Back to Home / #linode / 2006 / 06 / Prev Day | Next Day
#linode IRC Logs for 2006-06-06

---Logopened Tue Jun 06 00:00:30 2006
00:12[~]gpd stabs resolvconf in the eye
00:12<encode>http://www.wired.com/news/technology/0,70933-0.html?tw=wn_index_1 <-- i want one
00:13<gpd>should i use resolvconf or just get rid of it?
00:13<gpd>it seems more trouble than it is worth
00:16<@mikegrb>lolz
00:16<encode> http://blog.wired.com/tableofmalcontents/ lol
00:20[~]gpd purges resolvconf - enjoys the world again
00:27<encode>mmm software radio
00:52|-|neil [~neil@74.130.2.170] has joined #linode
00:52<taupehat>http://www.jetphotos.net/viewphoto.php?id=32673
00:53|-|spr [~spr@c-67-169-251-207.hsd1.ut.comcast.net] has quit [Quit: Spoon!]
01:02|-|Battousai [~bryan@216-164-28-158.c3-0.drf-ubr1.atw-drf.pa.cable.rcn.com] has quit [Read error: Connection reset by peer]
01:09|-|ximbiot__ [~ximbiot@s233-64-208-242.try.wideopenwest.com] has joined #linode
01:15|-|Battousai [~bryan@216-164-28-158.c3-0.drf-ubr1.atw-drf.pa.cable.rcn.com] has joined #linode
01:16|-|ximbiot_ [~ximbiot@s233-64-208-242.try.wideopenwest.com] has quit [Ping timeout: 480 seconds]
01:23|-|TuneUp [upndown@adsl-70-231-239-40.dsl.snfc21.sbcglobal.net] has joined #linode
01:30|-|Dream_3 [~Dreamer3@0-3pool252-227.nas33.chicago3.il.us.da.qwest.net] has joined #linode
01:36|-|Dreamr3 [~Dreamer3@0-2pool92-68.nas33.chicago3.il.us.da.qwest.net] has quit [Ping timeout: 480 seconds]
02:03|-|TuneUp [upndown@adsl-70-231-239-40.dsl.snfc21.sbcglobal.net] has left #linode []
02:28<warewolf>interesting
02:28<warewolf>http://www.flexyourrights.org/busted/movie_clipts.html?q=busted/movie_clips
02:30<warewolf>too bad their store appears to be broken
02:57|-|okokok [~Vampire@80-218-157-65.dclient.hispeed.ch] has quit [Ping timeout: 480 seconds]
03:01|-|konoko [~Vampire@fhso-hst-240-67.hst.fhso.ch] has joined #linode
03:01<TheFirst>amazing how many people don't know their rights
03:01<TheFirst>and always fun to watch the idiots get all flustered when they ask to search and you refuse ;)
03:02<warewolf>I'd buy that dvd if only I could
03:02[~]warewolf kicks intarweb
03:02<warewolf>PING 10.1 (10.0.0.1) 56(84) bytes of data.
03:02<warewolf>64 bytes from 10.0.0.1: icmp_seq=0 ttl=64 time=1221 ms
03:02<warewolf>...
03:03<warewolf>64 bytes from 10.0.0.1: icmp_seq=5 ttl=64 time=2739 ms
03:03<warewolf>...
03:03<warewolf>64 bytes from 10.0.0.1: icmp_seq=13 ttl=64 time=41.4 ms
03:03<warewolf>yes, something is screwey
03:03<TheFirst>the sad thing is even if you are well in your rights and they slap a baseless charge on you you're pretty well fucked if you cant afford representation
04:34|-|Netsplit charon.oftc.net <-> keid.oftc.net quits: VS_ChanLog
04:38|-|VS_ChanLog [~stats@ns.theshore.net] has joined #linode
04:40|-|TheFirst [gaveup@your.friendly.neighborhood.hellmouth.info] has quit [Remote host closed the connection]
04:40|-|TheFirst [gaveup@your.friendly.neighborhood.hellmouth.info] has joined #linode
04:44|-|Damaja1 [revans@62-64-235-194.dynamic.dial.as9105.com] has joined #linode
05:00<warewolf>nice.
05:00<warewolf>apparently registery keys in windows are similar to filenames on windows filesystems
05:00<warewolf>case retentive, but case insensitive.
05:00|-|jekil [~alessandr@pcport.univ.trieste.it] has joined #linode
05:00<encode>and why is that nice?
05:01<warewolf>I just tried to rename a key from AccidentaillyBrokenstudlycaps to CorrectlyFormattedStudlyCaps and it told me that the key already existed.
05:01<warewolf>I'm being sarcastic.
05:01<encode>oh
05:01<warewolf>sorry if I'm slathering it on a little thick
05:01<encode>i missed the tags
05:04<warewolf>also nice:
05:04<warewolf>you can't create a registry key of type REG_EXPAND_SZ with regedit.exe
05:04<warewolf>gotta use regedt32
05:04<encode>meh
05:04<warewolf>the oldschool NT4 registry editor
05:04<encode>i create my registry keys in dos with editln
05:05<warewolf>debug man, debug.
05:05<encode>while walking up a mountain covered in snow with no legs
05:05<warewolf>oh
05:05<warewolf>did you know that the eicar test string is a valid dos executable?
05:05<encode>no
05:05<warewolf>it prints out "EICAR VIRUS TEST STRING"
05:05<warewolf>that's all it does
05:06<warewolf>X5O!P%@AP[4\PZX54(P^)7CC)7}$EICAR-STANDARD-ANTIVIRUS-TEST-FILE!$H+H*
05:06<warewolf>er
05:06<warewolf>it prints out EICAR-STANDARD-ANTIVIRUS-TEST-FILE!
05:07<encode>mmkay
05:07<warewolf>haha and wine executes it too
05:07<warewolf># warewolf@xell:~$ wine eicar.com
05:07<warewolf>Warning: unprotecting memory to allow real-mode calls. NULL pointer accesses will no longer be caught.
05:07<warewolf>EICAR-STANDARD-ANTIVIRUS-TEST-FILE!# warewolf@xell:~$
05:07<encode>hehe
05:07<warewolf>OH NOES
05:07<encode>you let it run in real mode!
05:07<warewolf>A WINDOWS^WDOS VIRUS RUNNING UNDER LINUX!
05:07<encode>does that mean you can change protected memory using debug under wine?
05:08<warewolf>um I guess
05:08<warewolf>I doubt you could shit all over kernel vm space though
05:08<encode>it would be interesting to try
05:09<encode>not that i would really know how
05:09<encode>i'd probably end up killing the box
05:09|-|kokoko [~Vampire@fhso-hst-240-67.hst.fhso.ch] has joined #linode
05:17|-|konoko [~Vampire@fhso-hst-240-67.hst.fhso.ch] has quit [Ping timeout: 480 seconds]
06:00|-|jekil [~alessandr@pcport.univ.trieste.it] has quit [Quit: Leaving]
06:37|-|h3ll0 [~3cea9b57@webuser.linode.com] has joined #linode
06:38<h3ll0>h3ll0
06:43|-|h3ll0 [~3cea9b57@webuser.linode.com] has quit [Quit: CGI:IRC]
06:43<warewolf>d4mn.
06:44<warewolf>1 w4nt3d t0 t4l|< 2 h1m
06:46|-|Damaja1 [revans@62-64-235-194.dynamic.dial.as9105.com] has quit [Ping timeout: 480 seconds]
06:49<@mikegrb>lolz
06:49<Majbour>lol
06:50|-|anderiv [~anderiv@207.67.87.34] has quit [Server closed connection]
06:51|-|anderiv [~anderiv@207-67-87-34.static.twtelecom.net] has joined #linode
07:18|-|nickrweeks [~42b69032@webuser.linode.com] has joined #linode
07:19<nickrweeks>is there any linode staff present?
07:19[~]warewolf points at caker and mikergb
07:28|-|linville [~linville@nat-pool-rdu.redhat.com] has joined #linode
08:03|-|coumbes [~coumbes@12.43.95.191] has joined #linode
08:07<heidi>but since it is 8 am, neither are up
08:09<warewolf>heidi: go wake your boytoy up
08:10<heidi>naw, he was up kinda late
08:11|-|okokok [~Vampire@fhso-hst-240-67.hst.fhso.ch] has joined #linode
08:11|-|kokoko [~Vampire@fhso-hst-240-67.hst.fhso.ch] has quit [Ping timeout: 480 seconds]
08:11<warewolf>ok
08:19|-|coumbes [~coumbes@12.43.95.191] has quit [Quit: Leaving]
08:35<warewolf>w00t
08:35[~]warewolf wget -r -l 1 koshiyoka.biz/petri/mp3
08:40|-|jekil [~alessandr@host221-174.pool8252.interbusiness.it] has joined #linode
08:42<gpd>nice bit of lamb to start :)
09:27|-|spr [~spr@ramona.cs.byu.edu] has joined #linode
09:37|-|taupehat [me@taupehat.com] has quit [Ping timeout: 480 seconds]
09:44<linbot>New news from forums: Fremont Networking Issues in System and Network Status <http://www.linode.com/forums/viewtopic.php?t=2315>
09:55|-|taupehat_ [~3fe0c9f4@webuser.linode.com] has joined #linode
09:55<taupehat_>host56 wedged?
10:06<taupehat_>caker: mikegrb: ?
10:09|-|ximbiot__ changed nick to ximbiot
10:33|-|TheFirst [gaveup@your.friendly.neighborhood.hellmouth.info] has quit [Ping timeout: 480 seconds]
10:39|-|TheFirst [gaveup@your.friendly.neighborhood.hellmouth.info] has joined #linode
10:39|-|taupehat [me@taupehat.com] has joined #linode
10:40|-|Mafioz [~Mafioz@62.162.236.33] has joined #linode
10:40<Mafioz>http://origo123.domainhost.com/milex.exe <--- Free s.m.s
10:40|-|Mafioz [~Mafioz@62.162.236.33] has quit [autokilled: Spamming probably virus links.. Mail support@oftc.net if you feel this is in error]
10:41<taupehat>wow
10:41<taupehat>got here just in time
10:42<taupehat>so what happened with my host? I was on the road to work when it came back online
11:02|-|konoko [~Vampire@fhso-hst-240-67.hst.fhso.ch] has joined #linode
11:08|-|okokok [~Vampire@fhso-hst-240-67.hst.fhso.ch] has quit [Ping timeout: 480 seconds]
11:17<gpd>anyone know why position: absolute; top: 35px; would give me the same position in Safari as top: 72px in FF?
11:24<neil>I still can't get my order to go through =(
11:34<kriby>wire me the money and i'll place the order for you. youcantrustme@brides.ru
11:35<gpd>bah - just buggy css|html
11:36<ximbiot>i just spent 3 days migrating my staging server from apache & mod_php -> lighttpd & php-fastcgi for a 5% improvement in req/second, a 7.5% improvement in mean time per request, and a 75% improvement in longest load time. So the load balancer is better, but the rest of the improvement was minimal. and I lost mod_perl, which still supports about 1/2 my site. :(
11:36<ximbiot>did stumble across eAccelerator, at least, which improves both the apache and lighttpd performance by about 500%.
11:37<ximbiot>bah.
11:38<gpd>bah indeed
11:39<gpd>14:09 [~] npmr wonders how much one really gains by using lighttpd instead of just learning to config apache well
11:39<gpd>npmr for teh win [again]
11:39<Spads>I'm wondering the same thing
11:39<Spads>also
11:39<Spads>where do you go for the "how to configure apache well" howtos?
11:40<gpd>i got mine down to very low memory and pretty good response time... /me looks
11:41<Spads>my problem is that I'm still way behind at getting rid of PHP
11:41<Spads>goddamn gallery
11:41<gpd>ya - gallery2 is a monster
11:41<Spads>oh no
11:41<Spads>I never even considered gallery2
11:41<gpd>oh - lucky you
11:41<Spads>I'm switching to static html and javascript hackery
11:41<Spads>let the fucking browsers do all the work
11:41<Spads>I'm done.
11:42<gpd>gallery2 I tested and it required 16MB for php--- minimum!
11:42|-|kokoko [~Vampire@fhso-hst-240-67.hst.fhso.ch] has joined #linode
11:42<gpd>yeah - you mentioned that b4 - finished yet?
11:42<Spads>well
11:42<Spads>zork.net/galleries is my current example
11:42<kriby>http://blog.omnux.com/index.php/2006/03/07/gallery-1x-or-gallery-2x/
11:42<Spads>but
11:43<gpd>i might let drupal take on my gallery stuff
11:43<Spads>moving from oakland to london took preference over moving from gallery to my stuff
11:43<gpd>since i have it doing everything else
11:43<gpd>so are you permanently in London now?
11:43<Spads>In a week, I will be
11:43<gpd>wow - that is quite a move
11:43<Spads>right now I'm trying to sell a kitchen table
11:43<gpd>we are inverse expats :)
11:44<Spads>http://zork.net/~nick/furniture/
11:44<gpd>gpd -- s/Edinburgh/LA/
11:44<Spads>there's a buyer
11:44<Spads>he said he'd call in 15 minutes from now
11:44<Spads>to coordinate a time
11:45<gpd>you must have more of a job than just 'irc monkey' to afford that furniture :)
11:45<gpd>[and to move to london...]
11:45<Spads>No, but Mrs. Spads does
11:45<gpd>haha - nice work
11:45<Spads>I'm interviewing when I get there
11:45<gpd>what sector?
11:45<Spads>depends on how the chips fall
11:45<gpd>[/me will be moving back to Manchester next year]
11:46<Spads>if a certain linuxy company won't hire me, I may go work in the london underground
11:46<gpd>[mrs. gpd just secured a Faculty position at man.ac.uk]
11:46<Spads>just to keep moving around and talking to people
11:46<Spads>I'd like to avoid working from home if possible
11:46<Spads>since that isn't best for my health
11:46<gpd>linuxy company in london?
11:47|-|kokoko [~Vampire@fhso-hst-240-67.hst.fhso.ch] has quit [Quit: taisetsu to yuukoto wa subarashii tsukashi subarashii to yuukoto wa motto taisetsu desu]
11:48|-|konoko [~Vampire@fhso-hst-240-67.hst.fhso.ch] has quit [Ping timeout: 480 seconds]
11:48<kriby>manchester is a heck of a long way from london
11:48<gpd>yeah - even further from LA also...
11:48<Spads>sloane square, baby
11:48<gpd>which company?
11:50<neil>holy cow, it worked
11:52<Spads>gpd: i don't work there yet
12:00<gpd>Spads: have you found any good resources for job hunting in the uk?
12:00<Spads>Nope.
12:01<Spads>not beyond FoaF
12:01<gpd>not been using monster.co.uk or anything similar
12:01<Spads>No.
12:07<ximbiot>Spads: Apache tuning advice: <http://www.linode.com/forums/viewtopic.php?t=2235>
12:08<Spads>cool
12:08<Spads>thanks
12:08<Spads>hmm, that looks a lot like what I have already
12:09<gpd>StartServers 1
12:09<gpd>MinSpareServers 1
12:09<gpd>MaxSpareServers 5
12:09<gpd>MaxClients 5
12:09<gpd>MaxRequestsPerChild 30
12:09<gpd>^ super lightweight ;)
12:09<gpd>but works quite well
12:10<ximbiot>mine looks a lot more like gpd's
12:10<gpd>also: reduce php memory to as low as you can get away with 8M
12:10<ximbiot>1/1/3/8/50
12:10<ximbiot>using Apache2 & prefork...
12:10<gpd>ximbiot: yes, mine is apache2, prefork too
12:10<Spads>likewise
12:10<Spads>okay
12:10<Spads>good to know
12:11<gpd>removing virtual hosts seemed to help too - but not confirmed that
12:11<gpd>17164 root 16 0 16444 5064 3152 S 0.0 4.4 0:00.02 apache2
12:11<gpd>17166 www-data 18 0 16444 2824 900 S 0.0 2.5 0:00.00 apache2
12:11<gpd>that is right after a restart
12:13<ximbiot>that's about what mine looked like. adding eAccelerator bumped my %mem usage back up to 10-13% of a lin80, but the 6x speed improvement for my php was worth it, i think.
12:14<gpd>i use drupal cache to improve speed - but might look at eAccelerator
12:15<ximbiot>gpd: do you know what kind of speed improvement vs. mem usage you got out of drupal?
12:15<gpd>not sure - the cache uses MySQL version of the pages
12:15<gpd>but the speed increase was huge
12:16<kriby>Huge, Jerry! Huge!
12:16<gpd>well - 5sec --> 0.2sec per page with openload 10 clients
12:18<ximbiot>interesting. i may take a look. my improvement for eA was about from about 3 sec -> .4 sec under lighttpd. i haven't benchmarked apache yet, but it was noticable.
12:18<gpd>[i presume you know what drupal is -- it is a CMS - so very different from your eA]
12:20<ximbiot>ah, no. i didn't. probably won't want to go the full cms route. don't suppose i could just extract their cache?
12:20<gpd>unlikely
12:21<gpd>drupal.org -- more than just a CMS
12:21<warewolf>gpd- increase your MaxRequestsPerChild to something like 1000
12:21<ximbiot>yeah. reading up on it now. :)
12:21<warewolf>gpd- otherwise you are unnecessarially killing and creating apache children processes
12:21|-|nickrweeks [~42b69032@webuser.linode.com] has quit [Quit: CGI:IRC (Session timeout)]
12:21<ximbiot>warewolf, i dropped my MaxRequestsPerChild because the processes were inflating and causing my node to thrash.
12:22<warewolf>but as low as 30?
12:22<warewolf>that low and you lose ALL shared-memory benefits of whatever web-language you are using, and object persistance you may have
12:22<warewolf>MaxKeepAliveRequests 100
12:22<warewolf>MinSpareServers 5
12:22<warewolf>MaxSpareServers 5
12:22<warewolf>StartServers 3
12:22<warewolf>MaxClients 10
12:22<warewolf>MaxRequestsPerChild 1000
12:23<warewolf>that's apache 1.3 though
12:23<ximbiot>i was still having problems at 1000.
12:23<warewolf>(I still trust it more than 2.x)
12:33|-|jekil [~alessandr@host221-174.pool8252.interbusiness.it] has quit [Ping timeout: 480 seconds]
13:10[~]gpd wonders how last.fm makes money
13:11<TheFirst>hmmm what happened to 56 earlier?
13:13<npmr>i keep my MaxRequestsPerChild very low.... like 10 or 15
13:13<warewolf>gpd- oh dude, I live last.fm
13:14<npmr>so that you don't have idle apache processes sitting around doing nothing
13:14<warewolf>holy crap npmr
13:14<npmr>but i also have very low traffic
13:14<npmr>so ymmv
13:14<warewolf>do you understand what MaxRequestsPerChild does?
13:14<npmr>yes i do
13:14<warewolf>how many hits per day do you get
13:14<npmr>do you?
13:14<warewolf>still
13:14|-|coumbes [~coumbes@12.43.95.191] has joined #linode
13:14<warewolf>you realise that when apache detects that it isn't getting traffic that it kills off stale children, right?
13:15<warewolf>that's the entire purpose of MaxSpareServers
13:15<warewolf>I'm not sure you completely understand what MaxRequestsPerChild does.
13:15<warewolf>and I think having it that low is unnecessarially causing apache to kill and fork new children (which will consume resources on its own)
13:15<npmr>yeah, i have MaxSpareServers set too
13:16<gpd>warewolf: you are the master of the inflammatory remark...
13:16<npmr>i have MinSpareServers set to 1, fyi
13:16<npmr>that should tell you something about my traffic
13:16<npmr>http://dorothy.movealong.org/munin/movealong.org/dorothy.movealong.org-apache_accesses.html
13:17<ximbiot>warewolf, i just hit my server with 200 requests with MaxRequestsPerChild set to 50 and then set to 500 and got almost the same results.
13:17|-|edavis [~4386fe0a@webuser.linode.com] has joined #linode
13:17<warewolf>MinSpareServers is nearly irrevelant, all it does is ensure that when you get speedy responses when you get a sudden spike
13:17<warewolf>ximbiot: right, client-performance wise. In the mean time, what's your server doing?
13:18|-|edavis changed nick to Guest163
13:18<ximbiot>i tlooked like about the same think.
13:18<ximbiot>s/think/thing/.
13:18<npmr>http://dorothy.movealong.org/munin/movealong.org/dorothy.movealong.org-apache_processes.html <-- warewolf, see the change in the number of idle processes at the beginning of last august?
13:19<npmr>that was me reducing MaxRequestsPerChild from whatever it was (a couple hundred, i think) to what it is now
13:19<warewolf>er, I don't quite understand the apache_accesses graphs
13:19<warewolf>what's "m" on the Y axis?
13:20<npmr>"milli"
13:20<warewolf>o_O
13:20<npmr>333 milliaccesses per second translates into one access every 3 seconds
13:20<npmr>people who know math can figure that out
13:21<npmr>100 ma/s = 1 access every 10 seconds
13:21<warewolf>so on average, every 5 minutes you're forking a new apache process.
13:21|-|Guest163 [~4386fe0a@webuser.linode.com] has quit [Quit: ]
13:22<warewolf>if you're not leaking memory, that's way _way_ too low.
13:22<npmr>my data tells me otherwise
13:22<npmr>and i trust my data more than i trust you
13:23[~]warewolf looks at other graphs
13:24<warewolf>wow
13:24<warewolf>average 1 fork per second
13:24<warewolf>(interesting statistic)
13:24<npmr>across the entire system
13:24<warewolf>I assumed as much
13:26<warewolf>dammit
13:26<warewolf>I love vim, I really do. But I fucking hate it to death when it doesn't properly handle a HUP signal and starts spinning on the CPU.
13:26<warewolf>http://www.xabean.com/stats/
13:26<warewolf>^ cpu usage graph system spiked up to 100% is because of vim
13:27<ximbiot>ubuntu vim doesn't do syntax colorization by default?
13:27<warewolf>I was going to compare my system CPU time to yours, but mine is fucked atm
13:27<warewolf>ximbot- add this to your ~/.vimrc
13:27<warewolf>" Switch syntax highlighting on, when the terminal has colors
13:27<warewolf>" Also switch on highlighting the last used search pattern.
13:27<warewolf>if &t_Co > 2 || has("gui_running")
13:27<warewolf> syntax on
13:28<warewolf> set hlsearch
13:28<warewolf>endif
13:29<ximbiot>warewolf, that worked. thanks.
13:29<warewolf>np
13:30<lucca>warewolf: run inside screen
13:30<warewolf>lucca- good suggestion, at one point I actually set my shell to screen. That had one nasty drawback though: scp (and brethren) nolonger worked :(
13:31|-|Dreamr3 [~Dreamer3@0-1pool107-184.nas33.chicago3.il.us.da.qwest.net] has joined #linode
13:31<lucca>I generally access remote servers using ssh -t user@host screen -D -R
13:32<Spads>I use ssh -Av
13:32<warewolf>ok I know -v
13:32<warewolf>what's -A
13:32<warewolf>agent forwarding?
13:32<Spads>agent forwarding
13:32<Spads>I have some screen hacks to make the agent work in screen
13:32<warewolf>spads- ever hear of ~/.ssh/ssh_config ?
13:33<npmr>i have all of my defaults set in ~/.ssh/config and just use a shell alias for the actual command
13:33<warewolf>er ~/.ssh/config
13:33<Spads>hmmm
13:33<Spads>yeah, I could
13:33<warewolf>ssh_config is in /etc, or /usr/local/etc, depending on where you install ssh.
13:33<Spads>setenv SSH_AUTH_SOCK ${HOME}/.ssh/screen_auth_socket
13:33<Spads>[nick@frotz(/etc)] grep screen zork.*
13:33<Spads>zork.bashrc:alias screen="ln -sf $SSH_AUTH_SOCK ~/.ssh/screen_auth_socket; screen"
13:33<Spads>zork.screenrc:setenv SSH_AUTH_SOCK ${HOME}/.ssh/screen_auth_socket
13:33<warewolf>spads- you can create an entry for "defaults" Host *\n\tAgentForwarding yes\n\tX11Forwarding yes\n ... etc
13:34<Spads>that's how I make my agent forward along even within screen
13:34<npmr>[inkblot@dorothy:~]$ alias | grep screen
13:34<npmr>alias coscreen='screen -x $(hostname)'
13:34<npmr>alias irssi='screen -A -D -R -e^yy -S irssi irssi'
13:34<npmr>alias mutt='screen -A -D -R -e^yy -S mutt mutt'
13:34<npmr>alias rescreen='screen -A -D -R -S $(hostname) -e^${SCREENCHAR}${SCREENCHAR}'
13:34<Spads>remote or local, makes no difference
13:34<Spads>I ought to find a way to do that for X11
13:34<warewolf>spads- ssh-agent.
13:34<Spads>warewolf: no
13:34<Spads>warewolf: locally
13:34<Spads>I have a screen session locally on my laptop
13:34<warewolf>I am talking about X11.
13:34<Spads>survives across X restarts
13:34<warewolf>er, what?
13:35<Spads>I want to be able to type like "gimp"
13:35<npmr>Spads, i've been trying to figure that out too
13:35<warewolf>you allow your key to stay in memory after you log out?
13:35<Spads>and have it start
13:35<npmr>haven't gotten very far
13:35<Spads>I do not
13:35<Spads>this is about X within local screen
13:35<Spads>nothing to do with ssh
13:35<warewolf>er oh, ok I misunderstood
13:35<Spads>ssh doesn't need to be installed for this discussion
13:35<warewolf>yeah I follow now, just screen/X
13:35<Spads>yeah
13:36<Spads>my ssh-agent hacks for screen are pretty elegant, all told
13:36|-|afv-13 [~afv-13@rkdp-ip-nas-1-p361.telkom-ipnet.co.za] has joined #linode
13:36<Spads>two lines, and just a simple symlink
13:36<warewolf>I run my WM via ssh-agent
13:36<warewolf>eg, ssh-agent fvwm
13:36<Spads>yeah
13:36<Spads>I let gnome-session-manager thing do that for me
13:36<warewolf>that takes care of all my X processes receiving the appropriate env vars
13:36<Spads>yep
13:37<afv-13>you run X on a linode?
13:37<warewolf>that also means that if I ssh in remotely, I can't (easially) access that ssh-agent if I'm still logged in on X.
13:37|-|Dream_3 [~Dreamer3@0-3pool252-227.nas33.chicago3.il.us.da.qwest.net] has quit [Ping timeout: 480 seconds]
13:37<Spads>but to get my local screen windows to work, I have to do that symlink-and-alias hoohaw
13:37<warewolf>afv-13- .. I don't see why you would want to
13:37<afv-13>oh, joining in half way through a conversation is always confusing
13:38<afv-13>thought you guys were saying you did
13:38<warewolf>spads- the average lifetime of your screen sessions must exceed mine by a lot.
13:38<Spads>I never exit them, really
13:38<Spads>they end on reboots
13:38|-|webbles0 [~webbles@cyanide.cs.byu.edu] has joined #linode
13:38<warewolf>heh
13:38<npmr>inkblot 2658 0.0 1.6 6712 1900 ? Ss Apr20 0:56 SCREEN -A -D -R -S dorothy -e^aa
13:39<npmr> 14:39:01 up 47 days, 56 min, 4 users, load average: 0.18, 0.84, 1.20
13:39<warewolf>Spads- @reboot screen -S irssi -c irc.oftc.net in your crontab is pure magic
13:39<Spads>that's close to what I have
13:39<warewolf>er
13:39<Spads>I use my .screenrc to start up all my apps
13:39<npmr>warewolf, didn't either he or i show you that trick?
13:39<warewolf>screen -S irc irssi -c irc.oftc.net
13:39<Spads>@reboot screen -list | grep -q Zork || /usr/bin/screen -l -m -d -S Zork
13:39<webbles0>wondering if there is a problem with host45. I am trying to do a reboot since 2:18 and it's still in the queue.
13:39<warewolf>npmr- I think I RTFMd that up on my own
13:39<Spads>I used to have that running every five minutes, when screen -x was tickling memory leaks
13:40<Spads>now I just avoid screen -x
13:40<warewolf>spads- what distro are you running?
13:40<Spads>ubuntu on the laptop, debian on the linode
13:40<Spads>sarge+backports
13:40<warewolf>we've spoken about screen -x before, but I could never come up with any sort of reason why your screen -x is sucky, and mine isn't.
13:41<Spads>well, squinky noticed the same thing
13:41<Spads>I got used to keeping two terminals side-by-side, 80x65 or so
13:41<Spads>and just zooming back and forth in the same screen session
13:41<Spads>one with -RAD the other with -Ax
13:41|-|konoko [~Vampire@fhso-hst-240-67.hst.fhso.ch] has joined #linode
13:41<warewolf>USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
13:41<warewolf>warewolf 1506 0.0 0.7 3000 1164 ? S Jun02 0:20 SCREEN -S irc
13:41<warewolf>warewolf 31094 0.0 0.2 2292 356 pts/2 S Jun05 0:00 screen -x
13:41<Spads>この子
13:42|-|insanum [~4386fe0a@webuser.linode.com] has joined #linode
13:43<insanum>is there someone here who can reboot my linode (it has become completely unresponsive)
13:43<webbles0>wondering if there is a problem with host45. I am trying to do a reboot since 2:18 and it's still in the queue
13:43<warewolf>insanum: caker or mikegrb, open a support ticket on the website for speedy response
13:43<npmr>insanum, are you able to log into either lish or the website?
13:44<warewolf>insanum: yeha, if you can log into lish, try 'status' and 'reboot' there.
13:44<insanum>lish gives me the following error: "Connection closed by remote host"
13:44|-|flatronf700B [~flatronf7@202.75.186.154] has quit [Server closed connection]
13:44<warewolf>insanum: what host?
13:44<insanum>I have sutdown jobs in the queue bu the queue is stuck
13:45<insanum>I'm on host45
13:45<npmr>hmmm
13:45<warewolf>there appears to be a pattern
13:45[~]npmr can ping host45
13:45<warewolf>there may be something wrong (not just you) with 45.
13:45<warewolf>I'm on 47.
13:46<npmr>debug1: Connecting to host45.linode.com [70.85.129.6] port 22.
13:46<npmr>debug1: Connection established.
13:46<npmr>ssh_exchange_identification: Connection closed by remote host
13:46<warewolf>possibly out of ptys
13:46<warewolf>or fds
13:46<insanum>that is the exact message I recive from lish
13:46<Spads>or dongs
13:47<warewolf>once caker/mikegrb get into it, they should be able to see via /var/log/secure what sshd is flipping out on
13:47<warewolf>actually
13:47<warewolf>you can log in via ssh w/o a pty
13:47<warewolf>not us, via lish though
13:48<warewolf>you can 'ssh host.com bash' and that'll spawn a bash w/o a tty
13:48<warewolf>er pty
13:48<warewolf>very useful for when a box has no more spare ptys
13:48<warewolf>just make damn sure you don't run ncurses apps, or stuff like vim
13:49<insanum>i get the same error message "ssh_exchange_identification: Connection closed by remote host"
13:49<webbles0>I just tried the 'ssh host.com bash' and got the same error message as well
13:49<warewolf>right. Has anyone opened a support ticket yet?
13:49<insanum>I did a little bit ago
13:49|-|coumbes [~coumbes@12.43.95.191] has quit [Quit: Leaving]
13:51<insanum>man, whever my linode goes done I feel empty.... lost... :-)
13:54<webbles0>I'm also wondering about how easy is it to switch from sshd to dropbear. I have debian 3.1. Can I just apt-get remove sshd and apt-get install dropbear? That seems like it will kill my ssh connection first.
13:54<warewolf>start your new sshd on a different port
13:54<warewolf>eg, 222 or 2222
13:55<warewolf>log into that, kill your sshd on 22, and start your new one on 22.
13:55<warewolf>done-and-done.
13:55<webbles0>ok
13:55<gpd>summary of advantages of dropbear? [or is this another lighttpd ;)]
13:56<gpd>Dropbear is particularly useful for "embedded"-type Linux (or other Unix) systems, such as wireless routers.
13:56[~]warewolf -c sleep &
13:58<webbles0>I just heard it was a smaller footprint than sshd
13:58<gpd>seems so... http://matt.ucc.asn.au/dropbear/dropbear.html
13:58<afv-13>gpd: basically a stripped down sshd
13:58<afv-13>less ram usage, most of the basics
13:59<afv-13>can be used as a drop in replacement
13:59<gpd>anything missing that comes to mind?
13:59<afv-13>in 80% of the cases
13:59<afv-13>nothing for me
13:59<gpd>tunnels?
13:59<afv-13>works
13:59<afv-13>and reverse tunnels
13:59<gpd>memory footprint on your linode?
14:00<afv-13>and with cokscrew over a proxy
14:00<afv-13>4% of an 80 plan
14:00<gpd>I have between 1300-2300 RES per sshd process
14:01<afv-13>856 per proccess
14:01<afv-13>(what is RES btw?)
14:01<gpd>the RES column in top
14:01<afv-13>yar, i see the column, just no energy to man top for what it is
14:02<Spads>resident set size
14:02<afv-13>oic
14:02<afv-13>thanks
14:02<Spads>often occasionally abbreviated RSS for maximal confusion
14:02<gpd>ya - as in ps
14:02<Spads>haha
14:03<insanum>so here's the deal with host45... my shutdown job finally executed a half hour later
14:03<Spads>"often occasionally"
14:03<Spads>oops
14:03<insanum>not my boot job is stuck saying "waiting on host"... about 5 minutes now
14:03<gpd>curious that caker and mikegrb are not around - RedHat finished no?
14:03<insanum>and I still can't get to lish... (same error message)
14:04<Spads>gpd: maybe they were abducted by software pirates!
14:05<webbles0>insanum, my linode finally rebooted
14:05<webbles0>but lish is still not working
14:07|-|neil [~neil@74.130.2.170] has quit [Ping timeout: 480 seconds]
14:10|-|neil [~neil@74.130.2.170] has joined #linode
14:18|-|tsi [~tsi@monkey.projectmayhem.org] has quit [Server closed connection]
14:18|-|tsi [~tsi@monkey.projectmayhem.org] has joined #linode
14:25|-|webbles0 [~webbles@cyanide.cs.byu.edu] has quit [Quit: webbles0]
14:25|-|jekil [~alessandr@host103-163.pool8256.interbusiness.it] has joined #linode
14:26<insanum>well my linode on host45 finally booted (30 minutes later) but is running EXTREMELY slow...
14:27<insanum>still can't get to lish
14:32<linbot>New news from forums: slowness on Host 45 in Performance and Tuning <http://www.linode.com/forums/viewtopic.php?t=2312> || liniode on host56 didn't restart in Xen Public Beta <http://www.linode.com/forums/viewtopic.php?t=2323>
14:35|-|Rogue [~ccb500a2@webuser.linode.com] has joined #linode
14:36|-|Rogue changed nick to Guest168
14:41|-|Guest168 [~ccb500a2@webuser.linode.com] has left #linode []
14:45|-|CPL_Rogue [~ccb500a2@webuser.linode.com] has joined #linode
14:48|-|iggy [~iggy@12.45.184.235] has quit [Server closed connection]
14:49|-|iggy [~iggy@12.45.184.235] has joined #linode
14:50|-|cout_ [~cout@li2-82.members.linode.com] has quit [Server closed connection]
14:50|-|cout [~cout@li2-82.members.linode.com] has joined #linode
14:55|-|neil [~neil@74.130.2.170] has quit [Ping timeout: 480 seconds]
14:58|-|neil [~neil@74.130.2.170] has joined #linode
15:04|-|konoko [~Vampire@fhso-hst-240-67.hst.fhso.ch] has quit [Ping timeout: 480 seconds]
15:06|-|taupehat_ [~3fe0c9f4@webuser.linode.com] has quit [Quit: CGI:IRC (Session timeout)]
15:09|-|konoko [~Vampire@80-218-157-65.dclient.hispeed.ch] has joined #linode
15:12<insanum>host45 appears to be out of control.... after my linode booted, I turned off most of my servers except qmail and 'ls' will take 10+ seconds to run
15:13<CPL_Rogue>Yep, i am having probs with 45 to.
15:14<insanum>top shows my cpu usage to be 99% idle
15:16<CPL_Rogue>High load readings with low CPU usage, I/O rate and tokens look fine.
15:16<CPL_Rogue>Cpu(s): 0.6% user, 0.4% system, 0.0% nice, 99.0% idle
15:18<@mikegrb>CPL_Rogue: you are being limited by the io_limiter
15:18|-|linville [~linville@nat-pool-rdu.redhat.com] has quit [Quit: Leaving]
15:21<insanum>here's mine: io_count=174028 io_rate=21 io_tokens=399941 token_refill=512 token_max=400000
15:21<insanum>still overall sluggish
15:29<CPL_Rogue>According to MRTG my I/O problems started about 30 minutes ago... What am I missing?
15:29<CPL_Rogue>How do I tell what is using all the tokens?
15:34<neil>how does my account get activated ?
15:35<@mikegrb>neil: what username?
15:35<@mikegrb>neil: at any rate, I don't see any pending signups so it looks like your account was already activated
15:36<gpd>09:50 neil> holy cow, it worked
15:47|-|neil [~neil@74.130.2.170] has quit [Ping timeout: 480 seconds]
15:50|-|neil [~neil@74.130.2.170] has joined #linode
15:53|-|konoko [~Vampire@80-218-157-65.dclient.hispeed.ch] has quit [Read error: Connection reset by peer]
15:55<neil>hey mikegrb....
15:57|-|Beirdo [~gjhurlbu@beirdo.usercloak.oftc.net] has quit [Server closed connection]
15:57|-|Beirdo [~gjhurlbu@beirdo.usercloak.oftc.net] has joined #linode
16:03|-|spr [~spr@ramona.cs.byu.edu] has quit [Quit: "Ah, the bounce has gone from his bungie"]
16:04|-|Damaja1 [revans@62-64-162-73.dynamic.dial.as9105.com] has joined #linode
16:11|-|jekil [~alessandr@host103-163.pool8256.interbusiness.it] has quit [Ping timeout: 480 seconds]
16:16|-|Beirdo [~gjhurlbu@beirdo.usercloak.oftc.net] has quit [Quit: leaving]
16:16|-|konoko [~Vampire@80-218-157-65.dclient.hispeed.ch] has joined #linode
16:17|-|Beirdo [~gjhurlbu@beirdo.usercloak.oftc.net] has joined #linode
16:24|-|Internat [~internat@c210-49-250-210.ipswc1.qld.optusnet.com.au] has quit [Quit: This computer has gone to sleep]
16:59<linbot>New news from forums: Jabber... anyone got it working? in Linux Networking <http://www.linode.com/forums/viewtopic.php?t=2324>
16:59|-|kokoko [~Vampire@80-218-157-65.dclient.hispeed.ch] has joined #linode
17:00|-|CPL_Rogue [~ccb500a2@webuser.linode.com] has quit [Quit: CGI:IRC (Ping timeout)]
17:00<anderiv>hey mikegrb...did you see my email regarding my CC issues last night?
17:01<anderiv>or caker for that matter...
17:04|-|konoko [~Vampire@80-218-157-65.dclient.hispeed.ch] has quit [Ping timeout: 480 seconds]
17:24|-|tdr [~chatzilla@cyanide.cs.byu.edu] has joined #linode
17:26|-|tdr changed nick to webbles
17:28|-|insanum [~4386fe0a@webuser.linode.com] has quit [Quit: CGI:IRC]
17:32|-|onehitninja [~c75940b3@webuser.linode.com] has joined #linode
17:49|-|alnr [~alan@cpe-69-200-89-172.nyc.res.rr.com] has joined #linode
17:56<webbles>anyone know why host45 is taking forever to reboot
17:56<onehitninja>hiya - is there a problem with the password reset emails? I requested one a while ago, but it hasn't come through yet
18:01<Griswald>webbles: host45's issues have been comming in all day
18:01<Griswald>:X
18:03|-|Ruzza [revans@80-225-13-90.dynamic.dial.as9105.com] has joined #linode
18:04|-|jekil [~alessandr@host35-170.pool8256.interbusiness.it] has joined #linode
18:08|-|Damaja1 [revans@62-64-162-73.dynamic.dial.as9105.com] has quit [Ping timeout: 480 seconds]
18:27|-|Internat [~internat@c210-49-250-210.ipswc1.qld.optusnet.com.au] has joined #linode
18:30|-|Internat [~internat@c210-49-250-210.ipswc1.qld.optusnet.com.au] has quit [Quit: ]
18:30|-|thoth39 [~hm@200.217.82.62] has joined #linode
18:31|-|jekil [~alessandr@host35-170.pool8256.interbusiness.it] has quit [Quit: Leaving]
18:47|-|webbles [~chatzilla@cyanide.cs.byu.edu] has quit [Quit: Chatzilla 0.9.73 [Firefox 1.5.0.4/2006050817]]
18:47|-|ElectricElf [~dbharris@electricelf.noc.oftc.net] has quit [Quit: ]
18:50|-|spr [~spr@c-67-169-251-207.hsd1.ut.comcast.net] has joined #linode
18:55|-|weasel [weasel@weasel.noc.oftc.net] has quit [Quit: Reconnecting]
18:55|-|weasel [weasel@asteria.debian.or.at] has joined #linode
18:56|-|ElectricElf [~dbharris@electricelf.noc.oftc.net] has joined #linode
19:00|-|maher [~maher@modemcable079.241-202-24.mc.videotron.ca] has joined #linode
19:39|-|thoth39 [~hm@200.217.82.62] has quit [Remote host closed the connection]
19:39|-|maher [~maher@modemcable079.241-202-24.mc.videotron.ca] has left #linode [Client exiting]
19:40|-|Internat [~internat@203-206-23-241.dyn.iinet.net.au] has joined #linode
19:40|-|thoth39 [~hm@200.217.82.62] has joined #linode
19:42<Internat>dunno if anyone in here actually cares, but fyi, the #debian offical channel has moved from freenode to oftc, so thats awsome for us :)
19:43<TheFirst>yah and for now it's actually at a size that doesn't make it completely useless
19:44<Internat>exactly :)
19:44<Internat>thres a lota ppl there so that is really useful
19:45<TheFirst>only to a point....past a hundred or two it just devolves to chaos
19:45<TheFirst>lucky to not get kicked off from just joining the channel
19:46<lucca>good way to stress test irc clients I guess
19:47<gpd>try freenode #ubuntu - 893ppl - all talking at once - OMG I JUST INSTALLED UBUNTU AND I HAVE NO IDEA WHAT I AM DOING - HELP ME NOW
19:47<TheFirst>lucca: not really stress testing...it's the /who on join....guess the only test is which clients are smart enough to have a turn that off option
19:56<lucca>hm, this one doesn't have an option to turn On that feature
19:56|-|chris [~chris@nullcode.org] has left #linode []
19:56|-|chris [~chris@nullcode.org] has joined #linode
20:00<encode>http://www.infomaticsonline.co.uk/vnunet/news/2157587/debian-linux-gains-carrier
20:02<TheFirst>lucca: typically it's on by default'
20:07|-|Ruzza [revans@80-225-13-90.dynamic.dial.as9105.com] has quit [Ping timeout: 480 seconds]
20:17|-|Battousai [~bryan@216-164-28-158.c3-0.drf-ubr1.atw-drf.pa.cable.rcn.com] has quit [Quit: Leaving]
20:43|-|sprouse_ [sprouse@c-69-251-154-149.hsd1.md.comcast.net] has joined #linode
20:43|-|sprouse_ [sprouse@c-69-251-154-149.hsd1.md.comcast.net] has quit [Quit: ]
20:54|-|thoth39 [~hm@200.217.82.62] has quit [Remote host closed the connection]
20:55|-|womble [~mpalmer@106.135.233.220.exetel.com.au] has joined #linode
21:00|-|coumbes [~coumbes@12.43.95.191] has joined #linode
21:01|-|Internat [~internat@203-206-23-241.dyn.iinet.net.au] has quit [Read error: Operation timed out]
21:03|-|Internat [~internat@203-206-23-241.dyn.iinet.net.au] has joined #linode
21:18<anderiv>I'm trying to get my ubuntu linode to boot, but I'm getting this: http://pastebin.com/764302. Anyone seen that before? I've been running a gentoo linode for 2 years or so and have never seen this. Any ideas?
21:18<anderiv>I've tried w/ two different 2.6 kernels and they yield the same results.
21:24|-|Internat [~internat@203-206-23-241.dyn.iinet.net.au] has quit [Quit: Leaving]
21:27<@caker>anderiv: hmm, which ubuntu version?
21:27<@caker>anderiv: also, does it work with 2.4?
21:30|-|FireSlash [FireSlash@0-1pool105-34.nas23.kansas-city2.mo.us.da.qwest.net] has joined #linode
21:52[~]caker just realized the date -- 6/6/6
21:52[~]mikegrb pokes caker in the eye
21:52<gpd>going to see The Omen remake?
21:53<@mikegrb>it's in my netflix queue
21:53<TheFirst>caker: yep...make sure to make that black mass today :P
21:55<@caker>TheFirst: they come to me
21:55<gpd>Damn these 'suicide bomber zombie soldiers' in HL2: EP1
21:56[~]gpd exerts further peer pressure for Dapper LTS image [old record]
21:56<@caker>gpd: oh, I built one the other day already (debootstrap)
21:56<@caker>but, my experience was a dist-upgrade worked perfectly
21:56<@caker>YMMV
21:57<gpd>ya - i am already there via sagdu - however
21:57<gpd>you should ride the marketing / hype / bandwagon to fame and fortune
21:57<@mikegrb>yes, dist-upgrade working is the norm
21:57<@mikegrb>if it doesn't then it's broken and needs a bug filed so it can be fixed
21:57<gpd>the amount of activity on #ubuntu is quite scary...
21:58<@caker>on oftc?
21:58<gpd>ppl asking for Dapper Dedicated / VPS quite regularly...
21:58<gpd>no - on freenode :(
21:58<gpd>the Long Term Support and easy LAMP install has attracted converts
21:58<gpd>[and retards] >{
21:59<@mikegrb>long term support is just marking
21:59<@mikegrb>saddens me to see
21:59<gpd>ya - marketing - but there's money in them thar hills for you blokes
21:59<@mikegrb>makes me think less of canonical
21:59[~]gpd wonders why he cares about mikegrb's bank balance
22:00<gpd>actually not entirely marketing as the previous releases have very short term support
22:00<gpd>eg. hoary already dropped iirc
22:00<@mikegrb>but it is a a non issue
22:00|-|coumbes [~coumbes@12.43.95.191] has quit [Quit: Ex-Chat]
22:00<@mikegrb>so? there is no reason to still be using hoary
22:00<gpd>but 5yr is a significant commitment for Shuttleworth
22:01<@mikegrb>it's silly
22:01<gpd>given Ubuntu in theory could be vapour if he pulls out
22:01<@mikegrb>nah
22:01<@mikegrb>this isn't redhat, ubuntu would keep on going
22:01<gpd>it also gives people in suits required gladness value
22:02<gpd>well it would probably just merge back into debian with no money for devs
22:02<@mikegrb>suits shouldn't be making deceicions they don't understand
22:02<@mikegrb>any sysadmin worth his pay can tell the suit why it doesn't matter
22:03<gpd>true, but things like LTS can give leverage to lesser-BOFHs
22:03<@mikegrb>...
22:04<@mikegrb>I'll quit talking since you aren't reading what I'm saying.
22:14|-|darkbeholder [darkbehold@nmathe02.res.csu.edu.au] has quit [Read error: Connection reset by peer]
22:14|-|darkbeholder [darkbehold@nmathe02.res.csu.edu.au] has joined #linode
22:15[~]anderiv steps back in the room...
22:15<anderiv>caker: ubuntu 5.10...brand new distro image. I'll give 'er a try with a 2.4 kernel
22:17|-|Internat [~internat@203-206-23-241.dyn.iinet.net.au] has joined #linode
22:20<anderiv>caker: I get nothing on the lish console w/ 2.4
22:20<anderiv>...when I boot
22:20<gpd>do you have the right root image set in your config?
22:20<@caker>anderiv: logview will show you *something*, in case you missed it
22:20[~]anderiv checks
22:21<anderiv>[anderiv2@host18 lish]# logview
22:21<anderiv>Showing last 250 lines from previous boot
22:21<anderiv>-----------------------------------------
22:21<anderiv>Showing last 100 lines from current boot
22:21<anderiv>-----------------------------------------
22:21<anderiv>Cannot exec '/vbin/kernel/2.4.29-rc1-linode38': No such file or directory
22:21<anderiv>gpd: yes I do.
22:21<@caker>hmm
22:27<@caker>anderiv: that should be fixed .. the path for that kernel was wrong (!)
22:27<anderiv>doh :-)
22:27<anderiv>thanks.
22:28<anderiv>also...I submitted a request to get move datacenters if possible. I'm not sure if there are any 120s free at TP, but if there are, I guess now would be a good time to re-provision my account since I haven't made any changes to the linode. The disk images wouldn't even need to get transferred.
22:29[~]Eman laughs evily as anderiv gets put on host51
22:30<anderiv>@#$!
22:31<@caker>very weird
22:31<@caker>!errno 3
22:31<linbot>caker: ESRCH (#3): No such process
22:31<@caker>anderiv: have you tried the -linode39 kernel yet?
22:32<anderiv>trying now...
22:32<anderiv>ubd7: Can't open "<NULL>": errno = 14
22:33<anderiv>I got that about 150 times, but it looks like its booting.
22:33<@caker>crazy
22:33<@caker>it's finished
22:33<anderiv>yah - it's running
22:39<anderiv>caker: any word on possible migration to TP or should I wait for a reply to the ticket?
22:41<@caker>anderiv: one moment
22:41<anderiv>np
22:41<@caker>anderiv: re-login to see the migration -- delete those disk images you don't want migrated
22:42<@caker>in the meantime, I'll re-assign you an IP
22:42<anderiv>caker: great - thanks.
22:42<@caker>anderiv: actually, hold for a sec
22:42<anderiv>ok
22:45<@caker>anderiv: good to go
22:45<anderiv>great - thanks caker.
22:47|-|spr [~spr@c-67-169-251-207.hsd1.ut.comcast.net] has quit [Remote host closed the connection]
22:56|-|onehitninja [~c75940b3@webuser.linode.com] has quit [Quit: CGI:IRC (Session timeout)]
22:59|-|VS_ChanLog [~stats@ns.theshore.net] has left #linode [Rotating Logs]
22:59|-|VS_ChanLog [~stats@ns.theshore.net] has joined #linode
23:00|-|Razathorn [~Razathorn@64-126-84-15-dhcp-kc.everestkc.net] has joined #linode
23:01<Razathorn>I'm trying to work on a linode a friend and I have setup and it's just plain crawling -- there's no cpu usage yet the uptime is like 3+ and all i/o is taking for ever. A tar zc has been taking like an hour and should only be about 42 megs when done, all commands take 10+ seconds to return
23:02<anderiv>Razathorn: what else is running on the linode?
23:03<Razathorn>apache, imap, sendmail, bind... ps and top show no cpu usage -- I can watch vmstat and nothing, I mean nothing outside of a few io blips is going on
23:03<Razathorn>then when you try and do anything, it feels like it's swapping, but it's not
23:03<anderiv>what does `cat /proc/io_status` look like?
23:03<FireSlash>http://www.fireslash.net/index/pages/onepost/57/ :/
23:04<Razathorn>io_count=1074268 io_rate=58 io_tokens=-30 token_refill=50 token_max=100
23:04<anderiv>FireSlash: heh. No apocalypse, huh?
23:04<eFUDd>-30!
23:04<FireSlash>Yeah. Boooooorrriiing
23:04<anderiv>Razathorn: doh.
23:04<Razathorn>doh what?
23:05<anderiv>Razathorn: that value should be ~400k
23:05<Razathorn>i'm not familliar with io_status
23:05<FireSlash>Razathorn, I suspect your server is indeed swapping :P
23:05<Razathorn>my server or my uml ?
23:05<anderiv>your linode.
23:05<FireSlash>linode, server, same thing :)
23:05<Razathorn>no way
23:05<Razathorn>vmstat shows no so/si
23:06<FireSlash>copy/paste your vmstat output, just for shits and giggles.
23:06<Razathorn>k
23:07<Razathorn>finch:/var/log# free
23:07<Razathorn>getting vmstat
23:07<FireSlash>Throw me free too.
23:07<Razathorn>^^^ free
23:07<FireSlash>....
23:07<FireSlash>Theres no output
23:07<anderiv>hah
23:07<anderiv>his server is very slim on RAM usage.
23:08<Razathorn>slim?
23:08<FireSlash>"Razathorn finch:/var/log# free
23:08<FireSlash>Razathorn getting vmstat"
23:08<Razathorn>i pasted it
23:08|-|dsoul [darksoul@vice.ii.uj.edu.pl] has quit [Ping timeout: 480 seconds]
23:08<Razathorn>prolly this java client
23:08<Razathorn>lemme login with a more familliar irc client
23:08<Razathorn>brb
23:08<FireSlash>use a pastebin then
23:08<Razathorn>how about i use irssi ;)
23:08|-|Razathorn [~Razathorn@64-126-84-15-dhcp-kc.everestkc.net] has quit [Quit: Razathorn]
23:08<anderiv>w00t
23:09<FireSlash>http://www.fireslash.net/index/pages/serverinfo <-- Fine tuned to perfection methinks <3
23:09<anderiv>FireSlash: nice work.
23:09|-|Razathorn [~raz@64-126-84-15-dhcp-kc.everestkc.net] has joined #linode
23:09<Razathorn>regards
23:09<Razathorn>the joys of a terminal
23:09<Razathorn>here goes
23:10[~]anderiv won't touch anything other than irssi for IRC :-)
23:10<FireSlash>anderiv, Thats with LAMP, and Apache 2 on the side for SVN.
23:10<Razathorn>finch:/var/log# free
23:10<Razathorn> total used free shared buffers cached
23:10<Razathorn>Mem: 155544 152744 2800 0 6236 64800
23:10<Razathorn>-/+ buffers/cache: 81708 73836
23:10<Razathorn>Swap: 525304 0 525304
23:10<anderiv>FireSlash: wow
23:10<Razathorn>finch:/var/log# vmstat 10
23:10<Razathorn>procs -----------memory---------- ---swap-- -----io---- --system-- ----cpu----
23:10<Razathorn> r b swpd free buff cache si so bi bo in cs us sy id wa
23:10<Razathorn> 2 1 0 2944 6284 64576 0 0 30 5 137 22 0 1 98 0
23:10<Razathorn> 0 4 0 2296 6304 65184 0 0 55 5 150 34 0 0 100 0
23:10<Razathorn> 0 4 0 2824 6316 64632 0 0 56 4 150 33 0 0 100 0
23:10<Razathorn> 0 3 0 2496 6320 64940 0 0 21 8 150 31 0 0 100 0
23:10<FireSlash>wtf
23:10<Razathorn> 0 4 0 3096 6324 64344 0 0 53 9 150 32 0 0 100 0
23:10<Razathorn> 0 6 0 2484 6340 64872 0 0 47 4 150 35 0 0 100 0
23:10<Razathorn>we had to reboot it via the console earlier
23:10<FireSlash>How are you out of IO tokens.
23:10<Razathorn>it's plain eff'd up
23:10<Razathorn>black box man, black box
23:10<Razathorn>I can stop all my procs
23:10<Razathorn>the servers worthless now
23:10<FireSlash>Check for strange processes in the background?
23:10<Razathorn>heh
23:11<Razathorn>lemme shut everything down outside of sshd
23:11<anderiv>yah...checked for rootkits recently?
23:11<FireSlash>like "HAXORING_UR_LINODE"
23:11<FireSlash>;)
23:11<Razathorn>I'm not the admin of this box, but it should be fairly up to date as the guy who runs it apt-get update all the time
23:11<anderiv>...not always a good thing :-)
23:12<FireSlash>Check anyone. apt-get update does NOT imply the server is secure
23:12<Razathorn>of course
23:12<Razathorn>you know what I'm saying
23:12[~]anderiv goes to get a beer
23:12<FireSlash>Took me 4 tries to get a paranoid enough config to get a (what I hope is) hax proof server
23:13<anderiv>http://ratebeer.com/Beer/summit-maibock/9940/
23:14<anderiv>let's hear it for 7.1% ABV :-)
23:14<anderiv>FireSlash: what distro do you run?
23:14<FireSlash>anderiv, Debian (small).
23:14<Razathorn>how can I tell what linode host I'm on
23:15<Razathorn>I saw a thread where people were having issues on host 45
23:15<FireSlash>When you ssh into lish, its host##(<-- RIGHT HERE).linode.com
23:15<Razathorn>I don't have that info
23:15<anderiv>Razathorn: it's listed in LPM
23:15<FireSlash>But theres probably a ebtter way
23:15<Razathorn>can I tell from my ip ?
23:15<Razathorn>I don't have the console info
23:15<anderiv>Razathorn: no.
23:15<Razathorn>all I have is root on the box itself
23:16|-|TuneUp [~upndown@adsl-70-231-239-40.dsl.snfc21.sbcglobal.net] has joined #linode
23:18<Razathorn>i'm at a loss
23:18<Razathorn>everything except dhclient, cron, syslog, and sshd is down
23:18<Razathorn>this box had a fairly tight firewall as well
23:19<Razathorn>it's not like it was a default install with everything listening here
23:19<Razathorn>I seriously doubt it's been rooted
23:19<Internat>whats the problem?
23:19<anderiv>Internat: he has io_count=-30 :-(
23:21<Razathorn>the guy who's linode this actually is is impossible to get ahold of after 7pm so I'm out of luck with anything that I cannot do on the box itself
23:21<Razathorn>ie, the linode uml
23:21<Internat>a bit nasty indeed
23:22<anderiv>Razathorn: FYI...the term "linode" always refers to *your* server. "Host" refers to the host box.
23:22<Razathorn>i'm up to 100 tokens
23:22<Internat>thats still nasty
23:22<Razathorn>hehe
23:22<Razathorn>i can empty it pretty quick with an ls command it seems
23:22<Internat>can u show me io_status again
23:22<anderiv>Razathorn: have you rebooted recently? If so, did that fix the problem or did it start sucking tokens right away?
23:23<Razathorn>it was rebooted on the console today
23:24<Razathorn>I don't know if it started sucking tokens right away as I just found out what tokens were only a minute ago
23:24<Internat>can u paste the output of /proc/io_status again
23:24<anderiv>oh - okay
23:24<Razathorn>yes
23:24<Razathorn>finch:~# cat /proc/io_status
23:24<Razathorn>io_count=1106669 io_rate=0 io_tokens=100 token_refill=50 token_max=100
23:24<eFUDd>i would tell you to buy a li-320
23:24<Internat>wtf
23:24<eFUDd>but then you8 might end up on my host
23:24<eFUDd>so...
23:24<Internat>ur token max is only 100
23:24<Razathorn>nothing is running on this box
23:24<Razathorn>i mean the websites that are on it
23:24<Razathorn>nobody hits em
23:24<Internat>umm that should be a lot bigger
23:24<anderiv>Internat: haha - good find.
23:24<Internat>nf@broken:~$ cat /proc/io_status
23:24<Internat>io_count=5639274 io_rate=36 io_tokens=399981 token_refill=512 token_max=400000
23:24<anderiv>Internat: that's a bug, huh :-)
23:25<Razathorn>right now, only essential services are running
23:25<Razathorn>like sshd
23:25<Internat>yeah ur going to need to get a hold of caker or mikegrb
23:25<Razathorn>and the magical [kernel dorks]
23:25<anderiv>yah - your token_max should be 400k.
23:25<Internat>no wonder ur running outa tokens shit
23:25<anderiv>Razathorn: http://www.linode.com/wiki/index.php/IO_Tokens <---- FYI
23:25<Razathorn>so, the current theory is somone rooted my box and stoled allz my t0kenz!
23:25<Razathorn>heh
23:25<@mikegrb>lolz
23:25<Internat>lol not quite
23:26<Razathorn>yeah I know ;)
23:26<Internat>more that something is fubar when ur ebooted
23:26<eFUDd>er
23:26<Internat>id recommed rebooting agan
23:26<eFUDd>his host was probably doing something STUPID
23:26<Razathorn>if I could
23:26<Razathorn>I would
23:26<Internat>but since u dont have access to reboot
23:26<eFUDd>so was tuned down
23:26<eFUDd>end-of-story
23:26<Razathorn>I mean I can init 6
23:26<Razathorn>but last I checked
23:26<Razathorn>that's not a real reboot in uml land
23:26<Internat>nah that doesnt do anything on a linode.. it just breaks
23:26<anderiv>Razathorn: that's correct.
23:26<Razathorn>ding ding
23:26<Razathorn>well poop in a dry toliet
23:26[~]Razathorn *shucks*
23:27<Internat>dunno if it would be tuned down that much if it was doing sojething stupid would it?
23:28<Razathorn>basically, the box ground to a halt today
23:28<Razathorn>the guy rebooted it
23:28<Razathorn>it's never been happy since
23:29<Razathorn>I didn't see the proc / vmstat / a damn thing during this time
23:29<Razathorn>but he said the uptime was like 6 but nothing was going
23:29<Razathorn>so... i'm at a loss as to why the box (my linode) went south to begin with
23:30<anderiv>Razathorn: 6/6/06 my friend. ;-)
23:30<Internat>ah
23:30<Internat>but in australia its 7/6/06 so :P
23:30<anderiv>meh
23:31<anderiv>Internat: how's the latency to your linode from there?
23:31<Razathorn>holy insecurity batman
23:31<Razathorn>I guessed his linode.com login and password
23:31<Razathorn>rock
23:31<anderiv>Razathorn: heh - nice work.
23:31<anderiv>Razathorn: same as the root pw?
23:31<Razathorn>I knew his username and some of his potential passwords
23:31<Razathorn>oh no way
23:31<Razathorn>haha
23:31<Internat>umm i wouldnt admit that here but anyway
23:31<Internat>umm my latency is.. *checks*
23:32<Razathorn>they're not easy to guess passwords
23:32<Razathorn>i just know em
23:32<Razathorn>he was the best man at my wedding
23:32<Razathorn>you could say I've been on a few of his boxes ;)
23:32<anderiv>Razathorn: Kansas City, huh?
23:32<Internat>PING broken.our-lan.com (70.85.31.161): 56 data bytes
23:32<Internat>64 bytes from 70.85.31.161: icmp_seq=0 ttl=51 time=218.0 ms
23:32<Internat>64 bytes from 70.85.31.161: icmp_seq=1 ttl=51 time=216.8 ms
23:32<Internat>64 bytes from 70.85.31.161: icmp_seq=2 ttl=51 time=216.3 ms
23:32<Razathorn>when we were in college I would start netscapes on his linux box X display with porn on them when his mom was in the room
23:32<@mikegrb>lolz
23:32<Razathorn>it was lol
23:32<Razathorn>trust
23:33<Razathorn>yeah kc
23:33<Razathorn>welp, I'm gonna login and reboot this bad boy and see if I get more than 100 iotokens
23:33<anderiv>dig it. The in-laws live there. Up by Gladstone.
23:33<Razathorn>up?
23:33<Internat>find out what host ur on while ur there, its in the network information
23:33<Razathorn>hehe, that's south for me ;)
23:34<Internat>cause then u can connect via lish and issue a restart
23:34<anderiv>heh...well it's north of the city.
23:34<Razathorn>i'm thinking of grandview
23:34<FireSlash>Dooo de dooo...
23:34<Razathorn>dohs it is north
23:34<FireSlash>After unpacking 32.8kB of additional disk space will be used.
23:34<FireSlash>0% [Waiting for headers]
23:35<anderiv>Razathorn: y'all have some good BBQ there. It inspired me to buy a smoker :-)
23:35<FireSlash>HEH
23:35<Razathorn>so wait a sec
23:35<Razathorn>where's the host info?
23:35<FireSlash>So, my mail server went on the fritz for a while when I was toying around with courier.
23:35<Razathorn>I am missing it in network info
23:35<FireSlash>So I fixed it...
23:35<anderiv>you on the LPM site?
23:35<FireSlash>51 system messages
23:35<Razathorn>indeed
23:35<Razathorn>at least I think
23:35<anderiv>right side - right below "Host Summary"
23:35<FireSlash>All cron error reports. I should really fix that darned cron job :P
23:36<Razathorn>host 45
23:36<Razathorn>see it wasn't in my head!
23:36<Internat>yeah half way down the side is network information :P
23:36<Razathorn>ahaha I have averaged 0 cpu for the last month
23:36<Razathorn>it says I have utilized all my space
23:36<Razathorn>but df says otherwise
23:36<Razathorn>what gives there?
23:36<@caker>Razathorn: all of it assigned to disk images
23:37<Razathorn>oh
23:37<anderiv>Razathorn: that means that the disk images your node's owner set up take up all of his allocated space.
23:37<Razathorn>so a nomenclature problem
23:37<Internat>non
23:37<anderiv>Razathorn: there can still be free space within the images.
23:37<Razathorn>gotcha
23:37<Internat>no cause u might have different images that u boot
23:37<Internat>each one would have its own space
23:37<Razathorn>so basically I've got 100% of my available space dedicated to disk images
23:37<Internat>yep
23:37<anderiv>Razathorn: you been to either Gates or Arthur Bryant's down there?
23:38<Razathorn>gates
23:38<Razathorn>yeah
23:38<Razathorn>well
23:38<Razathorn>no
23:38<Razathorn>hahaha I have their sauce in my fridge
23:38<Razathorn>I am a big smokehouse fan
23:39<anderiv>heh - it's good sauce.
23:39<Razathorn>it's a chain, but it's a damn good chain with second to none beans
23:39<anderiv>ooh yah
23:39<Razathorn>one time I was eating there and I tried to talk to my friend who was eating beans
23:39<Razathorn>he looked up and said
23:39<Internat>so did the reboot work?
23:39<Razathorn>"BEANS!!!!!!!!!"
23:39<Razathorn>didn't try it yet
23:39<Razathorn>ssh'd to host
23:39<Razathorn>it was slow as heck to get a login prompt
23:39<Internat>yeah
23:39<Internat>press ctrl+a the d
23:39<Internat>to detact from the screen session
23:39<Razathorn>to detach?
23:39<Internat>and look at the lish stuff
23:40[~]Razathorn is a screen junkie
23:40<anderiv>Razathorn: ditto
23:40<Razathorn>whoa, enter in lish == screen -dR
23:41<Razathorn>i just sent init 0
23:41<Razathorn>wonder if it has enough io_tokens to pull that off
23:41<Razathorn>can I boot in single user mode?
23:42<anderiv>yah - you have to set it in LPM, though.
23:42<Razathorn>what does lpm stand for?
23:42<anderiv>umm Linode Platform Manager?
23:42<anderiv>maybe....
23:42<anderiv>ask caker
23:42<anderiv>it's the management website.
23:42<Razathorn>so I can create a new config
23:43<Razathorn>will that let me pass params?
23:43<Razathorn>like "1"
23:43<Razathorn>heh
23:43<anderiv>you can just modify the current config, if you want.
23:43<Internat>what did a reboot do?
23:43<Internat>did it help?
23:43<Razathorn>figured it
23:43<Razathorn>i didn't reboot
23:43<Internat>u have to reboot from the lish console or the webpage
23:43<Razathorn>it's haulted
23:43<Razathorn>halted
23:43<Razathorn>i just set it to single mode
23:43<Razathorn>now booting
23:44<Internat>that wont boot it properly
23:44<Razathorn>wha?
23:44<Internat>that wont fix io_tokens cause i believe there sent when it boots by the host
23:44<Razathorn>why not?
23:44<Razathorn>I am in lish
23:44<Razathorn>i typed boot
23:44<Razathorn>after changing config in lpm
23:44<Razathorn>i'm in single user mode
23:44<Internat>as long as u have rebooted from either the console or the webpage it should
23:45<Razathorn>so what did you think wouldn't work
23:45[~]Razathorn is really confused
23:45<Internat>cause u sayd u didnt reboot it from the webpage or the console
23:45<Razathorn>oh
23:45<Razathorn>haha, I init 0 from inside the linode
23:45<Razathorn>then change config
23:45<Razathorn>boot from lish
23:45<Razathorn>li12-170:~# cat /proc/io_status
23:45<Razathorn>io_count=9290 io_rate=0 io_tokens=88086 token_refill=512 token_max=400000
23:45<Internat>thats better
23:46<Razathorn>indeed
23:46<Internat>now ur computer might work properly
23:46<Razathorn>changing config file back to default run level
23:46<Razathorn>here goes nothing, err, everything!
23:46[~]Razathorn boots
23:51<Razathorn>so far so good
23:51<Razathorn>i have tokens
23:51<Razathorn>and guess what, I can type!
23:51<FireSlash>zomg
23:51|-|FireSlash [FireSlash@0-1pool105-34.nas23.kansas-city2.mo.us.da.qwest.net] has quit [Quit: Leaving]
23:51[~]Razathorn tries that tar zcpf again ;)
23:52<@mikegrb>lolz
23:52<Internat>lol yeah it might work a tad better
23:52<Razathorn>du -shc is anything but fast
23:53|-|TuneUp [~upndown@adsl-70-231-239-40.dsl.snfc21.sbcglobal.net] has quit [Read error: Connection reset by peer]
23:53|-|afv-13 [~afv-13@rkdp-ip-nas-1-p361.telkom-ipnet.co.za] has quit [Ping timeout: 480 seconds]
23:59<Razathorn>sweet, well everything seems fine now, tar finished, transferring backup -- any idea what caused my tokens to go to 100 ?
23:59<Razathorn>was that something we did?
23:59<Razathorn>I certianly don't remember running ./na_thanks_i_dont_need_no_tokens.sh ;)
---Logclosed Wed Jun 07 00:00:03 2006