.pacific.net.sg] has joined #linode 11:38 * r3z needs to remember how to bypass clamav now. 11:38 < Andrew_> its not a matter of remembering, its a matter of either googling, or randomly changing things 11:38 < Andrew_> or likely, both 11:39 < r3z> randomly changing things.... hrm. 11:39 < Andrew_> you know you want to.. 11:43 < r3z> Bypassing clamav is the easy part 11:43 < r3z> bypassing amavis and keeping spamassasin is what I need to figure out the best approach for., 11:45 < linbot> New news from linodelibrary: Setting up a LAMP Server on Fedora 11 11:45 -!- TheJoe is now known as TheJoe|ZzZz 11:48 -!- K-Zodron [~K-Zodron@irc.zezeniaonline.com] has quit [Read error: Connection reset by peer] 11:48 -!- K-Zodron [~K-Zodron@irc.zezeniaonline.com] has joined #linode 11:50 < r3z> ok just bypassed amavis and all of it for now. 11:50 < r3z> Will fix the problem later. 11:51 -!- SDjernes [~shawn@ip70-187-15-75.om.om.cox.net] has joined #linode 11:51 -!- SDjernes [~shawn@ip70-187-15-75.om.om.cox.net] has left #linode [] 11:58 -!- laser` [~laser@dyn244136.shef.ac.uk] has quit [Remote host closed the connection] 11:59 -!- r3z [~61591a6f@webuser.linode.com] has left #linode [] 12:02 -!- linny [~ca9c0afb@webuser.linode.com] has joined #linode 12:04 -!- ph_ [~ph^@cm-84.211.180.8.getinternet.no] has joined #linode 12:07 < erikh> spamass-milter 12:07 -!- linny [~ca9c0afb@webuser.linode.com] has quit [Remote host closed the connection] 12:10 < adj> heh. so installing suhosin in cent5 is a breeze 12:10 < adj> now to write a new vcl for varnish and this host is ready to go =) 12:25 < amitz> heh, I have this collection of video that must be run with combination of totem, mplayer, and vlc. heh 12:25 < amitz> wait, scratch that vlc. 12:25 < amitz> the state of decoder in linux is a mess. 12:27 < JshWright> I have a hard time finding a video that VLC won't run 12:27 < linbot> New news from linodelibrary: Setting up a LAMP Server on Gentoo Linux 12:27 -!- goofykinky [~chatzilla@CableLink-173-105-28.CPE.InterCable.net] has joined #linode 12:28 < amitz> JshWright: I'm about to take that challenge but the size is a whooping 150MB> 12:30 < goofykinky> hi, what kind of internet connection does the VPS have? 12:30 < JshWright> goofykinky: that depends on the datacenter 12:30 < amitz> [00000282] main input error: demux doesn't like DEMUX_GET_TIME <- for vlc 12:30 < Andrew_> an awesome one goofykinky 12:31 < goofykinky> lets say Dallas? 12:31 < Andrew_> I'm in australia, and its fast even from here.. 12:31 < amitz> goofykinky: Expect at least 50 mbps by default, and you can ask to raise it? if I get your question right? 12:32 -!- jroes_ [~jroes@74.207.225.154] has joined #linode 12:32 -!- lakin [~lakin@S01060013101832cd.cg.shawcable.net] has joined #linode 12:33 < JshWright> The DC in Dallas is operated by The Planet http://www.theplanet.com/data-centers/ 12:34 < vblank> Network connections are as fast as the slowest link. Has anybody ever had their linode be the slowest link? 12:34 < caker> You are: the slowest link! (dun dun dunnnnnn) 12:34 < JshWright> vblank: that's true for a 1-to-1 connection 12:35 < JshWright> 250 broadband users all pulling a large file are going to be able to saturate the linode's network interface 12:35 < JshWright> thus making it the bottleneck 12:36 < amitz> and that will happen for anything related to summer glau. 12:36 < goofykinky> well I just asked because I wanted to know whats the default thanks ;) 12:37 -!- elhippo [~elhippo@cpe-70-112-188-172.austin.res.rr.com] has joined #linode 12:37 < adj> if you are pushing over 50mbps off a linode you need to investigate how much a CDN will cost, imho 12:38 < JshWright> or If you're pushin 50mbps, you're gonna blow through your 360's bandwidth allocation in a matter of hours 12:38 -!- agentbleubleu [~agentbleu@lns-bzn-40-82-251-179-229.adsl.proxad.net] has joined #linode 12:39 < vblank> Sorry, just a sore topic with me. My manager insists we use gigE adapters everywhere and our rack switches have 2gig uplinks to the core. 12:39 -!- getsmart [~getsmart@88-149-241-21.dynamic.ngi.it] has joined #linode 12:39 < jed> vblank: I wouldn't worry about it 12:40 < vblank> thinking about network speed and collision domains and placing servers based on who is going to talk to who is a lost art 12:40 < vblank> it's every team for themselves! 12:40 < Guspaz|m> Intel's new Xeon platform has a 10GigE controller. Too bad 10GigE switches are insanely expensive. 12:40 < amitz> I wonder if there is a mechanism to torrent a web application.. 12:41 < JshWright> amitz: the overhead invovled would be way too high 12:41 < Andrew_> what do you mean torrent it? 12:41 < SpaceHobo> what indeed 12:41 * amitz is thinking about possible overheads. 12:42 -!- bluetegu [~bluetegu@93-172-4-119.bb.netvision.net.il] has joined #linode 12:42 < amitz> basically you have multiple people fulfilling the request to a certain webserver. 12:43 < amitz> it is as if they host some components of the server, maybe the images, videos, etc. 12:43 < Andrew_> night 12:43 < amitz> or the page itself. 12:43 -!- Andrew_ [~Andrew@c122-107-193-19.eburwd5.vic.optusnet.com.au] has quit [Quit: Leaving] 12:43 < adj> i dont want people hosting content for me that i dont trust 12:43 < Guspaz|m> The files served up in that case are too small to get any benefit from BitTorrent. 12:44 < adj> just use a good cache server and set content expirations properly 12:44 < Guspaz|m> Oh wait, you mean distributing the files themselves. 12:44 < adj> and offload large expensive files to a CDN 12:44 < amitz> adj: I think encryption has developed enough to make one person not knowing the whole content. 12:44 < Guspaz|m> You can offload small files to a CDN too, if you're high traffic. 12:44 < Guspaz|m> CDN bandwidth can cost 39% of Linode bandwidth or less. 12:45 < adj> amitz: its moot regardless. your stuffing a square problem in a round hole 12:45 < Guspaz|m> So it can make sense to do some offloading. 12:47 < amitz> I mean to have the webbrowser automatically read a torrent request for say an image to be displayed on the page, and torrent accordingly. 12:47 < vblank> i would optimize the flux capacitor 12:47 < amitz> so instead of maybe maybe 12:47 < amitz> I hope you get what I mean. 12:48 < adj> amitz: i get it. but why does it make sense to chunk out a file thats a few hundred kilobytes? 12:48 < amitz> and apply the same thing to downloading large file. Basically makes torrenting transparent, the user doesn't even know they torrent. 12:49 < amitz> adj: I thought maybe that will add up? 12:49 -!- dajhorn [~dajhorn@transmisor.vanadac.com] has joined #linode 12:49 < adj> in my experience torrents are WAAAY slower that direct downloads 12:49 < adj> than* 12:49 < adj> torrents solve the problem of asynchronous up/down rates from internet providers. not distribution 12:50 < amitz> adj: oh, I don't know about the slow part. I rarely torrent.. 12:51 < Guspaz|m> The file doesn't need to be split up into chunks, but the latency of joining a torrent swarm just to get a small image would cause page loads to take an order of magnitude longer (or more). 12:51 < amitz> adj: what you you mean? I thought downloading when connecting to ISP is always faster than uploading? 12:51 < Guspaz|m> Why bother? What are you doing that would have so many images that you'd not want to pay for them? 12:51 < straterra> imagebin 12:52 < Guspaz|m> If your cost of bandwidth is outstripping your ad earnings, you're doing something wrong. 12:52 < amitz> I'm not doing it. It's just something that cross my mind. The server for poor people :-p 12:52 < straterra> server for poor people is a 260 12:52 < straterra> err, 360 12:53 < Guspaz|m> Say an image is 200KB. The cost to deliver that image via Linode is ~$0.00002 12:53 < Guspaz|m> Less than half that via a CDN. 12:54 < Guspaz|m> That means that you need a CPM of about $0.019 to pay for that bandwidth, which is extremely low. 12:55 < amitz> hmm.. I have nothing to say for now but I'm sure one day it has it's own use. I'm also thinking of IBM's latest finding (2 months ago(?)) about making a client able to process a data while the data is still in encryption condition. I don't know about the detail yet but it seems promising... 12:56 -!- Redgore2 [~redgore@94-193-27-221.zone7.bethere.co.uk] has joined #linode 12:56 < adj> its use is not http traffic ;) 12:56 < adj> perhaps something like dropbox could use that 12:56 < amitz> coupled that with the transparent torrenting then you can tap into a super computer consisting of multiple bot controlled computers.. 12:56 < Guspaz|m> Content delivered to a web browser for rendering is extremely time-sensitive. It will never make sense to distribute it via P2P. Streaming content, perhaps, but never rendered content (images, text, etc) 12:58 -!- getsmart [~getsmart@88-149-241-21.dynamic.ngi.it] has quit [Quit: Ex-Chat] 12:58 < vblank> if it mades sense somebody would have done it already 12:58 * jforman is happy, moving his wordpress blog over to his linode 13:01 -!- sabufrancis [~sabufranc@115.240.80.226] has joined #linode 13:02 -!- sabufrancis [~sabufranc@115.240.80.226] has quit [] 13:02 < amitz> vblank: maybe the NSA does it :-p 13:02 -!- sabufrancis [~sabufranc@115.240.80.226] has joined #linode 13:02 < vblank> well if it's good for the NSA... 13:03 < vblank> there is a concept know as KISS, check it out 13:03 -!- Redgore [~redgore@94-193-27-221.zone7.bethere.co.uk] has quit [Ping timeout: 480 seconds] 13:03 < amitz> also considering the finding on resisting attack on torrent poisioning. The 4-8 years ago papers? 13:03 < sabufrancis> Hi. I am a new linode user. I need to run a program that uses libX11.so.6 I have a linode with Ubuntu 9.04 on it. Can someone help? 13:03 -!- Redgore [~redgore@94-193-27-221.zone7.bethere.co.uk] has joined #linode 13:03 < amitz> yeah, keep it simple. 13:03 < Guspaz|m> Are you trying to install the program from source, from a package, from a repository? 13:04 < amitz> anyway, I'll just return back to lurking. 13:04 < sabufrancis> The program does not require an installation (It is Rebol SDK) 13:05 < sabufrancis> The SDK works fine from my local Ubuntu 9.04 box 13:05 < Guspaz|m> Your local box likely has X installed 13:05 -!- Redgore2 [~redgore@94-193-27-221.zone7.bethere.co.uk] has quit [Ping timeout: 480 seconds] 13:05 < sabufrancis> I tried installing xfree at my linode using apt-get but I think the appropriate repository is not mentioned in the sources.list 13:06 < Guspaz|m> You should be able to install libx11 from the repository. 13:06 -!- goofykinky [~chatzilla@CableLink-173-105-28.CPE.InterCable.net] has quit [Quit: ChatZilla 0.9.85 [Firefox 3.5.3/20090824101458]] 13:06 < Guspaz|m> Probably the libx11-6 package 13:06 -!- jaiZ [~jaiZ@59.95.188.143] has quit [Quit: Leaving] 13:07 < sabufrancis> Thx. That worked. Now it is asking for libXext :( 13:07 -!- Twayne [~waynemilt@cpe-071-070-201-028.nc.res.rr.com] has quit [Ping timeout: 480 seconds] 13:07 < Guspaz|m> libxext6 package? 13:08 -!- agentbleubleu [~agentbleu@lns-bzn-40-82-251-179-229.adsl.proxad.net] has left #linode [] 13:08 < Guspaz|m> You can try running "aptitude search " 13:08 < Guspaz|m> For example, "aptitude search libxext" should get you that result. 13:08 -!- Bass10 [~Bass10@c-76-113-194-7.hsd1.mn.comcast.net] has quit [Quit: Leaving] 13:08 < sabufrancis> Great. Thanx 13:09 < Guspaz|m> You may not have aptitude installed by default, though, I'm not sure. It's a replacement for apt-get. 13:09 -!- Redgore2 [~redgore@94-193-27-221.zone7.bethere.co.uk] has joined #linode 13:09 < SpaceHobo> apt-cache search 13:09 < SpaceHobo> is the traditional apt way 13:09 < sabufrancis> I do have aptitude ;-) (sometimes in real life too) 13:10 < sabufrancis> Thanks a lot 13:10 < sabufrancis> WHenever I use apt-get I get a lot of locale warnings. But the apt-get install proceeds fine. How do I set the locale? (As u can see I am really a newbie in disguise) 13:11 < bd_> sabufrancis: debian or ubuntu? 13:11 < bd_> also which locale? 13:11 < sabufrancis> ubuntu 9.04 13:12 -!- andrew_j_w [~andrew@88-97-29-168.dsl.zen.co.uk] has joined #linode 13:12 < bd_> ... and which locale is it complaining about? (if unsure, run 'locale' from the command line and it'll complain about it some more) 13:12 < linbot> New news from linodelibrary: Installing and Configuring phpBB 13:13 < sabufrancis> "perl: warning: Setting locale failed." 13:13 < sabufrancis> "perl: warning: Please check that your locale settings:" 13:13 < bd_> p.linode.com <-- for long pastes 13:13 < sabufrancis> "LANGUAGE = (unset)," 13:16 < bd_> paste the whole thing in p.linode.com please, the (unset) line isn't the important one 13:16 -!- Redgore [~redgore@94-193-27-221.zone7.bethere.co.uk] has quit [Ping timeout: 480 seconds] 13:17 -!- Redgore2 is now known as Redgore 13:17 -!- Berto is now known as Guest158 13:18 -!- Turl [~Turl@host26.190-224-48.telecom.net.ar] has joined #linode 13:18 < sabufrancis> Thx. I've copied the Perl warning in p.linode.com 13:18 < sabufrancis> under "sabufrancis" 13:18 < bd_> en_IN, hm 13:18 * bd_ searches 13:18 < sabufrancis> I am from India 13:19 < bd_> okay, looks like it's included in the 'locales' package 13:19 < bd_> which is a bit odd, ubuntu usually puts these in language-pack-* packages 13:19 < dajhorn> sabufrancis: Run `sudo locale-gen en_IN.UTF-8` at the prompt. 13:19 < bd_> but anyway, install locales and it should be fine 13:19 < bd_> dajhorn: he's on locale en_IN not en_IN.UTF-8 13:20 < dajhorn> bd_: I get the same result on my system. en_IN is mapped to UTF-8. 13:20 < bd_> oh? 13:20 < bd_> well anyway: http://packages.ubuntu.com/search?searchon=contents&keywords=en_IN&mode=exactfilename&suite=jaunty&arch=any 13:20 < bd_> I think you'll need locales before localegenning it at least 13:21 < bd_> not sure if ubuntu automatically installs it 13:21 < dajhorn> bd_: It is a dependency of the ubuntu-minimal metapackage. 13:21 < dajhorn> bd_: If he doesn't have ubuntu-minimal, then he has borked his system. 13:22 < bd_> ah, okay 13:23 -!- supine [~marty@merboo.mamista.net] has quit [Ping timeout: 480 seconds] 13:23 < bd_> I'm mostly a debian person, you see :) 13:23 < bd_> so yeah, just locale-gen it then I guess 13:23 -!- zack [~zack@c-98-210-109-85.hsd1.ca.comcast.net] has joined #linode 13:23 -!- zack [~zack@c-98-210-109-85.hsd1.ca.comcast.net] has quit [] 13:24 < sabufrancis> I gave "sudo locale-gen en_IN.UTF-8" and that command went fine 13:25 < sabufrancis> But locale still says warnings such as "Cannot set LC_CTYPE to default locale: No such file or directory" 13:25 < bd_> try just en_IN then 13:25 < dajhorn> sabufrancis: Logout, and login. 13:26 < dajhorn> dajhorn: Then pastebin what the `locale` program tells you. 13:26 < bd_> dajhorn: that shouldn't affect anything -locale files are loaded at program start based on env variable settings 13:26 < sabufrancis> Great. No warnings now. Seems to have worked. Thanks 13:26 < bd_> and the env variables are likely being passed down from the ssh client 13:26 < bd_> oO 13:26 < bd_> weird 13:26 < dajhorn> :D 13:28 < sabufrancis> Great. I've been able to use the Rebol SDK ("rebcmd" specifically). I wanted to use it to run Fastcgi protocol on Cherokee, but alas no such luck :( Cherokee is still choking on Rebol. 13:28 < vblank> amitz: 13:29 -!- cpg [~cpg@c-76-126-208-117.hsd1.ca.comcast.net] has joined #linode 13:29 < sabufrancis> Anybody here using Cherokee web server on Ubuntu 9.04? 13:30 -!- Redgore2 [~redgore@94-193-27-221.zone7.bethere.co.uk] has joined #linode 13:30 -!- kelvinq [~kelvinq@bb121-7-88-165.singnet.com.sg] has quit [Quit: kelvinq] 13:31 < agentbob> can anyone recommend a free app for documenting mysql schemas? 13:31 < vblank> pen and paper 13:31 < vblank> lol 13:32 < mikegrb> lolz 13:32 < agentbob> rdrr 13:32 < vblank> mysqldump can produce a sql import file that describes the database, would that meet your need? 13:33 < agentbob> guess i'm wondering if there's some sort of equivalent to php's phpdoc for mysql 13:36 -!- supine [~marty@merboo.mamista.net] has joined #linode 13:36 -!- Redgore [~redgore@94-193-27-221.zone7.bethere.co.uk] has quit [Ping timeout: 480 seconds] 13:36 -!- Redgore2 is now known as Redgore 13:36 < vblank> in Phpadmin, in the structure tab, you can click on the data dictionary, thats a nice looking report 13:36 < vblank> phpMyAdmin 13:37 < Guspaz|m> Which, coincidentally, can (and should) be installed from the repositories ;) 13:37 -!- waldron [~waldron@c-24-218-5-62.hsd1.ma.comcast.net] has joined #linode 13:37 < waldron> Wow... irc. it's been a while! 13:38 -!- silverblade [~silverbla@cust116-dsl93-89-128.idnet.net] has joined #linode 13:39 < waldron> Just got my linode up, have installed Debian on advice of people I trust. 13:39 < waldron> Now I need to install/config full LAMP stack, but I'm something of a newb 13:39 < waldron> Can anyone here suggest a guide that isn't from 2004?? 13:40 < waldron> Or should I just go ahead and dive in? 13:41 < straterra> apt-get apache php mysql :P 13:41 < straterra> Or whatever the debian command is 13:41 < straterra> These days..it works pretty much out of box 13:41 < JshWright> http://library.linode.com/lamp-guides/debian-5-lenny/ 13:42 < vblank> If I was a newb I would use something like webmin to configure and maintain one you got the stack installed 13:42 < adj> waldron: http://www.google.com/search?hl=en&client=firefox-a&rls=org.mozilla%3Aen-US%3Aofficial&tbo=p&tbs=qdr%3Ad&q=debian+lamp+howto&aq=f&oq=&aqi= 13:42 < adj> there are 'debian lamp howto' matches from the last 24hours 13:43 < waldron> Awesome... I'm going to go try some stuff. Thanks all of you. brb. 13:43 < JshWright> all the cool kids aren't using LAMP anymore anyway... 13:44 < adj> no? 13:44 -!- queso [~queso@cpe-71-72-215-28.cinci.res.rr.com] has quit [Quit: Leaving...] 13:45 -!- loxs [~loxs@85-130-31-93.2073154325.ddns.cablebg.net] has quit [Ping timeout: 480 seconds] 13:45 -!- zack_ [~zack@c-98-210-109-85.hsd1.ca.comcast.net] has joined #linode 13:46 -!- peter [~jsmith@peter.mcinerney.id.au] has joined #linode 13:46 < vblank> there are some cool kids where I work. 13:46 -!- axod [56970259@webchat.mibbit.com] has joined #linode 13:46 < vblank> I once needed 'on-call' info to page admins after hours about issues 13:47 -!- sabufrancis [~sabufranc@115.240.80.226] has quit [Quit: sabufrancis] 13:47 < vblank> he developed a web application where I needed to sumbit xml 13:47 < vblank> uh. yeah. no thanks... 13:48 < JshWright> seriously... 13:48 < JshWright> shoulda used json ;) 13:48 < vblank> all for a pager number. and he didn't think it was unusal 13:48 < adj> haha, sounds like he was hinting that he didnt want to be bothered after hours 13:48 -!- elfgoh [~dingding@adsl30.dyn212.pacific.net.sg] has quit [Ping timeout: 480 seconds] 13:49 < vblank> so yeah, he would return a nice big juicy xml document that I finally managed to parse with perl after a quick references refresher, 13:50 < vblank> but I was so finished with that method we just used batch files to query the sql server and ftp to me...lol. 13:50 < mikegrb> lolz 13:50 < vblank> talking to ms sql directly from unix sux 13:50 < vblank> with perl, at least 13:51 < vblank> so yeah, very cool kid with his 'web service', lol 13:51 < mikegrb> lolz 13:52 < adj> why didnt you write a quick web page to pull the xml, parse it with xpath and format it? 13:52 * adj ducks 13:52 -!- lxsg [dcff07e8@webchat.mibbit.com] has joined #linode 13:52 -!- supine [~marty@merboo.mamista.net] has quit [Ping timeout: 480 seconds] 13:52 < vblank> adj: i just want a freaking pager number man, it's in a table 13:52 < lxsg> erm, how do i choose the VPS DC upon signup? 13:52 -!- mcinerney [~jsmith@peter.mcinerney.id.au] has quit [Ping timeout: 480 seconds] 13:52 < JshWright> lxsg: you do that when you create your linode 13:53 < adj> vblank: and the 'cool kid' decided (like a good admin) that giving out direct DB access is a Bad Thing, so he gave you an API 13:53 < lxsg> oh ok 13:53 < lxsg> thanks 13:53 < vblank> oh I have direct access, im on the same team with the dba's 13:53 < vblank> we went direct in the end with .bat files 13:53 < adj> not the point. 13:54 < adj> i digress. i don't know the whole situation. but i'm a huge fan of restricting DB access. 13:54 < vblank> my point is I shouldn't have been messing with soap and xml to figure out, programatically, who is on call 13:54 < adj> one day the db will move, and your stuff will break, requiring changes to all of it, instead of modifying one or two lines in the API 13:54 -!- loxs [~loxs@85-130-31-93.2073154325.ddns.cablebg.net] has joined #linode 13:56 -!- lxsg [dcff07e8@webchat.mibbit.com] has quit [] 13:56 -!- techman224 [~techman22@wnpgmb1316w-ds01-226-237.dynamic.mts.net] has quit [Remote host closed the connection] 13:56 < vblank> and I would have preferred lamp 13:56 < waldron> I'm OK not being a cool kid. The guide I found didn't work, need to try to find another 13:57 < vblank> wow that was fast 13:57 -!- Twayne [~waynemilt@cpe-071-070-201-028.nc.res.rr.com] has joined #linode 13:57 < waldron> http://www.howtoforge.com/installing-apache2-with-php5-and-mysql-support-on-debian-lenny-lamp 13:57 < vblank> waldron - start with one component. maybe lock your system down with a firewall first. then move on to simple one site apache config. 13:57 < vblank> then add mysql 13:57 < waldron> For example, didn't get the prompt for mysql root pw 13:57 < vblank> then add php 14:00 -!- zack_ [~zack@c-98-210-109-85.hsd1.ca.comcast.net] has quit [Quit: zack_] 14:03 -!- lakin [~lakin@S01060013101832cd.cg.shawcable.net] has quit [Quit: Ex-Chat] 14:03 -!- memenode [~libervisc@93-138-115-175.adsl.net.t-com.hr] has quit [Read error: Connection reset by peer] 14:03 -!- liberfiasco [~libervisc@93-141-75-79.adsl.net.t-com.hr] has joined #linode 14:03 -!- axod [56970259@webchat.mibbit.com] has quit [Quit: http://www.mibbit.com ajax IRC Client] 14:04 -!- lakin [~lakin@S01060013101832cd.cg.shawcable.net] has joined #linode 14:08 -!- TheJoe|ZzZz is now known as TheJoe 14:08 < JshWright> waldron: did you try the linode libary guide? 14:08 < JshWright> s/libary/library/ 14:09 < waldron> OK, I'm gonna try the linode how-to 14:09 < waldron> Hope I didn't pollute the whole thing with those other installs... 14:10 < waldron> what do you think?? 14:10 < waldron> prolly i could just reinstall the whole distro at this point, if I had to 14:11 -!- techman224 [~techman22@wnpgmb1316w-ds01-226-237.dynamic.mts.net] has joined #linode 14:11 < JshWright> waldron: that'd be my reccomendation 14:12 < waldron> mkay, I'll try that 14:12 < JshWright> http://library.linode.com/getting-started/ 14:12 < JshWright> then: http://library.linode.com/lamp-guides/debian-5-lenny/ 14:13 -!- lxsg [dcff0771@webchat.mibbit.com] has joined #linode 14:13 < lxsg> sorry to interrupt again, but may i ask is "disk image" in linode mean a disk partition? 14:14 < vblank> i think a disk image is a disk image 14:14 < vblank> you can partition your disk image 14:15 < lxsg> so you mean it looks like this in a hierarchy - Physical HDD > Disk Images > Disk Partitions? 14:16 -!- Bass10 [~Bass10@c-76-113-194-7.hsd1.mn.comcast.net] has joined #linode 14:16 < Guspaz|m> The disk images you create on linodes are virtual devices that are directly formatted with a filesystem (no partition table). 14:16 < Guspaz|m> You can partition them, but you shouldn't. 14:16 < Guspaz|m> Because you should just create different disk images. 14:16 < vblank> when you create a disk image, it is the disk, from the view of the linode 14:16 < Guspaz|m> Partitioning is a method of splitting a physical disk into logical volumes. But when you can create arbitrary volumes... 14:17 < Guspaz|m> Partitioning a linode disk image doesn't make any sense. 14:17 < vblank> i don't know that we were talking about doing it, just defining them 14:17 < straterra> HoopyCat: you around? 14:17 < Guspaz|m> Well, they can be defined as disks, not partitions. 14:18 < lxsg> ok, so i can just treat them as partition (though technically it's not, as you guys have said) 14:18 < Guspaz|m> No. 14:18 < Guspaz|m> Treat them as different disks. 14:18 < Guspaz|m> It's like you can just magically conjure into existance a disk of any size you like. 14:18 < lxsg> ok i'll treat them as individually physical HDDs in my mind 14:18 < vblank> yes, thats the idea 14:18 < lxsg> ok thanks guys! 14:18 < tarpman> lxsg: treat them as something you can make a filesystem on 14:18 < tarpman> lxsg: and don't worry about the details :P 14:20 < azaghal> What are people's experiences using Linode VPS for web/e-mail hosting? Particularly regarding disk usage? 14:20 < Guspaz|m> azaghal: Disk usage in what way? As in, does it have enough? 14:20 < Guspaz|m> It depends on what you're hosting. 14:20 < waldron> I did standard Debian last time, I'm thinking I'll go 64 bit, though, if someone in here tells me 64 bit addressing available in my virtual server 14:20 < Guspaz|m> waldron: It is, but it's pointless. 14:21 < azaghal> Guspaz|m: Mostly smaller companies, I think, but keep in mind many users _love_ to send big files through e-mail. 14:21 < vblank> vps is about root access. 14:21 < azaghal> I've just been looking at some web hosting companies, and they seem to offer lots of space to users. 14:21 < Guspaz|m> waldron: Unless you have a Linode with more than 4GB of RAM, don't bother. All it will do is waste memory without providing any real benefits. 14:21 < vblank> if you don't need it, you don't need a vps for hosting 14:21 < waldron> Good advice, thanks Guspaz 14:22 < azaghal> (I'm thinking of starting (yet another) web hosting company, though many thing would be providing site building). 14:22 < linbot> New news from forums: Munin shows HUGE slow MySQL queries - nothing in logs in Performance and Tuning 14:22 < azaghal> *main thing 14:22 < Guspaz|m> azaghal: If you want to manage the platform yourself, then Linode is an excellent high-performance scalable solution. If you don't want to, then reselling shared hosting is for you. 14:23 < azaghal> I want to mane it myself. I already have one VPS setup for my needs (Drupal, mail, jabber). 14:23 < azaghal> *manage 14:23 < azaghal> But I don't know how much hard-disk space small business users consume. 14:23 < JoeK> how much is a tld license? 14:23 < JoeK> D: 14:23 < Guspaz|m> Shared hosts can offer huge amounts of storage space because they're aggregating it. It can be oversold. If you sell 1TB of space to 100 people, and the average usage is 100GB, then you can probably get away with 20TB of space rather than 100TB of space. Or even less. 14:24 < Guspaz|m> Oh, I've no idea how much they consume. 14:24 < vblank> i would say less than a TB 14:24 < azaghal> Well, basically the only flaw to VPS is the HD space. 14:24 < azaghal> And it has many benefits, to be honest. 14:25 < JoeK> azaghal, i know a company that gives like 300gb space 14:25 < JoeK> for 30$ a month 14:25 < Guspaz|m> Well, more like Linode charges a lot more for disk space than others since they limit the number of drives in their hosts. 14:25 < JoeK> so, it depends on the company 14:25 < Guspaz|m> Or perhaps use hardware that only supports four drives, I don't know. 14:25 < dajhorn> azaghal: If you've never worked a medium or large email deployment, you'll be surprised at usage. And people will *instantly* complain if email goes down. 14:25 < JoeK> Guspaz, im in the webhosting business 14:25 < lxsg> erm one more question here, any way to download/backup a complete disk image? 14:25 < azaghal> dajhorn: I've worked at company that does, I've seen it :) 14:25 < JoeK> what you do is, if you run out of diskapce, and due to it being webhosting, you just reboot with more space (vpses in this case 14:26 < JoeK> so you can do lots of space for users 14:26 < azaghal> Uses don't get it that e-mail is not for file transfers... 14:26 < azaghal> *Users 14:26 < Guspaz|m> JoeK: At $2/GB, that's not really financially prudent. 14:26 < JoeK> not for linode 14:26 < JoeK> silly 14:26 < JoeK> im on a company which gives out DA and installatron included with vps plans(wont mention names) 14:27 < azaghal> DA? 14:27 < Guspaz|m> Unless you mean add more VPSes to your hosting cluster, then sure, that's economical at Linode. 14:27 < JoeK> but since its only webhosting you can upgrade disks without anybody knowing, and just claim it as maintence 14:27 < JoeK> direct admin, azaghal 14:27 < azaghal> Ah 14:27 < azaghal> To be honest, I'll offer deployment of servers at company's premises as well, hopefully for those who need lots of space. 14:33 < Guspaz|m> That makes management much more complex. 14:34 < azaghal> True, but I'd charge more for that. 14:34 < azaghal> And I'll setup NagiOS in any case. 14:35 < vblank> lo, uptime is 332 days for my node 14:37 < adj> nagios. the bane of my existance 14:38 < vblank> glroified pinger 14:38 < Guspaz|m> If a linode host fails, Linode staff will migrate you to another box for free and get you up and running again (assuming no data loss). If you're deploying hardware, you need to do it yourself. 14:38 -!- lxsg [dcff0771@webchat.mibbit.com] has quit [Quit: http://www.mibbit.com ajax IRC Client] 14:39 < vblank> pingers have their place though... 14:43 -!- Turl1 [~Turl@host26.190-224-48.telecom.net.ar] has joined #linode 14:43 -!- Turl is now known as Guest171 14:43 -!- Turl1 is now known as Turl 14:44 -!- Guest171 [~Turl@host26.190-224-48.telecom.net.ar] has quit [Ping timeout: 480 seconds] 14:45 -!- daMaestro [~jon@content.beatport.com] has quit [Quit: Leaving] 14:45 -!- daMaestro [~jon@content.beatport.com] has joined #linode 14:47 -!- johndbritton_ [~john@ool-4574433a.dyn.optonline.net] has joined #linode 14:47 -!- Smark[Gone] is now known as Smark 14:50 -!- SpaceHobo [~spacehobo@spacehobo.user.oftc.net] has quit [Ping timeout: 480 seconds] 14:50 -!- dajhorn [~dajhorn@transmisor.vanadac.com] has quit [Remote host closed the connection] 14:51 -!- dajhorn [~dajhorn@206.16.96.160] has joined #linode 14:52 -!- johndbritton [~john@ool-4574433a.dyn.optonline.net] has quit [Ping timeout: 480 seconds] 14:52 -!- johndbritton_ is now known as johndbritton 14:57 -!- johndbritton [~john@ool-4574433a.dyn.optonline.net] has quit [Quit: johndbritton] 14:58 < adj> nagios is waaaay more than a pinger 14:58 < adj> especially if you combine it with webinject 14:59 < vblank> our network team uses it. 14:59 < vblank> I like WhatsUP from Ipswitch 14:59 < rainman`> nagios is awesome 14:59 < rainman`> takes a little bit of effort to configure 14:59 < adj> whatsup can't handle anything remotely close to nagios though 14:59 < waldron> So... I got "Unable to fetch some archives, maybe run apt-get update or try with --fix-missing?" 15:00 < adj> try making it smart enough to monitor ospf links 15:00 < waldron> This happened last time, too... 15:00 < adj> impossible =) 15:00 < vblank> Well out of the box it can do services, up down, any windows metric, etc 15:00 < waldron> I'll try with the recommendation 15:00 < rainman`> at work i write loads of custom checks for nagios too 15:00 < adj> rainman`: same 15:00 < adj> whatsup flap detection blows too 15:00 < vblank> personally, for the unix team, we use IBM Tivoli Monitoring 15:00 < rainman`> for private, my nagios is pretty simple 15:01 < adj> but for a rapid monitor its nice and simple 15:01 < adj> tivoli is great, but pricey 15:01 < vblank> nah, i think we only paid 70k 15:01 < vblank> lolz 15:01 < adj> :P 15:01 < adj> imho, nagios could compete with tivoli if ethan opened up the vcs a bit more 15:01 < rainman`> don't forget the hours required to set up any solution too 15:01 < adj> he still requires patches to be sent to his email list! 15:01 < rainman`> free software isn't free :) 15:02 < vblank> it didn't come with with alerting, you have to roll your own 15:02 < vblank> it can run a command when a situation is true. our command is typically page the oncall person. 15:04 < vblank> The biggest thing for us is that it understands pSeries, processor entitlement, Virtual I/O, hardware management consoles, etc. 15:04 < adj> how do you like that pseries ;) 15:05 < vblank> once you go aix, you never go back, lol 15:05 < mikegrb> lolz 15:05 < Guspaz|m> a/ux ftw 15:05 < adj> because you can't? 15:05 < adj> hehe 15:05 < vblank> IT JUST WORK (TM) 15:05 < vblank> unlike my keyboard today 15:06 < adj> anyone know what it takes to reload ip_conntrack hashsize without a reboot? 15:06 < Guspaz|m> I wish I could have played with A/UX when I was growing up with 68k macs. 15:07 < adj> i'm dropping packets when i do more than about 4000 conncurrent http connections on my linode360 ;) 15:07 < vblank> oh yeah, they did that apple server with aix once... 15:07 < Guspaz|m> vblank: No, A/UX predates that. 15:07 < vblank> but a/ux is apple's right 15:07 < Guspaz|m> A/UX was Apple's unix distro from 1988 - 1995 15:08 < waldron> setting this thing up having no unix skills is a pain in the ass! 15:08 < Guspaz|m> It was System V Unix and had TCP/IP, and had the Mac (System 7) UI. 15:08 < waldron> using vi for first time. nobody should ever use vi for the the first time 15:08 -!- Redgore2 [~redgore@94-193-27-221.zone7.bethere.co.uk] has joined #linode 15:08 < Guspaz|m> So compared to macs at the time, A/UX would have been awesome. 15:08 < waldron> people who know vi should use vi. people who've never used vi should hire the people who know vi 15:09 < adj> waldron: try using nano 15:09 < vblank> Funny how things come full circle though... 15:09 * adj is a vim junkie 15:09 < Guspaz|m> Sadly, my mac didn't have an FPU. 15:09 < tarpman> waldron: try vim -y ('easy mode') 15:10 < Guspaz|m> TBH I don't know why Apple farted around with Copeland for so many years before just buying NeXT. They could have just moved A/UX mainstream much like Microsoft did with NT in Win2K/WinXP. 15:11 < vblank> Might be fun to aquire an se/30 and see it in action... 15:11 < vblank> according to the wiki article it would run on them 15:11 < Guspaz|m> If it has an FPU and a PMMU, I don't see why not. 15:12 < Guspaz|m> I had a MC68LC040, so I was out of luck 15:12 < Guspaz|m> I couldn't really run Linux either. 15:12 < Guspaz|m> This is a good site to see what it looks like BTW: http://www.applefritter.com/ui/aux/index.html 15:13 < Guspaz|m> It's like a modern Linux distro, merging a *nix backend and a usable UI, except two decades earlier :P 15:13 < vblank> looks pretty freaking cool 15:13 < adj> i suppose if my web server is crapping out at 20k connections in <5 seconds its probably ok 15:13 < adj> without a hardware firewall i dont think i can do much better 15:14 -!- K-Zodron [~K-Zodron@irc.zezeniaonline.com] has quit [Read error: Connection reset by peer] 15:14 -!- K-Zodron [~K-Zodron@irc.zezeniaonline.com] has joined #linode 15:14 < Guspaz|m> adj: You're dropping packets, not refusing connections? So, it's not a file descriptor limit? 15:14 < waldron> For those of you following the play-by-play... had to run apt-get update before apt-get install apache2 15:14 < adj> Guspaz|m: i have a huge FD limit set 15:14 -!- daMaestro [~jon@content.beatport.com] has quit [Quit: Leaving] 15:14 < adj> Guspaz|m: ip_conntrack's table is maxing out 15:14 < Guspaz|m> That can be increased. 15:15 < adj> but i dont think its worth raising the limit at the cost of ram on a 360 15:15 < adj> i already dropped the bucket size to 8 15:15 < adj> from cent's default of 16 15:15 < Guspaz|m> Well, if it's dropping packets, what else can you do other than upgrade the linode? 15:15 < Guspaz|m> err 15:15 < Guspaz|m> other than get a second linode 15:15 < adj> i'm just benching 15:15 -!- Redgore [~redgore@94-193-27-221.zone7.bethere.co.uk] has quit [Ping timeout: 480 seconds] 15:15 -!- Redgore2 is now known as Redgore 15:15 < Guspaz|m> Oh. 15:15 < adj> tbh, 4k concurrent on a 360 is pretty awesome, imho 15:15 -!- Smark is now known as Smark[Gone] 15:16 < JshWright> waldron: didn't the error message tell you to run apt-get update? 15:16 < Guspaz|m> Strangenesses: A/UX apps that used X would have per-window menus due to the conflicting paradigm between Mac (top of screen) and X (per-window) 15:17 < straterra> HoopyCat: Wakey wakey 15:18 -!- agentbleubleu [~agentbleu@lns-bzn-40-82-251-179-229.adsl.proxad.net] has joined #linode 15:18 < Guspaz|m> Yay for cheapy tuesdays! 15:21 < vblank> echo 131072 > /proc/sys/net/ipv4/ip_conntrack_max 15:21 < vblank> that helped someobody.. 15:21 < vblank> would have to do it on boot though 15:21 < linbot> New news from forums: Nginx Optimization Tips? in Performance and Tuning 15:22 < adj> vblank: odd. i wonder why they chose such a random number 15:22 -!- Bdragon|work [~bdragon@dpc6746139138.direcpc.com] has quit [Read error: Connection reset by peer] 15:22 < Guspaz|m> 131 thousand connections is a bit excessive. 15:23 -!- Bdragon|work [~bdragon@dpc6746139138.direcpc.com] has joined #linode 15:23 -!- agentbleubleu [~agentbleu@lns-bzn-40-82-251-179-229.adsl.proxad.net] has left #linode [] 15:23 < dajhorn> QuickDraw and important parts of the toolbox weren't threadsafe. They never did really fix that until Classic, and that was a sandbox. (Anybody got A/UX 3 media? I'll buy it.) 15:24 * dajhorn hugs his SE/30 15:24 < Guspaz|m> dajhorn: What's harder, writing a new OS from scratch (Copeland) or fixing QuickDraw. 15:24 < Guspaz|m> I mean, they ended up abandoning QuickDraw eventually anyhow. 15:24 < Guspaz|m> Could have just introduced a new graphical API. 15:25 < dajhorn> Guspaz|m: Yeah, and they did. Retrospect says that A/UX was a decade too soon. No gumption for that kind of major change. 15:25 < dajhorn> (Back then.) 15:25 < Guspaz|m> Unfortunate. But I'll note that support for it ended when Copeland was under development. 15:26 < Guspaz|m> And A/UX ran mac apps, it wasn't just Unix with the mac UI. 15:28 -!- zack_ [~zack@64-71-16-254.static.wiline.com] has joined #linode 15:30 < vblank> they just doubled their default, 65536 15:31 < Guspaz|m> What are they doing that needs more than 65536? 15:31 < vblank> http://rackerhacker.com/2008/01/24/ip_conntrack-table-full-dropping-packet/ 15:32 < vblank> very informative 15:32 < vblank> i guess iptables is the culprit 15:33 -!- SpaceHobo [~spacehobo@78-105-8-188.zone3.bethere.co.uk] has joined #linode 15:33 -!- v0lksman [~shayne@ottawa-hs-64-26-169-151.s-ip.magma.ca] has quit [Ping timeout: 480 seconds] 15:33 -!- SpaceHobo [~spacehobo@spacehobo.user.oftc.net] has quit [] 15:33 < Guspaz|m> It doesn't make any sense. He says he has a server under low load, but then says it's under heavy load... 15:33 < waldron> I think i'm gonna give up and let a pro do this 15:33 -!- SpaceHobo [~spacehobo@spacehobo.user.oftc.net] has joined #linode 15:34 < Guspaz|m> And that it was handling a high volume of net connections, but he doesn't say for what. 15:34 < vblank> idle clients... 15:34 < Guspaz|m> And: "Generally, the ip_conntrack_max is set to the total MB of RAM installed multiplied by 16. However, this server had 4GB of RAM, but ip_conntrack_max was set to 65536" 15:34 < Guspaz|m> You'll note what 4096 * 16 is... 15:35 < adj> yes. iptables is the performance bottleneck 15:35 < Guspaz|m> Unless you're under a DDoS attack or incredible load, it shouldn't be an issue. 15:36 < adj> right 15:36 < adj> i'm stress testing a new setup 15:38 < linbot> New news from forums: RAM usage on fresh install of debian in Performance and Tuning 15:38 < adj> ahh. and upping that to 32768 works 15:39 -!- Darxus [~darxus@panic.chaosreigns.com] has left #linode [] 15:40 < adj> nice. it peaks around 26k in the conntrack count when doing 2k concurrent and 50k total connections from apache bench 15:41 -!- kenichi [~kenichi@207.162.220.10] has joined #linode 15:43 < Guspaz|m> Are you actually expecting to handle that kind of load? 15:43 < linbot> New news from forums: Shared disk / synced disk between Linodes in Linux Networking 15:43 < Guspaz|m> I mean, you can get a lot out of a properly tweaked Linode, but... 15:43 < adj> Guspaz|m: not on a single box 15:44 < adj> i'm just testing a new setup to see how it performs on static content 15:44 < vblank> testing adj? you're fired. 15:44 < adj> and... it SCREAMS 15:44 < Guspaz|m> Only so long as the content is cached in RAM. 15:45 < straterra> Holy crap batman..my linode is 133 megs in to swap 15:45 < straterra> with..a lot of free memory 15:45 < Guspaz|m> That's normal. 15:45 < Guspaz|m> Turn down your swappiness if you don't like it. 15:46 < straterra> I didn't see it on my old kernel 15:46 < Guspaz|m> Was your old kernel from the 2.4 branch? 15:46 < straterra> No 15:46 < Guspaz|m> Perhaps you had an app suck up a bunch of RAM causing stuff to get swapped. 15:47 < adj> Guspaz|m: anything not dynamically generated will be in ram in a high load situation like this 15:47 < Guspaz|m> adj: Not if you have more diverse content than RAM. 15:47 < vblank> is swap on linux persistent ? once swapped in does swap decrease 15:48 < adj> Guspaz|m: the rev proxy will cache everything 15:48 < Guspaz|m> adj: Again, if you've got a gig of static content being loaded and 360MB of RAM, it can't all fit in RAM. 15:48 < adj> thats what i've been testing ;) how well a few hundred lines of VCL work in a high concurrency situation 15:48 < Guspaz|m> Also, what good is having a reverse proxy cache stuff in RAM? Shouldn't that be left to the disk cache? 15:49 < Guspaz|m> err 15:49 < Guspaz|m> what good is having a reverse proxy cache STATIC stuff in RAM. 15:49 < adj> no. it should be left in VM and the kernel should handle placing it where it needs to be 15:49 < Guspaz|m> And how is that different from letting the kernel decide if a given block needs to be read from disk or RAM? 15:50 < adj> Guspaz|m: http://varnish.projects.linpro.no/wiki/ArchitectNotes 15:50 -!- TheJoe is now known as TheJoe|ZzZz 15:50 < adj> that explains the concept a bit more completely than i can in IRC 15:52 < adj> Guspaz|m: needless to say, i have been *very* impressed with varnish. 15:54 < Guspaz|m> So, I read it... 15:55 < Guspaz|m> I don't agree with the author. 15:55 < adj> do you rev proxy anything? 15:55 < Guspaz|m> Effectively, his approach requires you to have all your data on the disk twice, with massive page files. 15:55 < Guspaz|m> The idea there is that, shove everything into virtual memory, let the kernel decide what to keep in RAM and what to page. 15:56 < adj> no no no. that would be innefficient. you won't really see much benefit (except for reducing the apache startup/teardown times) on a single host 15:56 < Guspaz|m> The problem with that is, if you have more content than RAM, some (much?) will get paged to disk, despite the fact that the data already lives on disk. 15:56 < adj> the benefit occurs when the rev proxy is placed in front of the web app cluster 15:56 < Guspaz|m> Modern operating systems will use all free RAM for the disk cache. So if you simply don't cache in RAM and jsut read everything from disk, the kernel will decide what to cache and what to leave on disk. In effect, the exact same thing, except no duplication. 15:57 < Guspaz|m> OK, so in that case, of a cluster, sure, you're not duplicating. 15:57 < vblank> you guys need more ram 15:57 < Guspaz|m> I'm talking strictly from a single-machine perspective. 15:57 < adj> Guspaz|m: ahh. i am not =) 15:57 < adj> Guspaz|m: i am just testing this on a single machine 15:58 < Guspaz|m> From a multi-machine perspective, it makes perfect sense. And you can allocate most of your disk to swap. If you've got a single machine, that architecture is silly and wasteful. 15:58 < adj> to work out the caveats and gotchas before a/b testing it in production against my existing squid accelerators 15:58 < adj> and varnish is blowing squid out of the water so far 15:58 < Guspaz|m> Can't you configure Squid to work in a vaguely similar manner? No disk cache, huge RAM cache (that is much larger than available RAM)? 15:59 -!- borris84 [~Lars@rrcs-96-10-93-114.se.biz.rr.com] has joined #linode 15:59 < Guspaz|m> Sure, it isn't designed with that paradigm in mind, so it'd be decidedly sub-optimal, but wouldn't it at least provide a gross approximation? 15:59 < adj> you can, but its not holding up as well as varnish is. without a bunch of systemtap scripts its hard for me to tell exactly why 15:59 -!- borris [~Lars@rrcs-96-10-93-114.se.biz.rr.com] has quit [Read error: Connection reset by peer] 16:00 < adj> let me dig up the link that first got me interested in switching from squid 16:00 < adj> http://www.bsdcan.org/2007/schedule/attachments/18-The_Varnish_HTTP_Accelerator_Poul-Henning%20Kamp 16:03 < vblank> You have to adjust the isolinear processors to dissipate the nucleogenic decay in the food replicators. 16:03 < vblank> (just found trenobabble generator 16:04 < Guspaz|m> LOL, I love the 16:04 < mikegrb> lolz 16:04 < Guspaz|m> LOL, I love the "allocation_modulus" variable on page 18 :P 16:04 < mikegrb> lolz 16:05 -!- kripken [~alon@CBL217-132-96-209.bb.netvision.net.il] has left #linode [Ex-Chat] 16:07 -!- TheJoe|ZzZz is now known as TheJoe 16:07 < TheJoe> Everytime there's a power cut, my router loses it's settings. Damn thing. 16:07 -!- ph_ [~ph^@cm-84.211.180.8.getinternet.no] has quit [Remote host closed the connection] 16:08 < adj> Guspaz|m: :P 16:09 < adj> Guspaz|m: plus the cli tools rock =) varnishtop is amazing 16:11 -!- ph_ [~ph^@cm-84.211.180.8.getinternet.no] has joined #linode 16:11 < Guspaz|m> Bah at the power savings of 12MWh/y, that's only like $800 here in Quebec :P 16:12 < adj> hehe 16:12 < Guspaz|m> Hmm, quite impressive. 16:12 < Guspaz|m> (finished presentatino) 16:12 < Guspaz|m> But the lack of GZIP compression is disappointing, and planned only for v2. 16:13 < adj> v2 is out =) 16:13 < Guspaz|m> Ah. 16:13 < Guspaz|m> And does it support gzip compression? :P 16:13 < adj> 2.0.4 is what i have, i think 16:13 < adj> let me check 16:14 < Guspaz|m> While my linode only admittedly runs one site with any significant load (and even then, only a million pageviews per month, so small), gzip works wonders. 16:14 < adj> yeah. it looks like gzip support was pushed to post-2.0 16:14 < Guspaz|m> Dang. 16:15 < adj> however, i also use gzip, and you can still let apache/lighty/nginx do that 16:15 < Guspaz|m> By putting it in front of the reverse proxy? 16:15 < adj> as long as the web server is writing the cache/expiration headers then the accel will cache the gzipp'ed objects no problem 16:16 < Guspaz|m> oh 16:16 < Guspaz|m> I do love me some gzip. Even though a lot of the content being served isn't (further) compressible, I still see almost 2:1 bandwidth savings at an almost unmeasurable increase in CPU usage. 16:17 < adj> the vcl i wrote honors the cache header info from apache first and foremost, then tries to guess based on file type, request method, etc whether or not to cache the obj and for how long 16:17 < adj> wow. i need to test that further 16:17 < adj> make sure i'm zipping everything possible 16:17 < Guspaz|m> Well, text content compresses very well. 16:17 -!- adnc [~numer@p5485299C.dip.t-dialin.net] has joined #linode 16:18 < Guspaz|m> So if you've got a page that has 20K of HTML, it adds up fast. 16:19 < adnc> hello, i've a directory which i can not delete. an ls -la shows this: 16:19 < adnc> d????????? ? ? ? ? ? tmp 16:19 < adnc> has someone got any ideas how i could remove this? 16:20 < adnc> it happened after a mount try of a fat32 filesystem 16:20 < Guspaz|m> Of course, my traffic wasn't that high to begin with, I think I went from like ~200kbit/s to ~100kbit/s, but the traffic is bursty enough that it's hard to tell. So I'm judging that based on the drop when I enabled it. 16:20 < Guspaz|m> adnc: It usually means that your current user doesn't have permission to see the mount, I think 16:20 < metap|pe> ls with inodes (not file names) and rm the inode 16:20 < adnc> Guspaz|m: it is root 16:21 < adnc> metap ok 16:21 < adnc> chown: cannot access `tmp': Stale NFS file handle 16:21 < adnc> thats the error I get 16:22 < metap|pe> nfs? wtf 16:23 < adnc> well i tried to mount a fat32 disc onto /mnt/tmp and that was the result of it. i now have the inode of that directory, but rm has no option to delete via inode number 16:24 < metap|pe> oh, i think you have to use find 16:25 < adj> yep 16:25 < adj> find -inum n -exec rm -rf {} \; 16:25 < adj> n is the inode num 16:25 < adnc> thank you, find is cool 16:25 < metap|pe> may need to escape the {} 16:25 < Karrde> '{}' 16:26 < adnc> same result 16:26 < vblank> is something mounted? 16:26 < adnc> ses:/mnt# find -inum 46843 -exec rm -rf '{}' \; 16:26 < adnc> find: `./tmp': Stale NFS file handle 16:26 < adnc> something is mounted but not on that directory 16:27 < metap|pe> cat /proc/mounts any help? 16:27 < adnc> i've one disc mounted but that is on /mnt/music 16:27 -!- andrew_j_w [~andrew@88-97-29-168.dsl.zen.co.uk] has quit [Quit: Konversation terminated!] 16:27 < vblank> unmount tmp and fsck it 16:27 < Guspaz|m> YOu did try unmounting it, right? 16:27 < adnc> rootfs / rootfs rw 0 0 16:27 < adnc> none /sys sysfs rw,nosuid,nodev,noexec 0 0 16:27 < adnc> none /proc proc rw,nosuid,nodev,noexec 0 0 16:27 < adnc> udev /dev tmpfs rw,size=10240k,mode=755 0 0 16:27 < adnc> /dev/disk/by-label/ROOT_FS / ext2 rw,noatime,errors=continue 0 0 16:27 < adnc> tmpfs /lib/init/rw tmpfs rw,nosuid,mode=755 0 0 16:27 < adnc> varrun /var/run tmpfs rw,nosuid,mode=755 0 0 16:27 < Karrde> !pb 16:27 < adnc> varlock /var/lock tmpfs rw,nosuid,nodev,noexec 0 0 16:28 < adnc> tmpfs /dev/shm tmpfs rw,nosuid,nodev 0 0 16:28 < adnc> devpts /dev/pts devpts rw,nosuid,noexec,gid=5,mode=620 0 0 16:28 < adnc> tmpfs /tmp tmpfs rw,nosuid,nodev 0 0 16:28 < Karrde> !pastebin 16:28 < linbot> http://p.linode.com <-- paste here, not in the channel 16:28 < adnc> tmpfs /var/log tmpfs rw,nosuid,mode=755 0 0 16:28 < adnc> tmpfs /var/tmp tmpfs rw,nosuid,mode=755 0 0 16:28 < adnc> /dev/sda1 /mnt/music ext3 rw,noatime,errors=continue,data=ordered 0 0 16:28 < adnc> guspaz sure, but it didn't work, i did reboot 16:28 < Yaakov> Ouch. 16:28 < adnc> Karrde: sorry, didnt know 16:28 < Guspaz|m> /var/tmp is mounted by tmpfs... 16:28 < adnc> yes 16:28 < adnc> thats fine 16:29 < Guspaz|m> You tried to mount a fat32 filesystem on top of another file system? 16:29 < adnc> no 16:29 < adj> adnc: what does lsof | grep '/mnt/tmp' show? 16:29 < adnc> it was not /var/tmp it was /mnt/tmp 16:29 < Guspaz|m> Oh 16:29 < adnc> adj it gave nothing back 16:29 < adnc> empty 16:29 < Guspaz|m> umount as root with a -f? 16:29 < adnc> adj since i rebooted there shouldnt be any open files 16:30 < adj> can you umount /mnt/tmp ? 16:30 < adnc> /mnt/tmp is not mounted 16:30 < metap|pe> if you've rebooted and it's still there, you may need to fsck it 16:30 < Guspaz|m> fsk that ucker. 16:30 < Guspaz|m> errm 16:30 < Guspaz|m> fsck that ucker 16:31 < vblank> or 16:31 < vblank> You might be able to adjust the emission of chronoton particles from the dilithium matrix to reinitalise the nucleogenic decay in the phase transition coils. 16:31 < adnc> it is a compact flash on an embedded board 16:31 < Guspaz|m> That doesn't mean you can't fsck it. 16:31 < adnc> vblank: with or without fluxcompensation? 16:32 < vblank> well I thought that was implied :) 16:32 < adnc> guspaz yes right 16:32 < Guspaz|m> 1.21 JIGGAWATTS?!?!? 16:32 < adnc> vblank: ohh, sure, yes, i forgot 16:32 -!- huseyin [~huseyin@ces200.info] has joined #linode 16:33 < vblank> lol 16:33 < mikegrb> lolz 16:33 < huseyin> hello 16:33 < metap|pe> ehlo 16:33 < adnc> merhaba 16:33 < Guspaz|m> Bonjour 16:33 < adnc> ;) 16:33 -!- lxsg [dcff0783@webchat.mibbit.com] has joined #linode 16:33 < adnc> guten tag 16:33 < Guspaz|m> Hajimamashite (and I probably butchered that) 16:34 < Guspaz|m> Maybe that was supposed to be an e. 16:34 < lxsg> nice to meet you~ 16:34 < lxsg> ah 16:34 < lxsg> am back with another question 16:34 < lxsg> and yea it's an e as in "...memashite" 16:35 < bd_> 初めまして、どうぞよろしくお願いします 16:35 < lxsg> ok i gotta need google translate for that 16:35 < Guspaz|m> My Japanese isn't good enough to spout anything but half-rememebered phrases. 16:36 < Karrde> 私はガラスを食べられます。それは私を傷つけません。 16:36 < Guspaz|m> You eat glass? 16:37 < lxsg> heh u used google translate too 16:37 < straterra> this knoppix disk isn't utf8 enabled 16:37 < Guspaz|m> knoppix is still around? 16:37 < metap|pe> remember, only YOU can prevent non-utf8 knoppix disks. 16:38 < straterra> yes 16:38 < lxsg> ok, have anyone of you tried changing the SSH port for your linode? 16:38 < lxsg> i tried modifying this file: /etc/ssh/sshd_config 16:38 -!- waldron [~waldron@c-24-218-5-62.hsd1.ma.comcast.net] has quit [Quit: waldron] 16:38 < lxsg> uncommented #Port 22 16:38 < lxsg> to Port 2222 16:38 < straterra> You have to restart sshd 16:38 < lxsg> i tried rebooting from the linode manager 16:38 < lxsg> and 16:39 < lxsg> /etc/init.d/ssh reload 16:39 < lxsg> and i'm still on Port 22 16:39 < vblank> isn't that /etc/services 16:39 -!- waldron [~waldron@c-24-218-5-62.hsd1.ma.comcast.net] has joined #linode 16:39 < lxsg> oh? 16:39 < caker> no 16:40 < waldron> Hey, wanted to thank you guys for helping me with my pathetic attempts to setup a raw linux server 16:40 < caker> lxsg: grep Port /etc/ssh/sshd_config <-- what does that show? 16:40 < JshWright> waldron: you get it up and running? 16:40 < waldron> That means you, JshWright, and all others who helped me. 16:40 < lxsg> omg, it's showing Port 22 16:40 * caker bonks 16:41 < waldron> you'll be pleased to know I have a qualified ops guy helping me out 16:41 < lxsg> let me figure out what's going on 16:41 < waldron> My guy has apache up, and he says mysql, php apache module, plus python (because he's one of THOSE) 16:41 < waldron> @ 16:41 < tarpman> @ 16:42 < JshWright> mod_python or mod_wsgi? 16:42 < lxsg> oh damn, i edited the wrong config 16:42 < waldron> dunno. 16:42 < lxsg> i did it on /etc/ssh/ssh_config 16:42 < lxsg> thanks caker! 16:42 < waldron> OK guys, I'm out. Thanks again. 16:42 < metap|pe> all the cools kids are using mod_wsgi 16:43 -!- waldron [~waldron@c-24-218-5-62.hsd1.ma.comcast.net] has quit [] 16:43 < Guspaz|m> All the cool kids use netcat to run their web servers, then sit at a terminal manually typing HTTP replies. 16:43 < jed> ^ +1 16:43 < tarpman> it's the only way to develop fast fingers. 16:44 < JshWright> hey jed, congratz on the libcloud committer status 16:44 < jed> I have committer status? 16:44 < JshWright> I thought so... 16:44 < jed> nah, just got alex to commit 16:44 < jed> if he gave it to me it's news... 16:45 < JshWright> "Also, I've made oremj and jedsmith both committers on the cloudkick/libcloud branch." 16:45 -!- lakin [~lakin@S01060013101832cd.cg.shawcable.net] has quit [Quit: Ex-Chat] 16:45 < jed> hah 16:45 < jed> I didn't even know 16:45 < JshWright> unless there's another jedsmith... 16:45 < jed> thanks Jsh <3 16:46 < jed> thank you for pointing it out, and thank you for the congrats :) 16:46 < jed> <-- durp durp 16:46 < JshWright> you are welcome for both, sir 16:46 -!- vblank [~ecron@li-50.members.linode.com] has quit [Quit: ircII EPIC4-2.8 -- Are we there yet?] 16:47 -!- lxsg [dcff0783@webchat.mibbit.com] has quit [Quit: http://www.mibbit.com ajax IRC Client] 16:47 < huseyin> have got a very complex system 16:48 < adj> Guspaz|m: final numbers are in. apache w/o varnish on a 360 is doing 1001.49 [#/sec] (mean) w/ varnish is 1264.95 16:48 -!- Turl [~Turl@host26.190-224-48.telecom.net.ar] has quit [Quit: Leaving.] 16:48 < adj> so, not a huge impact on a single host 16:48 < adj> i'm interested what happens when i put it in front of a cluster doing 10mbps avg 16:49 < Guspaz|m> Unless it's caching dynamic content, I'm surprised that there's any advantage at all. 16:49 < Guspaz|m> Then again, Apache is a pig. 16:49 < adj> yeah. its the pig factor for sure 16:49 -!- daMaestro [~jon@content.beatport.com] has joined #linode 16:49 -!- loxs [~loxs@85-130-31-93.2073154325.ddns.cablebg.net] has quit [Quit: Leaving] 16:49 < adj> and i'm prefork, worker would probably be near identical 16:50 < Guspaz|m> I'd be curious to see what lighttpd/nginx would do 16:50 -!- huseyin [~huseyin@ces200.info] has quit [Quit: Leaving] 16:51 < Guspaz|m> Well, Lighttpd 1.5 with linux-aio-sendfile 16:52 < hcl2> xbox360 ? 16:52 < adj> hehehe, xbox =) 16:53 < Guspaz|m> Huh? 16:53 < hcl2> anyone interested in a vhost control panel that can manage nginx/lighthttpd setups? 16:54 < adj> only if its written in haskell 16:54 < adj> :P 16:55 < Guspaz|m> *shudder* 16:58 < Guspaz|m> One of these days, I should look into why nginx is killing off lighttpd, and see what all the hoopla is about. 17:00 < BarkerJr> can you use the 'ip' command to setupt 6to4? 17:02 < BarkerJr> I think I'm gonna order up another vps so I can play with ip and see what it does 17:07 -!- superkuh [~hukrepus@c-24-245-50-49.hsd1.wi.comcast.net] has quit [Read error: Connection reset by peer] 17:08 < JshWright> bah... how come the built in Django "serializer" modules only works on QuerySets...? I don't want to serializer a queryset, I want to serialize a model instance 17:09 -!- Kerem [~kerem@keremdurmus.com] has quit [Read error: Operation timed out] 17:10 < JoeK> error: command 'gcc' failed with exit status 1 <- anybody know that error when compiling something with python? D: 17:12 < JshWright> JoeK: not without some more context, no... 17:12 < JshWright> exit status 1 just means "it didn't work" 17:13 < linbot> New news from forums: Think this would work? - Multiple paying users on one Linode in Feature Request/Bug Report 17:15 -!- v0lksman [~shayne@ottawa-hs-64-26-169-151.s-ip.magma.ca] has joined #linode 17:17 -!- dajhorn [~dajhorn@206.16.96.160] has quit [Quit: ChatZilla 0.9.85 [Firefox 3.5.3/20090924122522]] 17:22 -!- goofykinky [~chatzilla@189.156.142.148] has joined #linode 17:25 -!- dzjepp [~menace2s@adsl-76-239-28-192.dsl.emhril.sbcglobal.net] has joined #linode 17:31 < Peng_> BarkerJr: Probably. 17:32 < Peng_> BarkerJr: Yes. [ref: http://tldp.org/HOWTO/Linux+IPv6-HOWTO/configuring-ipv6to4-tunnels.html ] 17:34 < BarkerJr> thx 17:41 -!- goofykinky [~chatzilla@189.156.142.148] has quit [Ping timeout: 480 seconds] 17:42 -!- linville [~linville@sapphire.tuxdriver.com] has quit [Quit: Leaving] 17:42 < Peng_> BarkerJr: A http://tunnelbroker.net/ tunnel would probably be faster. 17:42 -!- hercynium [~hercynium@c-98-216-53-107.hsd1.ma.comcast.net] has joined #linode 17:43 < BarkerJr> yeah, the problem is that they don't support rdns 17:43 -!- Redgore [~redgore@94-193-27-221.zone7.bethere.co.uk] has quit [Quit: Peace and Protection 4.22.2] 17:43 < BarkerJr> I'm using an HE tunnel now, but am seriously looking at switching to 6to4 17:44 < bob2> why 17:44 < straterra> BarkerJr: how don't they support rdns? 17:44 < bob2> of course HE support rdns 17:44 < straterra> I have rdns working with HE 17:44 < BarkerJr> see the forum thread 17:44 < straterra> which one? 17:44 < bob2> what forum thread 17:44 < BarkerJr> I posted on the forum that they don't support rdns 17:44 < bob2> then you posted a lie 17:44 < straterra> I'm using RDNS with them.. 17:45 < bob2> since they do 17:45 < Peng_> Which forum, HE's or Linode's? 17:45 < BarkerJr> http://www.linode.com/forums/viewtopic.php?p=23799#23799 17:45 < bob2> which is true, but unrelated 17:46 < bob2> they delegate it to the nameservers you ask them to 17:46 < bob2> for the subnet that you ask for 17:46 < BarkerJr> for the subnet, but not if you only want one ip 17:46 < straterra> BarkerJr: check it out..get the AAAA for projectstfu.com... 17:46 < straterra> Which is 2001:470:8a81::2 17:46 < straterra> check the reverse of that 17:46 < bob2> BarkerJr: so get a subnet 17:47 < bob2> BarkerJr: you can't delegate rdns for a single IP 17:47 < bob2> without insane cname trickery, anyway 17:47 < BarkerJr> in 6to4 you can 17:47 < bob2> "You shouldn't use your point-to-point address assigned by HE for day-to-day work. They are not clear in explaining this, but it makes sense once you're familiar with it. " 17:47 < straterra> Get the damn subnet..like you're supposed to get anyway..and set the reverse on that one ip 17:47 -!- internat [~nf@123-243-184-161.static.tpgi.com.au] has quit [Read error: Connection reset by peer] 17:47 -!- goofykinky [~chatzilla@189.156.142.148] has joined #linode 17:48 < Peng_> What's wrong with only using 1 IP out of your /64? 17:48 -!- internat [~nf@123-243-184-161.static.tpgi.com.au] has joined #linode 17:48 < straterra> Nothing 17:48 < Peng_> It's not actually any more difficult. 17:48 < Peng_> straterra: Right. 17:48 < BarkerJr> I prefer to use 1 ip out of my /48 17:48 < Peng_> Then you're wasting address space. 17:48 < straterra> So..use 1 ip 17:48 < Peng_> :P 17:48 < Peng_> Shy get a /48, then? 17:48 < straterra> Set up rdns for the /64 that you use 17:48 < Peng_> Why* 17:49 < BarkerJr> anyway, tell me why HE is better than 6to4 17:49 < bob2> no one cares what you use 17:49 < BarkerJr> I'm curious... haven't heard a good reason yet 17:49 < bob2> misinformation should be corrected, though 17:50 < BarkerJr> I see no misinformation 17:50 < straterra> uhm..not being able to do rdns with HE is misinformation 17:50 < bob2> 07:44:40 < BarkerJr> I posted on the forum that they don't support rdns 17:50 < bob2> ^ false 17:50 < BarkerJr> ok, the typical setup for HE can't do rdns 17:50 < Peng_> 6to4 is more dynamic. You have no idea where you'll be routed. With HE, it's consistent. 17:51 < bob2> they do, just not for a single ip, which they don't intend for you to use 17:51 < Peng_> BarkerJr: False. 17:51 < bob2> that's not the typical setup 17:51 < straterra> a whole subnet is a typical setup 17:51 < BarkerJr> HE gives you the configs for a single IP, not a subnet, so that is the typical setup 17:51 < bob2> don't think so 17:51 < straterra> They give you the config to get it up and going 17:51 < straterra> They aren't going to teach you everything about networking 17:51 < BarkerJr> or are you saying that the typical user doesn't use the config they are given? 17:52 < Peng_> s/everything/anything/ *cough* 17:52 -!- bob2 [rob@crumbs.ertius.org] has left #linode [] 17:52 < straterra> If someone has an IPv6 tunnel...its kinda assumed they should know what they are doing 17:52 -!- bob2 [rob@crumbs.ertius.org] has joined #linode 17:52 < BarkerJr> obviously not, or they wouldn't give that config out 17:52 < Peng_> 6to4 still gives you a /48, no? 17:52 < Peng_> BarkerJr: That config is just to get you started. It's not everything. 17:53 < Peng_> HE's shitty documentation doesn't change what the correct setup is. 17:53 < BarkerJr> but if getting you started gives you ipv6, why would you look further? 17:53 -!- bob2 [rob@crumbs.ertius.org] has left #linode [] 17:54 < Peng_> BarkerJr: Cuz you care about doing it correctly? And you want rDNS? 17:54 < straterra> ffs 17:54 < straterra> I'll host your damn rdns 17:54 < BarkerJr> the typical user wouldn't care 17:54 < straterra> the typical user wouldnt care about ipv6 17:56 < Peng_> The typical user doesn't even know about tunnelbroker.net. 17:56 < Peng_> Hahaha, my Linode's nearest 6to4 router (or whatever you call 192.88.99.1) is run by HE. 17:57 < BarkerJr> man, I want native ipv6 17:57 < BarkerJr> well, both tunnelbroker and HE's 6to4 are in fremont :) 17:57 < straterra> tunelbroker IS HE 17:58 < BarkerJr> so, should I have said "HE's tunnelbroker"? 17:58 < straterra> Peng_: I'm number 15 on the list of lowest latency ipv6 tunnels to the NY HE tserve 17:58 < Peng_> BarkerJr: Where's your 'node? 17:58 < BarkerJr> fremont 17:58 < BarkerJr> is it low on the latency scale? 17:59 < Peng_> Woah, jtsage is #1 and #11 on the Dallas tunnel server. I'm nowhere. :( 17:59 < Peng_> Even though I'm like 1 rack over from him. 17:59 < BarkerJr> I'm #11 in toronto 18:00 < straterra> Peng_: wanna know something crazy? To the tserve, my latency is lower over ipv6 than it is ipv4 18:00 < Peng_> Do you have to do anything special to place on that list? 18:00 < straterra> Nope 18:00 < BarkerJr> I was running a tracert to hoopycat over ipv6 and it was faster than ipv4, too 18:01 < Peng_> I've seen lower latency over IPv6 for something, but I forget what. Probably HoopyCat's Newark node. 18:03 < BarkerJr> does that mran HE gives higher priority to ipv6? 18:03 -!- chendo [~chendo@atfdrn.lnk.telstra.net] has joined #linode 18:04 < chendo> hey, can you transfer linodes between accounts? 18:04 -!- Guspaz|m [cffdca03@webchat.mibbit.com] has quit [Quit: http://www.mibbit.com ajax IRC Client] 18:04 < Peng_> BarkerJr: In my case, the IPv4 route was very different. 18:04 < Peng_> I can't ping my "Server IPv6 address". That's not supposed to be the case, right? 18:05 < BarkerJr> dunno 18:05 < BarkerJr> chendo: I don't know, but you might as well open a ticket and ask 18:05 -!- Harvey [~45f93086@webuser.linode.com] has joined #linode 18:06 < Harvey> hey guys 18:06 < chendo> okay 18:06 < BarkerJr> hi Harvey 18:06 < Harvey> how do I find out how many cores my linode 360 has? 18:06 < chendo> cat /proc/cpuinfo 18:06 < BarkerJr> linodes have 4 virtual CPUs, and their hosts have 8 cores 18:06 < Peng_> Unless you're on an old UML box. 18:08 < Harvey> thanks guys 18:08 -!- ph_ [~ph^@cm-84.211.180.8.getinternet.no] has quit [Remote host closed the connection] 18:09 < BarkerJr> np 18:11 < Harvey> so if i need to enter how many daemon processes should run (for mod_wsgi in django) then i should enter 4? 18:11 < BarkerJr> yeah 18:12 < Harvey> okay thanks 18:18 -!- megatron27 [~firdaus@124.13.187.16] has joined #linode 18:19 -!- Bohemian [~Bohemian@32.140.250.179] has joined #linode 18:23 < BarkerJr> you know the root of all this confusion is, straterra? I didn't notice that the client ipv6 is not part of the routed /64 :( 18:23 < Peng_> OK, so can any of you ping6 your "Server IPv6 address"? 18:23 < Peng_> BarkerJr: I did that too. I set up rDNS for the wrong zone. Kind of a pain to fix, too. 18:23 < BarkerJr> yeah 18:24 -!- sc0field [~rajiv@201.82.94.122] has quit [Ping timeout: 480 seconds] 18:24 < BarkerJr> I can ping 2001:470:1f04:3c4::1 fine 18:24 < straterra> Peng_: yeah.. 18:25 < straterra> I set up my ipv6 border firewall for my /48 today 18:25 < Peng_> Weird that I can't. Wonder what I screwed up? 18:26 < megatron27> damn, who the hell specifies a character limit in a summary, usually it's a word limit :| 18:26 < megatron27> I read it as word limit :| 18:26 < megatron27> so now I have to cut cut cut 18:27 < Peng_> But that does explain why I'm not on the lowest-latency tunnel list. (Although even #20 is <2 ms, so I might not get on it anyway.) 18:27 < straterra> Sweet.. 18:27 < straterra> Only like 200 more GB left of this DD 18:28 < megatron27> hungry 18:30 < BarkerJr> well, I'm not on the low-latency list, either 18:33 -!- PHPdiddy [~johns@comp.stedwards.edu] has quit [Quit: Leaving.] 18:35 -!- Harvey [~45f93086@webuser.linode.com] has quit [Remote host closed the connection] 18:44 -!- TheJoe is now known as TheJoe|ZzZz 18:46 < straterra> http://twitpic.com/jnggv 18:47 < BarkerJr> hehe 18:47 * Peng_ bangs his head on the wall. I hate IPv6 tunnels. :( 18:48 < BarkerJr> I decided I really don't care about rdns, anyway 18:50 < megatron27> what happened to that organization that wanted to stimulate IPv6 adoption using pr0n 18:50 < thornheart> I am creating a custom Linux install for a Linode 360. Should I use LVM or a default partition setup? 18:50 < Peng_> megatron27: http://www.ipv6experiment.com/ ? 18:50 < Peng_> megatron27: I dunno. It was really gonna come online this time, he swears, and then it vanished. 18:50 < Peng_> (I'm sure he didn't really swear.) 18:51 < megatron27> that site is taking forever to load up on my connection 18:51 < Peng_> megatron27: Yeah, I can't connect. Or if I can, it's really slow. 18:51 < megatron27> but i think that's the one 18:53 < dioz> why not use he.net 18:53 < dioz> tunnelbroker 18:56 -!- hfb [~hfb@pool-96-247-114-183.lsanca.dsl-w.verizon.net] has quit [Quit: Leaving] 18:56 -!- adnc [~numer@p5485299C.dip.t-dialin.net] has quit [Quit: bye] 18:57 < Peng_> dioz: What do you mean? 18:59 < dioz> what is the issue you're having? 18:59 < BarkerJr> how can my additional ip have a different gateway? 19:00 < Peng_> I can't ping my tunnel server over IPv6. 19:00 < Peng_> "Address unreachable". 19:00 < dioz> using what? 19:00 < dioz> ping6 ? 19:00 < Peng_> Indeed. 19:00 < dioz> spam your ifconfig 19:01 < dioz> while i smoke a cig 19:01 < dioz> bbs 19:02 < Peng_> dioz: http://cheezum.mattnordhoff.com/tmp/interfaces . The tunnel otherwise works, though. 19:02 -!- supine [~marty@115.128.34.71] has joined #linode 19:03 < Peng_> I'm really not smart enough to be allowed to configure this. :P There are 17 different ways to do it, and I don't know the differences. 19:03 < BarkerJr> so, is setting up my second ip as simple as "ip address add 74.207.247.89/24 dev eth0" ? 19:07 -!- zack_ [~zack@64-71-16-254.static.wiline.com] has quit [Quit: zack_] 19:08 -!- sc0field [~rajiv@201.82.94.122] has joined #linode 19:09 -!- zack [~zack@64-71-16-254.static.wiline.com] has joined #linode 19:09 -!- zack [~zack@64-71-16-254.static.wiline.com] has left #linode [] 19:09 -!- zack [~zack@64-71-16-254.static.wiline.com] has joined #linode 19:11 < dioz> that's weird 19:12 < dioz> i do mine a bit different 19:12 < Peng_> Yeah, like I said, there are a lot of variations. 19:12 < dioz> ifconfig sit0 up 19:12 < dioz> ifconfig sit0 inet6 tunnel ::endpoint 19:12 < dioz> ifconfig sit1 up 19:12 < HoopyCat> the more you run over a raccoon, the flatter it gets 19:13 < dioz> ifconfig inet6 add clientipvd/64 19:13 < dioz> route -A inet6 add ::/0 dev sit1 19:13 < HoopyCat> Peng_: fremont<->newark tends towards a slightly lower ipv6 latency vs. ipv4 19:14 < HoopyCat> straterra: what? 19:14 < Peng_> Blaah ifconfig. 19:14 < dioz> well then 19:14 < straterra> HoopyCat: You're too late! 19:14 < dioz> modprobe ipv6 19:14 < Peng_> :P 19:14 < dioz> if you have ipv6 mod there's another way to do it 19:15 < Peng_> I don't have the ipv6 module. I use the Linode kernel, so it's built-in. 19:15 < Peng_> HoopyCat: <3 19:15 < Peng_> HoopyCat: Can you ping6 your HE "Server IP address"? 19:15 < HoopyCat> straterra: aww dang 19:15 -!- Bohemian_ [~Bohemian@32.140.250.179] has joined #linode 19:15 < HoopyCat> Peng_: yes 19:15 -!- Bohemian [~Bohemian@32.140.250.179] has quit [Read error: Connection reset by peer] 19:15 -!- Bohemian_ is now known as Bohemian 19:16 < Peng_> Dang. What is wrong with my setup? 19:21 -!- User12482 [~User82934@70.134.68.122] has joined #linode 19:23 < dioz> why do you have 19:23 < dioz> up ip addr add 2001:470:1f0f:b4::123:1/128 dev he-ipv6 19:23 < dioz> that isn't part of your router /64 19:23 < Peng_> Yes it is. 19:24 < Peng_> 2001:470:1f0f:b4::/64 is my routed /64. 19:24 < dioz> yeah i just saw that 19:24 < dioz> i should read 19:24 < BarkerJr> y'know what? I'm gonna setup both he and 6to4 on my server :) 19:25 < HoopyCat> Peng_: oh wait, i thought you said ping, not ping6. no, i appear to get address unreachable. 19:25 < SelfishMan> !urmom 19:25 < linbot> SelfishMan: Yo momma's so slow, she got beat by both the Hare and the Tortoise (803:0/7) [rmoum] 19:25 -!- Bohemian [~Bohemian@32.140.250.179] has quit [Quit: Bohemian] 19:26 < dioz> ip tunnel add ipv6 mode sit remote ENDPOINT local LOCALIP ttl 225 19:26 < Peng_> HoopyCat: Oh, cool, I'm not alone. I wonder what's wrong, though? 19:27 < SelfishMan> What can't you ping? 19:27 < HoopyCat> Peng_: dunno. i've never tried before, though. :-) 19:27 < SelfishMan> I can ping both my IPv6 tunnel IP and the server 19:27 < Peng_> SelfishMan: You can ping6 the tunnel server? 19:27 < SelfishMan> s/ping/ping6/ 19:28 < Peng_> SelfishMan: Well, we can't. Address unreachable. 19:28 < SelfishMan> 64 bytes from 2001:470:1f0e:19e::1: icmp_seq=1 ttl=64 time=1.06 ms 19:28 -!- megatron27 [~firdaus@124.13.187.16] has quit [Quit: Ex-Chat] 19:28 < Peng_> Who in the world thought networking computers was a good idea? :P 19:28 < SelfishMan> !mtr 2001:470:1f04:ac2::1 19:28 < linbot> mtr combines the functionality of traceroute and ping into one easy to use tool, and the output can be useful for determining where the source of a problem is. It can be downloaded from http://www.bitwizard.nl/mtr/ or http://winmtr.sourceforge.net/ for Windows. MTR summaries can be retrieved in-channel using the command !mtr-CITY where CITY is fremont, atlanta, newark or dallas. 19:28 < SelfishMan> dammit 19:29 < HoopyCat> Peng_: i believe it to be a routing thing on my end, as the unreach comes from my end 19:29 < HoopyCat> PING 2001:470:1f06:f41::1(2001:470:1f06:f41::1) 56 data bytes 19:29 < HoopyCat> From 2001:470:1f06:f41::2 icmp_seq=1 Destination unreachable: Address unreachable 19:29 < SelfishMan> !mtr-fremont 2001:470:1f04:ac2::1 19:29 < linbot> SelfishMan: [mtr] 2001:470:1f04:ac2::1: 1 hops, no loss, last hop average RTT was 0.9ms (These results brought to you by urmom and the number 42) 19:29 < Peng_> From 2001:470:1f0e:b4::2 icmp_seq=1 Destination unreachable: Address unreachable 19:29 < Peng_> HoopyCat: I think you based your config on mine? 19:30 < Peng_> Or not. I don't remember. 19:30 < SelfishMan> For the record, I can ping 2001:470:1f0e:b4::2 19:30 < HoopyCat> Peng_: or the other way around, one of the two :-) 19:30 < HoopyCat> frankly, i don't care enough to care about it 19:30 < dioz> up ip route add ::/0 via ::216.218.224.42 dev he-ipv6 19:31 < dioz> that is what seems weird to me 19:31 < HoopyCat> that's, what, 1/2^128 of the IPv6 internet i can't reach? 19:31 < dioz> in your little scriptor 19:31 < SelfishMan> up /sbin/ip route add ::/0 dev $IFACE 19:31 < Peng_> HoopyCat: I originally tried yours, but it didn't work. After mine was working, you fixed yours, but I don't remember what happened next. 19:31 < Peng_> s/what happened next/the details/ 19:31 < BarkerJr> here's my script now: http://wiki.barkerjr.net/tmp.txt 19:32 -!- River_Rat [~me@174-22-133-200.clsp.qwest.net] has joined #linode 19:32 < Peng_> I have too many things on my mind. Firefox just crashed, too. 19:32 < BarkerJr> :( 19:34 < Peng_> I wonder what'll happen to my IPv6 connections if I quickly restart the interface? 19:34 < HoopyCat> Peng_: either nothing or something 19:34 -!- mawolf [~mw@189.146.15.89] has quit [Quit: mawolf] 19:34 -!- RiverRat [~me@97-112-147-130.clsp.qwest.net] has quit [Read error: Operation timed out] 19:34 < Peng_> HoopyCat: Or something else! 19:34 < SelfishMan> Peng_: if you bounce the interface it will kill it 19:35 -!- silverblade [~silverbla@cust116-dsl93-89-128.idnet.net] has quit [Remote host closed the connection] 19:35 < Peng_> Aww. 19:36 < Peng_> I almost want to buy another Linode just to test this. 19:36 < SelfishMan> DO IT 19:37 < tylerdu> just do it 19:37 < Peng_> Just buy a Linode, or just break my connections? 19:37 < SelfishMan> buy a linode 19:37 < tylerdu> buy a linode 19:37 < HoopyCat> science, pro-rated to the day! 19:37 < Peng_> But I'm on annual billing, so it'll be like 10 months before the pro-rated money actually gets used. 19:37 < SelfishMan> mikegrb promises free setuping if you buy another node 19:38 < Peng_> s/money/refund/ 19:38 < SelfishMan> eh? 19:38 < SelfishMan> wait, with annual billing all your nodes have to be annual? 19:38 < Peng_> No. I doubt it, anyway. 19:39 -!- River_Rat [~me@174-22-133-200.clsp.qwest.net] has quit [Read error: Operation timed out] 19:39 * SelfishMan is confused 19:39 -!- verb [~blarg@bespoke.blarg.org] has quit [Quit: leaving] 19:39 -!- River-Rat [~me@174-22-141-251.clsp.qwest.net] has joined #linode 19:40 < Peng_> Say I buy a Linode, then cancel it in an hour. I'll pay $20 and then be credited $19 or whatever, right? My other Linode's next payment is 2010-07-01 (or 2010-06-01?), so the $19 will only benefit me then. 19:40 < HoopyCat> Peng_: you can use that the NEXT time you need to use science 19:41 < SelfishMan> it works! 19:41 < HoopyCat> and really it's just $20 i spend more than that on cask ale to forget the meaninglessness and direness of my life you know? 19:42 < SelfishMan> mmm...new linode 19:42 < HoopyCat> if linode had a way for you to put $1 into my linode account, i'd totally use my standing credit for this 19:44 < SelfishMan> Linode screwed themselves with the whole prorated to the day thing 19:44 < SelfishMan> I think several people (including myself) would keep a node all month just for testing that would mostly idle 19:44 < exor674> SelfishMan: yes, prorated to the nanosecond would be way more awrsome! 19:45 < HoopyCat> SelfishMan: i'd want to, but i can't afford to 19:46 < Peng_> It would be neat to be able to buy a Linode for an hour, NearlyFreeSpeech.Net-like. 19:48 < HoopyCat> $0.0277/hr starts getting a little awkward 19:50 < SelfishMan> I don't know. Linode has hit my credit card for <$0.03 before 19:50 -!- zack_ [~zack@64-71-16-254.static.wiline.com] has joined #linode 19:51 -!- zack_ [~zack@64-71-16-254.static.wiline.com] has quit [Remote host closed the connection] 19:51 -!- zack [~zack@64-71-16-254.static.wiline.com] has quit [Read error: No route to host] 19:51 < SelfishMan> !extras 19:51 < linbot> Available Add-ons: Disk: $ 2 per 1GB/month. RAM: $ 5 per 90MB/month. Transfer: $ 10 per 100GB/month. IP: $ 1 per IP/month 19:51 -!- zack [~zack@64-71-16-254.static.wiline.com] has joined #linode 19:52 -!- SDjernes1 [~shawn@ip70-187-15-75.om.om.cox.net] has joined #linode 19:52 < Peng_> What did you spend $0.03 on? 19:52 < SelfishMan> I upgraded, downgraded, added and/or removed multiple nodes at the same time 19:52 < Peng_> Ah. 19:53 < SelfishMan> credits didn't quite cover the charges 19:53 < SelfishMan> I think that was the day that tasaro turned to caker and said "user resizing. nao plz" 19:53 < HoopyCat> SelfishMan: shoulda waited a day, you anti-business commie asshole 19:54 < SelfishMan> HoopyCat: but I needed it right then! 19:55 -!- DephNet[Paul] [~Paul@94-194-180-156.zone8.bethere.co.uk] has quit [Ping timeout: 480 seconds] 20:00 -!- kelvinq [~kelvinq@bb121-7-88-165.singnet.com.sg] has joined #linode 20:01 -!- woz [~woz@87-194-162-4.bethere.co.uk] has joined #linode 20:03 -!- daMaestro [~jon@content.beatport.com] has quit [Quit: Leaving] 20:08 < Peng_> Wouldn't they be charged more than $0.03 in CC fees and whatever? 20:08 -!- supine [~marty@115.128.34.71] has quit [Ping timeout: 480 seconds] 20:09 -!- goofykinky [~chatzilla@189.156.142.148] has quit [Quit: ChatZilla 0.9.85 [Firefox 3.5.3/20090824101458]] 20:10 -!- kenichi [~kenichi@207.162.220.10] has quit [Quit: kenichi] 20:10 < Twayne> . 20:10 < HoopyCat> Peng_: probably, hence SelfishMan being an anti-business commie asshole. :-) i figure it's rare enough in practice that it's probably not worth coding around too vigorously, tho 20:11 -!- panpainter [~panpainte@gw.onehub.com] has joined #linode 20:12 -!- dzjepp [~menace2s@adsl-76-239-28-192.dsl.emhril.sbcglobal.net] has quit [Read error: Connection reset by peer] 20:12 -!- User12482 [~User82934@70.134.68.122] has quit [Ping timeout: 480 seconds] 20:13 < Peng_> Indeed. 20:13 -!- User12482 [~User82934@70.134.68.122] has joined #linode 20:22 -!- Mathew [~Elvis@cpc4-flit1-0-0-cust346.lutn.cable.ntl.com] has quit [Read error: Connection reset by peer] 20:27 < straterra> http://twitpic.com/i8bq0 20:29 -!- woz [~woz@87-194-162-4.bethere.co.uk] has quit [Quit: woz] 20:29 < linbot> New news from forums: NginX Setup on Multiple Domains in Linux, Apache, Mysql and PHP (LAMP) Forum 20:30 -!- supine [~marty@office.rbery.bulletproof.net] has joined #linode 20:33 -!- FTA [~4a651d4a@webuser.linode.com] has joined #linode 20:33 < FTA> Hi, guys I wanted to know if I can run pptp on xen I know it doesnt work on virtuozzo 20:34 < straterra> err yeah 20:34 < straterra> not sure why you wouldnt be able to 20:34 < FTA> yah I wanna make sure its possible 20:34 < FTA> because virtuozzo doesnt work 20:35 < straterra> I run pptp and openvpn in my linode 20:36 < FTA> nice it works fine right? 20:36 -!- Turl [~Turl@host26.190-224-48.telecom.net.ar] has joined #linode 20:36 < straterra> yup 20:38 < FTA> are there any specials? 20:38 < straterra> I don't think so 20:40 < BarkerJr> do host loads go down over time? 20:40 -!- FTA [~4a651d4a@webuser.linode.com] has quit [Read error: Connection reset by peer] 20:40 < JshWright> FTA: sign up for 12 months and get a 10% discount 20:40 < BarkerJr> the host I setup on a few hours ago has been red since 20:40 < JshWright> BarkerJr: the "Host Load" indicator in the control panel is useles (sayeth caker) 20:40 -!- Bohemian [~Bohemian@24.92.32.92] has joined #linode 20:45 < BarkerJr> that said, would you recomend I replace a "low" with a "medium" host? 20:45 -!- Bohemian [~Bohemian@24.92.32.92] has left #linode [] 20:46 < Peng_> BarkerJr: But it's useless. 20:48 < BarkerJr> uh huh 20:54 -!- panpainter_ [~panpainte@gw.onehub.com] has joined #linode 20:55 -!- panpainter_ [~panpainte@gw.onehub.com] has quit [] 20:55 -!- Joao_1980 [brkkh@89.180.70.71] has joined #linode 21:01 -!- panpainter [~panpainte@gw.onehub.com] has quit [Ping timeout: 480 seconds] 21:07 < BarkerJr> do cancelling IP addresses not produce credit? 21:07 < caker> it does 21:07 < BarkerJr> oh nm, the credit for the vps and ip are in one invoice 21:07 < BarkerJr> makes sense :) 21:08 < Peng_> jtsage: ping 21:08 < BarkerJr> jtsage: ping6 21:10 < Peng_> Heh. 21:12 -!- Smark[Gone] is now known as Smark 21:18 -!- Turl [~Turl@host26.190-224-48.telecom.net.ar] has quit [Quit: Leaving.] 21:20 -!- User12482 [~User82934@70.134.68.122] has quit [Quit: Client Quit] 21:21 -!- SDjernes1 [~shawn@ip70-187-15-75.om.om.cox.net] has left #linode [] 21:23 -!- Harry_Mudd [~jon@c-71-235-73-29.hsd1.ct.comcast.net] has joined #linode 21:24 < jforman> caker: i signed up for a linode last night, and have since been acclimating myself to it. do you guys send out any sort of email verification, "hey welcome" and "we love your credit card long time" every month? 21:27 -!- bob2 [rob@crumbs.ertius.org] has joined #linode 21:28 -!- Joao_1980 [brkkh@89.180.70.71] has quit [Quit: Scp Script 2003 ! pegue o seu em www.scoop.com.br] 21:30 < amitz> jforman: we're going to hug your card and we love your card are done every month yes but I forgot about the initial welcome. 21:30 < caker> jforman: hey -- welcome. You'll get a detailed invoice email, and then a payment receipt email once your CC goes through 21:31 < straterra> jforman: after three months you get the brand...and the keycard to the underground meeting room 21:32 < caker> secret handshake email goes out at 6 mo 21:32 < JoeK> all i got was a free month 21:32 < JoeK> o.o 21:34 < HoopyCat> THERE IS NO CABAL 21:36 < jforman> straterra: judging from this channel, i'm not sure i'm willing to give up the shame that is required to enter said meeting room 21:36 < HoopyCat> shame is an investment 21:36 < straterra> jforman: Have you seen me in a bra? 21:37 < jforman> if you have to ask, no, and i refuse to in any context in the future 21:37 < straterra> Oh come on 21:37 < HoopyCat> there are only two places to see straterra in a bra... half the URLs posted in this channel, and the avatar of anyone who uses my e-mail address to sign up for some crap-ass social networking thing 21:37 < straterra> Be a good sport.. 21:37 < jforman> caker: ah okay, was just curious. and thanks :) been great so far, i have a 360 in atlanta. great ping times from boston 21:41 < HoopyCat> The company reserves the right to re-instate the fuel supplement for all customers at up to $9 per Linode per day if the NYMEX oil price exceeds $70 per barrel. 22:04 -!- jtaji [~jtaji@c-76-124-234-212.hsd1.pa.comcast.net] has joined #linode 22:07 < MJCS> fuck the windows key is stuck in my server...I dont have a virtual keyboard nor can i restart as it is my production server and people are connected taking some e-learning tests 22:07 < MJCS> any ideas 22:08 < MJCS> oh and citrix does not allow me to pass the windows key 22:08 < bob2> get a new keyboard 22:08 < metap|pe> preferably one without a windows key 22:09 < MJCS> haha 22:09 < MJCS> no 22:09 < MJCS> i just reped into a computer that has a virtual keyboard and then rdped through that 22:09 < MJCS> that fixed it 22:10 < jed> what does ubuntu call it 22:10 < jed> the Awesome key 22:10 < jed> they have some silly name for it so they don't say windows key 22:11 < jed> Multi key? I forget. 22:11 < palintheus> super-key is what I've heard 22:11 < jed> that's it 22:11 < jdub> which dates back far longer than ubuntu 22:11 < palintheus> right 22:11 < jed> oh, I'm unaware 22:11 < jdub> probably 20 years 22:11 < metap|pe> oh yeah, xev shows it as super_l or something 22:12 < jed> 20 years? 22:12 < metap|pe> or more 22:12 < jed> I understand command and meta and the like going back 20+, but super is a surprise 22:13 < bob2> and hyper 22:13 < jed> wikipedia discusses the space-cadet keyboard, whose super key was of a different meaning 22:16 < jdub> jed: that's where all the names come from, on through emacs and X 22:17 < jed> I really meant the super key as it relates to the windows key, that association 22:17 < jed> I wasn't aware of the super key's history, and I've been duly informed 22:17 < jed> its association with the windows key does not 20 years back go :) 22:17 -!- Twayne [~waynemilt@cpe-071-070-201-028.nc.res.rr.com] has quit [Ping timeout: 480 seconds] 22:17 < jdub> "Ah, fuck it, Microsoft have added an ego key to the keyboard... how do we use this in X?" 22:18 < jdub> "Dunno. Pick the next Space Cadet idiot key name and use it." 22:20 -!- Talman [~Talman|Aw@174-20-33-9.mpls.qwest.net] has joined #linode 22:21 < palintheus> it's awesome that system76 and ZaReason have an ubuntu key now http://system76.com/images/nb1_keyboard_large.jpg 22:23 < checkers> oh dear 22:23 < checkers> that just looks like a recipe for confusion 22:23 < palintheus> why? 22:23 < bob2> at least make it a tux key or something 22:23 < bob2> or a vague diamond shape like acer does 22:23 -!- Talman|Ghosting [~Talman|Aw@97-127-12-112.mpls.qwest.net] has quit [Ping timeout: 480 seconds] 22:24 < palintheus> again, why? system76 sells preinstalled ubuntu... 22:24 < checkers> the only people who would know that 'windows key = ubuntu key' are the people who would realise that you can sitll use a windows key on ubuntu anyway 22:24 < checkers> given the entire internet so far calls it a 'windows key' it just seems like a cruel joke to play on unsuspecting users 22:25 < palintheus> checkers: I don't see many unsuspecting users buying from either company 22:25 < jdub> heh 22:25 < jdub> unsuspecting user buys ubuntu pre-installed 22:30 < Pryon> with any luck, doesn't notice 22:51 -!- hercynium [~hercynium@c-98-216-53-107.hsd1.ma.comcast.net] has quit [Quit: Leaving] 22:53 -!- maher [~maher@user-0cevfbp.cable.mindspring.com] has joined #linode 22:54 -!- maher [~maher@user-0cevfbp.cable.mindspring.com] has quit [] 23:02 -!- elfgoh [~dingding@adsl187.dyn212.pacific.net.sg] has joined #linode 23:16 -!- Guest158 [~roberto@pool-71-177-159-19.lsanca.fios.verizon.net] has quit [Quit: Leaving] 23:18 < linbot> New news from forums: Understanding configuration profiles in Sales Questions and Answers 23:23 -!- User12482 [~User82934@70.134.68.122] has joined #linode 23:36 -!- kelvinq [~kelvinq@bb121-7-88-165.singnet.com.sg] has quit [Quit: kelvinq] 23:38 -!- jordanb [~jordanb@adsl-99-21-161-249.dsl.chcgil.sbcglobal.net] has joined #linode 23:38 < jordanb> Hi. I had a friend put me as a referral a long time ago and it's still in a 'pending' state. Does anyone know how long that's supposed to take? 23:39 -!- panpainter [~panpainte@c-66-235-62-16.sea.wa.customer.broadstripe.net] has joined #linode 23:42 < bd_> jordanb: http://blog.linode.com/2008/07/31/referral-system-annual-discounts/ 90 days 23:43 < jordanb> bd_ Thanks 23:44 < jordanb> I suppose it hasn't been 90 days yet. 23:47 < MJCS> my logic fails me 23:48 -!- panpainter [~panpainte@c-66-235-62-16.sea.wa.customer.broadstripe.net] has quit [Quit: panpainter] 23:54 -!- jordanb [~jordanb@adsl-99-21-161-249.dsl.chcgil.sbcglobal.net] has quit [Quit: jordanb] 23:56 -!- ryanc [~ryanc@173-13-151-173-sfba.hfc.comcastbusiness.net] has quit [Ping timeout: 480 seconds] 23:57 -!- User12482 [~User82934@70.134.68.122] has quit [Quit: Client Quit] --- Log closed Tue Sep 29 23:59:11 2009