Back to Home / #linode / 2005 / 08 / Prev Day | Next Day
#linode IRC Logs for 2005-08-27

---Logopened Sat Aug 27 00:00:36 2005
00:12|-|FireSlash [Liger@125.207-191-238.gbrdialin.gbronline.com] has quit [Read error: Connection reset by peer]
00:21|-|JonR [~jon@mn-10k-dhcp2-190.dsl.hickorytech.net] has left #linode [Leaving]
00:22|-|tj [~tj@65-100-5-175.eugn.qwest.net] has quit [Quit: leaving]
00:22<@linbot>New news from forums: maek exim with mysql in Email/SMTP Related Forum <http://www.linode.com/forums/viewtopic.php?t=1840>
00:31<taupehat>maek?
00:36<Battousai>exim?
00:39<@mikegrb>exim rocks my socks
00:40<Battousai>exim is teh lunix bad
00:42<taupehat>1heh
00:45|-|spr [~spr@c-24-10-237-76.hsd1.ut.comcast.net] has quit [Quit: Spoon!]
00:53<@linbot>New news from forums: make exim with mysql in Email/SMTP Related Forum <http://www.linode.com/forums/viewtopic.php?t=1840>
01:36|-|Newsome [~sorenson@216-190-206-130.customer.csolutions.net] has left #linode [Leaving]
01:55|-|besonen [~besonen@dsl-db.pacinfo.com] has quit [Quit: Leaving]
03:57|-|adamg [~misthos@zeus.misthos.com] has joined #linode
04:08|-|emcnabb [~emcnabb@67-41-182-102.slkc.qwest.net] has quit [Ping timeout: 480 seconds]
04:18|-|emcnabb [~emcnabb@67-41-182-102.slkc.qwest.net] has joined #linode
05:30|-|Ascii [GZnycm9E6Z@panix5.panix.com] has joined #linode
06:50|-|whice [~whice@as-71.ar43.vc.ukrtel.net] has quit [Ping timeout: 480 seconds]
07:23|-|alone_boy [~Firari@85.96.151.20] has joined #linode
07:29|-|Jeremy [stormy@crstn.bc.ipv6.smart-serv.net] has quit [Ping timeout: 480 seconds]
07:30|-|Jeremy [stormy@crstn.bc.ipv6.smart-serv.net] has joined #linode
07:30|-|alone_boy [~Firari@85.96.151.20] has quit [Quit: -\xBB\xAB- Ka\xA2ak v3 -\xBB\xAB- www.Hayta.net & www.Kacak.net -\xBB\xAB--]
07:38|-|Jeremy [stormy@crstn.bc.ipv6.smart-serv.net] has quit [Ping timeout: 480 seconds]
08:10|-|Jeremy [stormy@crstn.bc.ipv6.smart-serv.net] has joined #linode
08:17|-|NeonNero [neonnero@home.neonnero.net] has joined #linode
08:27|-|NeonNero [neonnero@home.neonnero.net] has quit [Quit: Everybody who believes in telekinesis, raise my hand!]
08:31|-|spr [~spr@c-24-10-237-76.hsd1.ut.comcast.net] has joined #linode
09:00|-|tompuppy [~tom@proxy.tom.kao.name] has joined #linode
09:00<tompuppy>!seen caker
09:00<@linbot>tompuppy: caker was last seen here 1 day, 11 hours, 19 minutes, and 55 seconds ago saying: <caker> 2.6 vs 2.4 ?
09:01<tompuppy>!seen mikegrb
09:01<@linbot>tompuppy: mikegrb was last seen here 8 hours, 21 minutes, and 27 seconds ago saying: <mikegrb> exim rocks my socks
09:05<heidi>they are sleeping
09:23|-|ronpoz [~ronpoz@ool-45700306.dyn.optonline.net] has joined #linode
09:24|-|FireSlash [Liger@105.207-191-238.gbrdialin.gbronline.com] has joined #linode
09:25|-|Guest43 [~misthos@zeus.misthos.com] has joined #linode
09:25|-|adamg [~misthos@zeus.misthos.com] has quit [Read error: Connection reset by peer]
09:37<tompuppy>heidi: i see
09:39|-|adamg [~misthos@zeus.misthos.com] has joined #linode
09:39|-|Guest43 [~misthos@zeus.misthos.com] has quit [Read error: Connection reset by peer]
09:51|-|Guest43 [~misthos@zeus.misthos.com] has joined #linode
09:52|-|adamg [~misthos@zeus.misthos.com] has quit [Read error: Connection reset by peer]
10:06|-|FireSlash [Liger@105.207-191-238.gbrdialin.gbronline.com] has quit [Read error: Connection reset by peer]
10:34|-|besonen [~besonen@dsl-db.pacinfo.com] has joined #linode
10:34|-|Newsome [~sorenson@216-190-206-130.customer.csolutions.net] has joined #linode
10:49|-|dc0e [~dc0e@pcp0010756487pcs.howard01.md.comcast.net] has quit [Quit: going down]
11:08|-|Ascii [GZnycm9E6Z@panix5.panix.com] has left #linode []
11:10|-|jekil [~alessandr@151.82.0.153] has joined #linode
11:11<jekil>the linode renew and credit card payment is automatic?
11:13<fuzzie>yes
11:15<jekil>thanks
11:16|-|joe changed nick to Guest586
11:18|-|tizen [tiz@ADent5754.ResNet.Dal.Ca] has joined #linode
11:22|-|whice [~whice@as-71.ar43.vc.ukrtel.net] has joined #linode
11:32|-|dddd44 [~dhb55@SPARCed.org] has joined #linode
11:32<dddd44>excuuse me i have an ip address.. but how do i get the dns name for this ip address??anyone??
11:35<@mikegrb>"host <ipaddress>"
11:40<dddd44>what about in windows version??
11:42|-|dddd44 [~dhb55@SPARCed.org] has quit [Quit: Leaving]
11:42<fuzzie>heh.
11:46|-|Guest43 changed nick to adamg
11:54|-|dddd44 [~dhb55@tor-proxy.qnet.com] has joined #linode
11:54<dddd44>can anyone tell me the hostname for this???218.111.166.30 thank
11:54<fuzzie>it doesn't resolve.
11:54<dddd44>how you resolve??
11:55<fuzzie>30.166.111.218.in-addr.arpa not found: 3(NXDOMAIN)
11:55<fuzzie>and, you contact your ISP.
11:57<jekil>if I use all 100% month cpu? the linode is shutted down?
12:00<iggy>I wish
12:00<fuzzie>you'll probably hit memory/disk limits far before you hit cpu limits
12:01<jekil>fuzzie: 1 month, linode idle, 13% used
12:01<fuzzie>it obviously isn't idle if you've used 13%
12:02<fuzzie>i've used 2% of a host CPU on a fairly active server
12:02<iggy>that's for the month so far
12:02<jekil>fuzzie: idle=no services, only munin for monitoring
12:03<iggy>why are you monitoring a server that runs no services?
12:03<fuzzie>take a look to see if there are any processes which have used a lot of cpu time?
12:03<jekil>fuzzie: no, there isn't
12:03<fuzzie>if you're going by the web interface, it might be giving you inaccurate results, you'd want to ask mikegrb
12:03<iggy>better yet, why are you running a server that runs no services
12:04<fuzzie>which reminds me
12:04<fuzzie>is there a way to get the bandwidth usage data for my linode via a script or something?
12:05<fuzzie>i'd like to code something which gives me a warning if i near the limit
12:05<iggy>yeah, there's something in the forums about it
12:05<jekil>fuzzie: yes, i am reading this 13% on website
12:05<fuzzie>ah, i'll go look there
12:05<jekil>fuzzie: before was 1%, and one day i read 12%
12:05<iggy>fuzzie: search terms "xml interface" maybe
12:06|-|whice [~whice@as-71.ar43.vc.ukrtel.net] has quit [Ping timeout: 480 seconds]
12:06<fuzzie>http://www.linode.com/forums/viewtopic.php?t=776 :-)
12:08|-|kas_3 [U2FsdGVkX1@60.36.181.86] has joined #linode
12:08<@mikegrb>fuzzie: http://www.linode.com/forums/viewtopic.php?t=776
12:08<@mikegrb>oh
12:08<@mikegrb>too slow I am
12:08<@mikegrb>xml has changed a bit, more data in it
12:09<fuzzie>looks good
12:09|-|whice [~whice@81.30.163.26] has joined #linode
12:10<fuzzie>once every two hours is the recommended time between requests?
12:15|-|dddd44 [~dhb55@tor-proxy.qnet.com] has quit [Ping timeout: 480 seconds]
12:25|-|linkz [~jestha@p6847-adslbkksp7.C.csloxinfo.net] has joined #linode
12:25|-|linkz [~jestha@p6847-adslbkksp7.C.csloxinfo.net] has left #linode []
12:27|-|kas_3 [U2FsdGVkX1@60.36.181.86] has quit [Remote host closed the connection]
12:28|-|kas_3 [tor@c212-151-198-78.swipnet.se] has joined #linode
12:36|-|dddd44 [~dhb55@24-52-166-190.lndnnh.adelphia.net] has joined #linode
12:43|-|kas_3 [tor@c212-151-198-78.swipnet.se] has quit [Ping timeout: 480 seconds]
12:49<dddd44>excuse me can i use sock to connect to internet instead of http proxy???
12:49<dddd44>]
12:51<[|^__^|]>dude, linodes don't need socks or proxies, dddd44
12:52<SupaZubon>linodes are all hosted right on a DMZ-type fat pipe
12:55|-|kas_3 [~dhb55@71-32-45-14.tcsn.qwest.net] has joined #linode
12:55|-|kas_3 [~dhb55@71-32-45-14.tcsn.qwest.net] has quit [Remote host closed the connection]
13:00|-|dddd44 [~dhb55@24-52-166-190.lndnnh.adelphia.net] has quit [Ping timeout: 480 seconds]
13:06|-|JonR [~jon@mn-10k-dhcp2-190.dsl.hickorytech.net] has joined #linode
13:14<JonR>goddamn apache/mysql.
13:15<JonR>I can't even frigging log in.
13:17<Battousai>lish
13:17<Battousai>it is teh your friend
13:17<JonR>yeah
13:17<@caker>Filename Type Size Used Priority
13:17<@caker>/dev/ubdb partition 132088 132084 -1
13:17<@caker>OK io_count=7990422 io_rate=599 io_tokens=-152 token_refill=512 token_max=400000
13:17<@caker>^-- JonR
13:17<JonR>caker: yeah, I'm aware.
13:18<JonR>I just want to run ps aux and then kill some stuff quick.
13:18<JonR>I don't really understand, I've been trying to keep apache/mysql on a diet, but it's not working.
13:18<@caker>JonR: try logging in now
13:18[~]JonR does.
13:19<@caker>JonR: apache 2? perhaps you placed limits under the wrong MPM
13:19<JonR>I'm still out of ram/swap, though, so even with the io lifted...
13:19<iggy>look at caker's my.cnf and turn off the 40 modules in httpd.conf that you don't need
13:19<JonR>no, apache 1.3
13:19<iggy>those are good starts
13:19<fuzzie>we went through the my.cnf thing yesterday
13:19<JonR>yeah
13:19<iggy>ok
13:20<JonR>I used to have both apache and apache-ssl running... And it was fine. I stopped running apache-ssl, since I realized I don't need it, and I get this.
13:20<fuzzie>i'd be suspicious of something hammering your site and you not limiting the number of apache processes heavily enough
13:20<JonR>fuzzie: I don't think apache 1.3 has a max server processes limit...
13:20<JonR>I've got mod_dosevasive, though.
13:20<@caker>JonR: it does: MaxClients
13:20<fuzzie>it does.
13:20<JonR>hm.
13:20<fuzzie>it defaults to 50
13:21<@caker>yeah, it's 150 on RH-based distros ..
13:21<fuzzie>which is going to cause you big problems if something decides to use them all.
13:21<fuzzie>yikes
13:21<JonR>Anyhow, I've got StartServers 1, MinSpareServers 1, MaxSpareServers 2, iirc.
13:21<JonR>okay, I'm in.
13:21<@mikegrb>you also need to set maxclients
13:21[~]JonR nods.
13:21<fuzzie>but, i'd definitely check the http logs to see if you're being hammered when you have the problems
13:21<fuzzie>and MaxClients is the important one
13:22<JonR>Well, I'm logged in... no bash prompt yet. :/
13:23<JonR>aha, prompt
13:24<JonR>fuck it, i'm sure it's apache, i'll just frigging reboot
13:31<JonR>meh
13:32<JonR>I had MaxClients 64
13:34|-|dddd44 [~dhb55@dip-107-127.centaurus.dsl.breisnet.com] has joined #linode
13:37<JonR>Is 10 a good number for MaxClients?
13:37<Battousai>clients?
13:37<JonR>yeah
13:37<Battousai>that includes keepalives right?
13:37<JonR>i had 64.
13:37<JonR>i think
13:38<fuzzie>i have it set to 20, here, and it's more than fine, but it depends exactly what you're running in them
13:38<Battousai>10 would be alright for slow sites
13:38<Battousai>20 is probably good for most stuff
13:38[~]JonR has been looking at his apache traffic...
13:39<JonR>getting a hella lotta hits from msnbot.
13:39<fuzzie>haha, another one.
13:39[~]JonR puts them in IPtables.
13:39<JonR>fuck msn
13:39<fuzzie>they'll just switch IPs
13:39<fuzzie>to another subnet
13:39<fuzzie>it's great
13:39<fuzzie>i have a robots.txt solely to block msnbot
13:40<@caker># cat ../htdocs/robots.txt
13:40<@caker>User-agent: msnbot
13:40<@caker>Disallow: /forums/
13:40|-|tompuppy [~tom@proxy.tom.kao.name] has quit [Quit: Leaving]
13:40<@caker>their bot ends up there for days
13:40<JonR>heh... problem is, I got a lot of virtual servers.
13:40<@caker>very annoying
13:40<Battousai>i know my site only sees more than 20 or so concurrent clients for about 6 or 7 days a month
13:40<JonR>can i easily block msnbot from the whole thing, or do I need a robots.txt for each?
13:40<dddd44>excuse me ..if i connect tor network by
13:40<dddd44>sock5.. do the receiver know i connect by socks5 protocol or direct connection???anyone
13:41<fuzzie>you need one for each. maybe a mod_rewrite rule or .. aliases?
13:41<@caker>JonR: Disallow: /
13:41<JonR>caker: but how would that work for virtual servers? I'd need a robots.txt in every server.
13:41<@caker>JonR: ahh, yeah each one
13:41<JonR>yeah, mod_rewrite sounds like the key
13:44<JonR>hm.
13:44<JonR>i think they just come from one subnet...
13:45<iggy>hah, I've been fighting with msnbot all day
13:45<iggy>ended up having to iptables -j DROP all the IP's I could find
13:46<JonR>hm, iptables -A INPUT -s 65.54.188.0/24 -j DROP seems to have done the trick
13:47<iggy>207.46.98.52 is another address it uses
13:47<JonR>oh?
13:48<fuzzie>i watched it switch between four subnets before i gave up and just used robots.txt on all my domains
13:48<JonR>meh
13:48<@caker>64.4.8.94 - - [27/Aug/2005:14:47:59 -0400] "GET /wiki/index.php/Linode HTTP/1.0" 200 14163 "-" "msnbot/1.0 (+http://search.msn.com/msnbot.htm)"
13:49<iggy>oooh, I'll have to get that one too
13:49<JonR>my only concern is they seem to be using the same ip range as hotmail, and some of my users recieve mail from hotmail users.
13:50<iggy>so just block port 80
13:50<JonR>point.
13:50<iggy>--dport
13:51<JonR>I'm a little slow today. :/
13:52|-|TroubLed [~eX@194.54.56.139] has joined #linode
13:52<TroubLed>hoppaaa
13:52<TroubLed>selam melam
13:53|-|Martin [~Lynx@194.54.56.139] has joined #linode
13:54|-|Zangi [~KLBK2005@194.54.56.139] has joined #linode
13:54<TroubLed>hepppaaaaa
13:54<Martin>helloo
13:54<TroubLed>haddii bakeeemmm
13:55<SupaZubon>HAVE YOU THE BRAIN WORMS????
13:55<Martin>secter
13:55[~]Zangi SEVGIM SINIRSIZ AMA ANLIYABILENE, DOSTLUGUM SONSUZ AMA PAYLASABILENE, SEN HARIKASIN AMA BENIM OLDUGUN SURECE
13:55|-|Zangi [~KLBK2005@194.54.56.139] has quit [Quit: ]
13:55<fuzzie>fascainting
13:55<fuzzie>fascinating, even.
13:56<Martin>y\xFCr\xFC len amc\xFDk
13:56<Martin>\xFD fuck your grandmother
13:56<Martin>\xFD fuck your grandmother
13:56<Martin>\xFD fuck your grandmother
13:56<Martin>\xFD fuck your grandmother
13:56<Martin>\xFD fuck your grandmother
13:56<Martin>\xFD fuck your grandmother
13:56<Martin>\xFD fuck your grandmother
13:56<Martin>\xFD fuck your grandmother
13:56<Martin>\xFD fuck your grandmother
13:56<Martin>\xFD fuck your grandmother
13:56<Martin>\xFD fuck your grandmother
13:56<Martin>\xFD fuck your grandmother
13:56<Martin>\xFD fuck your grandmother
13:56<Martin>\xFD fuck your grandmother
13:57<SupaZubon>Martin: Your character set seems to be confused
13:57<SupaZubon>Martin: I'm running UTF-8, and I can't see those little characters you're putting in
13:57<SupaZubon>Martin: try unicode!
13:57<Martin>SupaZubon : anan\xFDn am\xFDn\xFDndan su getirem emi ?_
13:57<SupaZubon>Martin: forfiku vin, trolo!
13:57<Martin>kim lan buran\xFDn root u ?
13:57<iggy>SupaZubon: it looks like a funny y with an accent over it
13:58<SupaZubon>iggy: I see little diamond question marks
13:58<fuzzie>Me too.
13:58<Martin>iggy : \xFD fuck you ?
13:58<iggy>well, that's more interesting than what I see
13:58<iggy>Martin: no thanks, my hand looks better than you
13:58<Martin>secter amc\xFDk o zaman
13:58|-|Martin [~Lynx@194.54.56.139] has left #linode []
13:59<SupaZubon>aaaaaand SCENE
13:59<fuzzie>Aw, that was far more amusing than the normal "i'm 300mb into swap and my box is slow" activity in here.
13:59<[|^__^|]>it sure beat "HELLO HOW DO I INSTALL THE GENTOO ON MY DELL" questions
13:59<taupehat>http://www.funfreepages.com/flash/stfu_noob.php
13:59<taupehat>=]
13:59<[|^__^|]>flash, huh
14:00<taupehat>yeah
14:00<[|^__^|]>thanks, however, comma, no thanks
14:00<taupehat>hah
14:00<taupehat>with a nick like that
14:00<taupehat>I'll guess you prefer .wmv
14:00<taupehat>=P
14:00<[|^__^|]>the last thing I need is another puzzle piece
14:00<[|^__^|]>yeah well, xine can play .wmv
14:01[~]JonR has only ever been 128MB into his swap!
14:01<TroubLed>la bi susun a.q fuck you
14:01<taupehat>hah
14:01|-|kas_3 [debian-tor@blackhat4life.org] has joined #linode
14:01<taupehat>see if you can gete caker to recount the time I tried to run open-xchange on my 64
14:02[~]taupehat got put in no-tickets jail for that one
14:02<fuzzie>i have yet to dip into swap on my 128, it's eerie
14:02<Battousai>taupehat: somehow that's not surprising to me
14:03<taupehat>haha
14:03<taupehat>yeah
14:03<taupehat>since open-xchange is this giant tomcat monster
14:03[~]Battousai lays openxchange out
14:04<taupehat>with a postgresql backend
14:04[~]taupehat was being clueless
14:04<Battousai>yeah
14:04<Battousai>you really need a bit more than 64mb ram for that
14:04<Battousai>something like a penguin computing cluster would do, maybe
14:04<iggy>maybe
14:04<[|^__^|]>haha
14:04<[|^__^|]>Scyld!
14:05[~][|^__^|] was employee number 2 at Penguin
14:05<[|^__^|]>and CTO #1!
14:05<[|^__^|]>hahaha
14:05<[|^__^|]>I actually just had dinner with Donald Becker last night
14:05<[|^__^|]>he's the current CTO!
14:05<[|^__^|]>we're going to start a band: CTOs of Penguin
14:05<Battousai>please make him give my school another one
14:05<taupehat>heh
14:05<[|^__^|]>we'll be a laptop band, so we'll need dancing girls so the crowd doesn't get bored
14:05<Battousai>i'm sick of the realtime students getting all the funtime on it
14:05<[|^__^|]>haha
14:06[~]taupehat has not-so-fond memories of installing the Tulip driver, and much fonder memories of trying again a year later when it was greatly improved.
14:06|-|TroubLed [~eX@194.54.56.139] has quit [Quit: ]
14:06<Battousai>taupehat: tulip?
14:06<SupaZubon>well, it didn't help when Netgear switched from the DEC chips to some SMC no-name
14:06<Battousai>that's in the kernel...
14:06<SupaZubon>that happened when i was at Penguin still
14:07<taupehat>Battousai: in the olden days when I started using linux it wasn't
14:07<taupehat>and it was a bit of a bear to configure
14:09|-|dddd44 [~dhb55@dip-107-127.centaurus.dsl.breisnet.com] has quit [Ping timeout: 480 seconds]
14:16|-|kas_3 [debian-tor@blackhat4life.org] has quit [Ping timeout: 480 seconds]
14:38|-|funkycow [~Ap0ll0@loves.girls.who.have.bigtits.and.bigbutts.ca] has joined #linode
14:44|-|cow [~Ap0ll0@loves.girls.who.have.bigtits.and.bigbutts.ca] has quit [Ping timeout: 480 seconds]
15:01|-|adamg [~misthos@zeus.misthos.com] has quit [Quit: ]
15:07|-|adamg [~misthos@zeus.misthos.com] has joined #linode
15:14[~]caker wonders what the *@194.54.56.139 guy was all about
15:24[~]rillian hopes MaxClients 30 will keep our 64 out of swap for a while
15:25<gpd>rillian: install monit... get it to shut down apache when loadavg goes up...
15:26<rillian>I guess that would have gotten us control back sooner
15:26<rillian>but isn't exactly ideal either
15:27<rillian>in theory adding robots.txt would be enough, but the problem with badly-written spiders is you never know how much they're respect the exclusion
15:27<gpd>restarting apache and mysql normally frees up a lot of memory
15:28<fuzzie>30 should be okay, if not ideal, if you're not serving anything complex
15:28<rillian>s/complex/popular/ :)
15:28<fuzzie>not realy
15:28<rillian>no, the problem was the 100 python processes each using 2% of ram
15:28<@linbot>... but the problem is that necessary with urgency :), rillian ...
15:28<fuzzie>i mean, the 'popular' will be handled by setting maxclients
15:29<fuzzie>but if you're running scripts which take up a lot of RAM, that will kill you
15:29<gpd>python as cgi?
15:29<rillian>yes
15:29<gpd>or gnu mailman?
15:29<rillian>trac
15:30<fuzzie>you've considered mod_python?
15:30<rillian>a spider hit it with no query delay
15:30<gpd>oh, I was looking at trac yesterday...
15:30<rillian>it's a nice system
15:30<rillian>or will be when it more mature
15:30<fuzzie>i use .. mod_throttle? .. to limit requests
15:30<rillian>but it's not particularly light weight
15:30<rillian>fuzzie: does that work on a location basis?
15:31<fuzzie>yup
15:31<rillian>would be nice to throttle trac while leaving static pages unlimited
15:31<fuzzie><Location /blah> SetHandler throttle-blah </Location>
15:31<gpd>it uses sqlite... do you have mysql too?
15:31<fuzzie>oh, no, ThrottlePolicy. but, anyway, it works
15:31<rillian>ok, we'll do that when we get some non-trac content :)
15:32<fuzzie>i use it to throttle concurrent connections
15:32<fuzzie>mostly to stop idiot download managers which open 20+ connections from using up all my apache children
15:33[~]rillian isn't finding mod_throttle for apache2
15:33|-|Guest43 [~misthos@zeus.misthos.com] has joined #linode
15:34|-|adamg [~misthos@zeus.misthos.com] has quit [Read error: Connection reset by peer]
15:34<fuzzie>oh, heh. there isn't one. didn't think of that. but there's a mod_bandwidth port over at http://www.ivn.cl/apache/ it seems
15:35<gpd>which apache2? mpm/worker/prefork?
15:39<rillian>yes, looks like that one would work
15:39<rillian>gpd: prefork
15:39<rillian>we were using php at some point, although not currently
15:40<rillian>fuzzie: thanks for the pointers
15:50<rillian>gpd: the spider was actually hitting the svn browser, so I'm not sure a "real" database would have helped
15:50<rillian>we're using postgres for most other things
15:51<gpd>what other services do you run? can you trim any?
15:51<fuzzie>heh. is the svn browser any good?
15:51<fuzzie>i'm having viewcvs, and viewsvn is .. insanely heavy
15:51<fuzzie>s/having/using/
15:52<fuzzie>and viewcvs has interesting issues
15:52<rillian>it's not as good for tracing history as viewcvs
15:53<rillian>http://onlinegamegroup.com/projects/libstrophe/browser/trunk/libstrophe/ for an example
15:53<gpd>http://trac.adiumx.com/browser << another example... (how I found it)
15:54<fuzzie>it looks rather nice..
15:54<rillian>yep. trac's major feature is "pretty" :)
15:55<rillian>that's a newer version than we're running
15:59<rillian>gpd: there's not really anything we can trim
15:59<rillian>it was an allocation tuning problem more than anything
16:13<@linbot>New news from forums: tomcat 5.5 apache 1.3 talking properly on debian in Linux, Apache, Mysql and PHP (LAMP) Forum <http://www.linode.com/forums/viewtopic.php?t=1841>
16:25|-|Guest43 changed nick to adamg
16:28|-|ronpoz [~ronpoz@ool-45700306.dyn.optonline.net] has quit [Quit: Trillian (http://www.ceruleanstudios.com]
16:35|-|JonR [~jon@mn-10k-dhcp2-190.dsl.hickorytech.net] has left #linode [Leaving]
16:41<@caker>http://www.pbs.org/cringely/pulpit/pulpit20050825.html
16:41<@caker>^-- good read
16:41<@caker>What could Apple do to take down Windows, with or without the help of Intel?
16:41<@caker>Every one of those iPods is a bootable drive. What if Apple introduces OS 10.5, its next super-duper operating system release, and at the same time starts loading FOR FREE the current operating system version -- OS 10.4 -- on every new iPod in a version that runs on generic Intel boxes? What if they also make 10.4 a free download through the iTunes Music Store?
16:41<@caker>Smart.
16:42<fuzzie>it'd mean a massive audience for the first person to hack 10.5 onto generic hardware. :P
16:42<gpd>where are they making money out of all this?
16:44<@mikegrb>the first sip is always free
16:44<gpd>a tick box on dell.com: Windows [ ] or OSX [X] :)
16:45<@linbot>I learned it by watching YOU, Dad.
16:45<warewolf>wouldn't happen.
16:45<warewolf>'cause microsoft won't let dell sell any os other than windows.
16:46<warewolf>also on dell: have you noticed how many times dell has said to the public "We're looking into putting AMD chips in our systems"
16:46<warewolf>and then NEVER actually doing it?
16:46<@caker>Dink. dong ding dong dink
16:47<warewolf>it's a tactic to get Intel to give 'em cheaper prices
16:48<Battousai>no it isnt
16:48<Battousai>cheaper prices are a tactic by intel to get dell not to switch
16:48<warewolf>same difference
16:52<warewolf>I DON'T SEE YOU ON GOOGLETALK
16:52<warewolf>er
16:52<warewolf>misfire
16:59|-|honeyzzz changed nick to honey
17:15<gpd>bored bored bored...
17:19<honey>aw
17:24|-|abock [~aaron@user-112038b.dsl.mindspring.com] has joined #linode
17:24|-|abock [~aaron@user-112038b.dsl.mindspring.com] has left #linode []
17:32|-|jekil [~alessandr@151.82.0.153] has quit [Quit: This computer has gone to sleep]
17:36<rillian>gpd: if you're bored, write a gtk (or macosx) app that talks to mailman's moderation queue over the net
17:39<honey>oh god, what a timely comment. that is *exactly* what i need, after trying mailman out yesterday, and missing it
17:40<honey>i need to be able to unmoderate users via something other than the web. else i'm back to old old majordomo
17:42<rillian>I find the web interface fine for small lists
17:43<rillian>but it's a usability nightmare when you have more than a handfull of lists with more than a handfull of queued messages
17:43<rillian>but you're right that mailman 2.0 worked much better
17:44<gpd>can't you moderate users from the commandline?
17:44<rillian>you probably can from a server shell
17:44<rillian>I've never tried
17:46<honey>i looked for a way
17:46<honey>im sure i could if i knew python or looked into mailman's withlist
17:47<honey>but i have to weigh up if the curve's worth it - some other stuff missing i need too
17:47<rillian>failing that, you can load the mailman machinery into python and filter the queue manually
17:47<honey>rillian, that's where i'd have a learning curve - and the point of swapping to mailman would be to lesson the hack factor for me
17:47<honey>i already have a hacked majordomo, but can do some special stuff with it
17:47<rillian>as I said, it would be really nice to have a remote app that just notified you of new messages and let you easily examine and accept/reject/drop them
17:48<gpd>the new messages web interface isn't that bad...
17:48<gpd>I find the user tables the worst...
17:48<honey>like, i interpret specific majordomo bounce messages by procmail->script according to the bounce type and respond. all mailman bounces seem to just say "bounce - go to the web"
17:48<honey>rillian, yes, like you can in ancient majordomo :)
17:49<honey>gpd, i'm only 2 hour's old with mailman. by "new" do you mean 2.1.6? i have 2.1.5 just now
17:49<gpd>neither... i just meant the moderation pages... but come to think of it those pages are a pain...
17:50<honey>i'd really like to swap to mailman. but i need to be able to administer lists by email completely, or at the very worst by shell commands, including unmoderation
17:51<rillian>does seem odd that mailman doesn't support moderation over email
17:51<honey>i guess i need an unmoderate.py. but i also need the bounce messages to contain bounce reason so i can process them with procmail. most of all, i need to set taboos on body content
17:51<rillian>that's probably much easier to add than a whole remote queue manipulation gadget
17:51<honey>yes rillian, it does
17:51<rillian>it does support moderation over email?
17:52<honey>i presume i'm not just missing something? i can't see any way to set taboos on body content
17:52<honey>can't see anyway, rillian
17:52<honey>after playing for 2 hours (including first installation)
17:52<honey>i could still be missing something, but others are indicating these features don't exist too
17:52<gpd>just spotted: listadmin - command line mailman moderator queue manipulation
17:53<honey>gpd, can you show me where?
17:53<honey>is it an add-on?
17:53<gpd>apt-get install listadmin for me
17:53<gpd>http://freshmeat.net/projects/listadmin/
17:53<honey>oo. if it's a debian package, it's kind of officialish
17:53[~]honey looks
17:54|-|tierra [~tierra@c-24-10-173-249.hsd1.ut.comcast.net] has joined #linode
17:54<honey>looks interesting. it does moderation of users too?
17:55<gpd>no clue... just saw it in apt-cache search
17:55<honey>thanks! worth looking at at least
17:56<honey>looks from the sample as if it's to do with moderation mail, not affecting moderated user lists...
17:56<gpd>the freshmeat pages says subscription requests are covered
17:56<honey>ah
17:56<gpd>listadmin is a command-line alternative to Mailman's Web interface for administering mailing lists. It allows you to inspect messages or subscription requests, approve or discard them manually, or discard messages automatically when certain conditions are true. listadmin can also be configured to run out of cron to do routine cleaning.
17:57<honey>right - but administrer the requests to subscribe. i need to unmoderate people automatically after a time
17:58<gpd>not sure what that even means? you mean remove their 'requires moderation' bit if they don't post for x weeks?
17:58<honey>close - stop them being moderated (mail blocked, effectively) two weeks after joining
17:59<honey>i do it currrently via majordomo (with ugly hacks) and it really helps list sanity
17:59<gpd>how many new users per day?
17:59<honey>the idea is new users can't post for two weeks, they have to just read
17:59<honey>not that many, but it needs to scale and be handsfree
17:59<gpd>you must have high tech users
18:00<honey>it is currently and has worked for 5 years
18:00<gpd>my users can barely subscribe!
18:00<honey>hmm? it doesn't have anything to do with the users :)
18:00<gpd>every day i get: PLEASE UNSUBSCRIBE ME FROM THIS LIST
18:00<honey>they're just told they can'[t mail for two weeks
18:00<gpd>despite a message at the bottom saying how to do it...
18:00<honey>then they get a mail saying "ok! you have a voice now"
18:00<fuzzie>gpd: you just have stupid users, i expect.
18:00<gpd>the irony is that all my users have PhDs... :)
18:01<gpd>evey single one of them!
18:01<fuzzie>thus proving meaninglessness. :P
18:01<honey>it means they have to read silently for two weeks, thats all. they can't jump in and try and sell anything, or not understand list ethos. works a treat and no-one complains
18:01<honey>gpd, they're often the worst at reading :)
18:01<gpd>true
18:02<fuzzie>if it's an automatic list, i guess the trick is to subscribe about 50 addresses at once, wait two weeks, and then slowly use addresses from the pool to spam. :-)
18:02<honey>not tried majordomo2 - looks very devel still, after 5 years
18:02<honey>i'd really like a bounce handler
18:03<gpd>aren't forums replacing mailing lists?
18:03<honey>fuzzie :) they subscribe themselves at random intervals :) the software also remembers those who were given voices, and if they leave and rejoin, they get on immediately
18:03<thelsdj>i still prefer mailing lists over forums
18:03<@mikegrb>honey: the pobox.com buy has a mailing list he wrote
18:03<honey>it's all based on appalling shell script wot i wrote
18:03<@mikegrb>he also wrote a perl module that handles bounces that it uses
18:03<thelsdj>i hate having to go somewhere to get updates, would rather have it come to me
18:03<fuzzie>i tend to just beg people for newsgroups
18:03<@mikegrb>I'm using his perl module currently
18:04<honey>mikegrb, i have a feeling i tried that aeons ago... but can't recall details
18:04<fuzzie>but mailing lists are an acceptable alternative
18:04<gpd>thelsdj: the forum emails you or use RSS...
18:04<honey>it's a standalone bounce handler?
18:04<honey>oh goodness, forums aren't replacing mailing lists :)
18:04<thelsdj>bleh, having to click on something in an email is too much work, and most forums don't have rss of every post
18:05<@mikegrb>oh
18:05<@mikegrb>looks like it is a service
18:05<thelsdj>gpd: grooving out to radio soulwax
18:05<honey>plaintext-only mailing lists are wildly more useful. search your own mailbox archives forever, etc etc
18:05<@mikegrb>http://search.cpan.org/~freeside/Mail-DeliveryStatus-BounceParser-1.4/BounceParser.pm <-- perl module
18:05<@mikegrb>pretty easy to use
18:05<honey>mikegrb, hm thanks. i'll look
18:05<honey>mailman and mj2 both claim to handle bounces well - mj1, not
18:06<@mikegrb>using it for processing bounced outbound linode.com messages
18:06<@mikegrb>though manually right now
18:07<honey>ok so it's standalone
18:07<gpd>thelsdj: Erick Morillo
18:07<honey>helpful, thanks
18:08<gpd>thelsdj: soulwax em 05/01/02 ?
18:08<honey>can i ask some rather basic questions about the future evils of Sender ID and Domainkeys?
18:08<@mikegrb>of course
18:08<honey>are my hotmail users likely to be blocked in November?
18:09<fuzzie>isn't SPF good enough?
18:09<honey>and if i don't want them to be, and accept being bullied by MS, is it just a matter of adding some TXT records
18:09<@mikegrb>honey: you mean messages to hotmail users or messages from?
18:09<honey>well, MS are implenting their own version, Sender ID in November
18:09<honey>to, mikegrb
18:09<honey>from a mailing list
18:10<@mikegrb>there was an opposition to the draft filed with the IETF or who ever
18:10<honey>it's also a mess, because it looks open to spoofing of IPs to me, which spammers will do
18:10<@mikegrb>just this last week
18:10<honey>whereas at least Yahoo's Domainkeys involves key exchange
18:11<honey>oh really mikegrb?
18:11<fuzzie>how would you spoof an IP in an SMTP exchange?
18:11<fuzzie>it's not as if it's UDP
18:11<@mikegrb>honey: yeah, check NANOG list archives
18:11<honey>fuzzie, my naive understanding on first reading is that all Sender ID requires is that tokens in your mail match TXT records for your domain
18:11<honey>mikegrb, thanks, will.
18:11<fuzzie>no, it requires the sending server match the server listed in the TXT record for your domain.
18:11<honey>will an opposition with IETF block it for long?
18:12<thelsdj>gpd: no, the original 10 2 many djs
18:12<@mikegrb>http://search.cpan.org/~freeside/Mail-DeliveryStatus-BounceParser-1.4/BounceParser.pml
18:12<@mikegrb>er
18:12<@mikegrb>http://search.cpan.org/~freeside/Mail-DeliveryStatus-BounceParser-1.4/BounceParser.pml
18:12<honey>fuzzie, right, but can't a spammer also generate the same records?
18:12<@mikegrb>gah
18:12<@mikegrb>http://www.merit.edu/mail.archives/nanog/msg10888.html
18:13<fuzzie>honey: sure, all it does is make sure that the server an email is sent from is a valid one for the domain.
18:13<fuzzie>so a spammer could just register domains and spam from those.
18:13<honey>fuzzie, maybe i'm not understanding it right. i'm going on this article, largely: http://www.esecurityplanet.com/views/article.php/3398431
18:13<@mikegrb>honey: there was a study showing that SPF/sender id had a much higer acceptance rate among spammers
18:13<fuzzie>but you can just block the domains.
18:13<honey>mikegrb, yes, i've seen that server. they love it
18:13<@mikegrb>as many places foolishly bypassed other checks
18:14<fuzzie>it's not a very good solution to spam, but it fixes faked return-path addresses.
18:14<@mikegrb>right
18:14<honey>fuzzie, the article claims that " IP address spoofing renders Sender ID moot"
18:14<fuzzie>well, as far as i'm aware that's bullshit
18:14<@mikegrb>tcp ip spoofing is not trivial
18:14<honey>mikegrb, the logic would dictate that therefore you score UP anyone as spam who implements Sender ID :)
18:14<fuzzie>and it doesn't particularly look like the most professional site
18:14<@mikegrb>you pretty much need to be in the route of the target
18:14<honey>uh, SPF
18:15<fuzzie>a lot of major mail providers already provide SPF
18:15<fuzzie>so that'd not be useful, unfortunately
18:16<honey>hmm ok. so what will yahoo do, implement Sender ID to accept hotmail.com mail AND implement Domainkeys? and gmail implement both?
18:16<fuzzie>you don't have to implement Sender ID to recieve mail from hotmail users
18:16<fuzzie>just to send it to them.
18:16<honey>fuzzie, yeah, i mean send
18:16<honey>to hotmail users from their mailing lists
18:17<@mikegrb>the problem is, sender id and spf don't handle mailing lists gracefully
18:17<honey>i'm trying to work out how tis will pan out, and what i'll need to do within the next 6 months, if i wish to keep mailing hotmail, msn, yahoo users directly from mailing lists
18:17<honey>and yes, the headers get munged
18:18<honey>also a problem for DomainKeys
18:18<@mikegrb>well, I think msn will likely be the only problem
18:18<honey>if yahoo push that
18:18<@mikegrb>yahoo isn't stupid enough to require it
18:18<@mikegrb>ms just thinkgs they can force everyone to do it
18:18<honey>you think? but will the world give in and implement Sender ID?
18:18<@mikegrb>no
18:18<honey>(surely hotmail as well as MSN users)
18:18<@mikegrb>they will turn it on
18:19<honey>i've already lost most of my AOL users
18:19<@mikegrb>people will complain they aren't getting their email
18:19<@mikegrb>they will turn it off
18:19<honey>through their idiot user reporting
18:19<honey>you think?
18:19<@mikegrb>yeah
18:19<honey>it's a nice thought
18:20<honey>they're scoring down non-Sender ID already i think - they say November they'll block
18:20<@mikegrb>right
18:21<honey>if i *wanted* to implement Sender ID on mailing lists, is it just a matter of setting up TXT records? or will i have to play with mailing list headers?
18:21<honey>(not that i'm inclined to - i already told my AOL and hotmail users to leave)
18:21<@mikegrb>headers too
18:22<@mikegrb>as I'm pretty sure spf/senderid both look at the header from, not just the envelope sender
18:22<honey>add some, mikegrb? or something more complex?
18:22<@mikegrb>which is the big problem for mailing lists
18:22<fuzzie>SPF does not require looking at the header from.
18:22<honey>i'm actually confused what the problem is
18:22<fuzzie>Their recommendation is to use the envelope sender.
18:22<honey>i thought that only DomainKeys was a problem for mailing lists
18:22<fuzzie>But this doesn't mean everyone's going to be competent, sadly.
18:23<honey>sorry to be dim. Reply-path: ?
18:23<fuzzie>honey: the mail address sent at the start of the SMTP command.
18:23<honey>ah, so during the HELO?
18:24<@mikegrb>the MAIL FROM
18:24<honey>ah
18:24<fuzzie>no, during .. yes, that
18:24<fuzzie>http://spf.pobox.com/faq.html#whichfield
18:24<honey>thanks
18:24<honey>i'll read, cool
18:24<fuzzie>which basically says "if you want to protect From, use DomainKeys"
18:24<@mikegrb>heck, exchange doesn't even send bounces to the envelope sender
18:24<fuzzie>nice.
18:24<@mikegrb>yeah
18:25<@mikegrb>appearently the issue is it accepts the message for delivery, realizes it can't deliver it, then sends the bounce
18:25<honey>right, but if my concern isn't to use these to protect myself from incoming spam (i can do that fine without), but sending TO people who implement it - i don't get to choose schemes
18:25<@mikegrb>but it has already discarded the envelope sender
18:26<thelsdj>mikegrb: you mean when i saw exchange do that the other day it wasn't misconfigured? its actually just the way it works?
18:26<@mikegrb>yeah
18:26<thelsdj>i thought it was insane that it accepted the mail then later bounced it
18:26<@mikegrb>that's why you've got to add an Errors-To header
18:26<@mikegrb>for exchange
18:26<fuzzie>honey: you should not have problems with adding a TXT field for SPF.
18:27<honey>fuzzie, and no issues with sending from mailing lists?
18:27<fuzzie>i do not know if Microsoft will accept plain SPF as their insane system, though.
18:27<honey>yeah
18:27<fuzzie>but considering no-one else gives a damn about their proprietary system
18:27<@mikegrb>fuzzie: that's the big problem with senderid
18:27<fuzzie>they'd be suicidal to block everything without it.
18:27<honey>well, if everyone ignores it, it'll sink yes
18:27<fuzzie>whereas SPF is more reasonable to require.
18:27<@mikegrb>if you have spf records, you don't know if people are going to use them as spf records or sender id
18:28<honey>i'm worried the big companies won't. i see bigger corporations falling over themselves to ban outgoing mail based on AOL's insane abuse reports, at a lower threshold than AOL
18:28<fuzzie>it's only the additional stuff in sender id that is a problem, though?
18:28<@mikegrb>which is what is likely to hold it up from becomming an experimental rfc
18:28<honey>because they're scared of AOL
18:28<@mikegrb>well, sender id uses similiar records but interprets them differently
18:29<@mikegrb>and when you publish a record, you don't know which rules will be used
18:29<fuzzie>right, so they're even more crazy
18:29<@mikegrb>exactly
18:30<honey>does anyone have a clue what Gmail will do. i'm currently recommending stranded AOL users to go there
18:30<@mikegrb>they will probably be doing the most sane thing of the three
18:30<@mikegrb>er four
18:31<honey>TEOS?
18:32<@mikegrb>!acronymn TEOS
18:32<@mikegrb>!acro TEOS
18:32<@linbot>mikegrb: TEOS could be Tetra Ethyl Oxysilane, or Tetraethylorthosilicate
18:32<@mikegrb>yes
18:32<[|^__^|]>The End Of Sex
18:32<[|^__^|]>Forever.
18:33<[|^__^|]>The End Of Sexychat
18:33<[|^__^|]>No more world-wide intersexychat!
18:33<thelsdj>OH NOES!
18:34<honey>"Trusted Email Open Standard" - unsure if it ever took off
18:34<[|^__^|]>Doesn't seem to have
18:35<honey>ok, so um, what was mikegrb's fourth :)
18:35<[|^__^|]>the fourth of mikegrb is a sacred holy day
18:36<@mikegrb>honey: aol, yahoo, msn, gmail
18:36<[|^__^|]>The month of mikegrb was created in the fourth century BS (Before Sexychat) by cakerius the third.
18:36<@mikegrb>honey: gmail will likely be the most sane
18:36<honey>oh, you meant webmail providers. thought you meant schemes
18:36<@mikegrb>right
18:36<honey>indeed
18:36<honey>i'm just wondering *how* sane - if they'll use any
18:37<honey>they're spam protection works pretty well for me
18:37<honey>(their)
18:37<@mikegrb>though if I was making TFT stuffs, I'd definitly use Tetraethylorthosilicate
18:37<honey>:)
18:38<honey>*if* one of more of the schemes manages to get a foothold, smaller providers will need to implement them, or they'll get a layman's reputation "oh, mailer x doesn't work with hotmail"
18:38<honey>which i guess is MS's hope
18:40<@mikegrb>in and ideal world it would be "don't use hotmail, it doesn't work with anything else"
18:40<@mikegrb>but far to many idiots
18:40<honey>i've already told my listees that. plus AOL
18:40<honey>or warned them it'll break
18:40<honey>i first recommended the AOL users used hotmail. til i read this
18:41<honey>"frying pan" and "fire" springs to mind
18:48|-|maher [~maher@charter-241-206.caltech.edu] has joined #linode
18:50<honey>gpd, faintly worrying comment on manpage of listadmin: "Mailman 2.1 is not supported." - does it mean 2.1 or 2.1.x :)
18:56<gpd>you might ask people on /j -freenode mailman, it is not very active but they get back to you eventually
18:59<honey>ah, there's an idea
18:59<honey>thanks
19:00<honey>yeah, it's kind of small :)
19:06<@caker>wtf
19:06<@caker>http://www.gnu.org/music/free-software-song.au
19:07<SupaZubon>um
19:07<SupaZubon>welcome to 1997
19:07<@mikegrb>mmm unix audio
19:07<@caker>yeah
19:07<honey>oh not THAT song? oh god!
19:07<SupaZubon>that used to be up on jwz.org with the name why-cooperation-with-rms-is-impossible.au
19:07<honey>the RMS song?
19:07<honey>it's terrifying
19:07<SupaZubon>it's half of a modal folk song
19:09<@caker>What's all the drama behind this jwz guy?
19:09<honey>i once had a mild fight with him on freenode
19:09<honey>which was fun :)
19:09<@mikegrb>ha ha jwz
19:09<@caker>honey: regarding what?
19:09<honey>he's just a bit opinionated i guess
19:09<@mikegrb>I'm banned from his blog
19:09<honey>caker, can't remember - something trivial about RHEL
19:09<honey>it was more an attitude thing, which i wasn't having!
19:10<@mikegrb>on account of suggesting everyone was stupid including him because the question was stupid
19:10<@mikegrb>don't remember the topic
19:10<honey>i used to quite like some of his rants
19:10<honey>but not so much now :)
19:10<@mikegrb>effectively how to turn a computer on -> push the power button type of simplicity
19:11<@mikegrb>someone replied with "do you know who jwz is!?!? if he can't figure it out it is not simple and none of us are likely to"
19:11<honey>it made me appreciate RMS a bit more, in spite of the fact i'd be scared spending time with him on my own
19:11<gpd>http://en.wikipedia.org/wiki/Jwz heh
19:11<honey>ha!
19:12[~]gpd begins /wiki/gpd article
19:12<honey>jwz just has a lot of deep web/linux history to his credit, but he does seem to be too attitudinal
19:12<honey>if that's a word
19:14<@mikegrb>"Some people, when confronted with a problem, think "I know, I'll use regular expressions." Now they have two problems."
19:14<honey>jwz: " Mailman is very popular mailing list manager software. It's a pain in the ass to use, and has a braindead security model. Here's a rant about it. "
19:14<honey>http://www.jwz.org/doc/
19:14<@mikegrb>"Linux is only free if your time has no value."
19:14<honey>interesting, let's give it a read
19:14<@caker>Hah
19:15<gpd>http://www.jwz.org/gruntle/ <-- grumbles about most things...
19:15<honey>(i quite like that last quote actually, mikegrb)
19:15<honey>first point against mailman is incorrect. let's move on
19:16<honey>(and i only installed it yesterday!)
19:18<honey>his historical docs like this had some interest: http://www.jwz.org/gruntle/nomo.html
19:24|-|maher [~maher@charter-241-206.caltech.edu] has quit [Remote host closed the connection]
19:36|-|dc0e [~dc0e@pcp0010756487pcs.howard01.md.comcast.net] has joined #linode
19:47<gpd>are there laws in the US regulating noise levels in an office ?
19:53|-|Pique [~seck@216.86.157.41] has joined #linode
20:01<honey>gpd, thanks for the obvious suggestion to visit #mailman. getting good help now. 3-line unmoderate.py script :)
20:02<gpd> nice :)
20:13<@mikegrb>gpd: yes
20:13<@mikegrb>gpd: OSHA regulates that
20:14<@mikegrb>http://www.osha.gov/pls/oshaweb/owadisp.show_document?p_table=INTERPRETATIONS&p_id=19093
20:14<@mikegrb>gpd: 90 decibels
20:15<gpd>wow that is pretty loud!
20:16<gpd>and I am moaning about a noisy refrigerator... :(
20:16<gpd>whilst listening to Paul Oakenfold at about 90db :)
20:16<@mikegrb>http://www.dir.ca.gov/title8/5096.html
20:17<@mikegrb>we had excess of 120db at work
20:17<gpd>measuring noise level is not trivial... ?
20:17<@mikegrb>and now I can't hear worth a damn
20:17<gpd>what caused that? computer stuff?
20:17<@mikegrb>somewhat trivial
20:18<@mikegrb>a nuclear reactor
20:18<@mikegrb>the steam plant is noisy
20:18<gpd>why did you work there? for the danger?
20:18<@mikegrb>I'd probably be worse off had I not worn ear plugs
20:18<@mikegrb>mmm danger
20:18<gpd>mmm radiation
20:18<@mikegrb>quite low
20:19<gpd>I use 32P... which is fairly safe
20:19<@mikegrb>less then being outside
20:19|-|Pique [~seck@216.86.157.41] has quit [Quit: ]
20:19<gpd>until you start picking your nose in the middle of an experiment
20:19<@mikegrb>over a year my exposure was something like 2 or 3 milirem
20:19<gpd>our safety officer used to work in a nuclear sub
20:20<@mikegrb>er .2 or .3
20:20<gpd>MILI!!! wow... tha tis high...
20:20<gpd>oh
20:20<@mikegrb>limit is something like 100 milirem/yr
20:20<@mikegrb>local limit
20:20<@mikegrb>still well within federal regulations
20:21<gpd>we also use 3H but that scares me... can't tell where it is...
20:21<@mikegrb>we put crabs in a blender
20:21<@mikegrb>from the river
20:21<@mikegrb>blend them up nice and smooth
20:21<@mikegrb>spread the paste into a tray
20:21<@mikegrb>measure with radiac
20:21<@mikegrb>check for enviromental exposure
20:21<gpd>waring blender I hope...
20:22<gpd>do you relate to Homer Simpson?
20:22<@mikegrb>yes
20:22<@mikegrb>matt groen worked in nuclear power
20:22<@mikegrb>in the navy
20:22<gpd>heh
20:22<gpd>I suspect there is a
20:22<@mikegrb>mr burns was modled after adm rickover
20:23<@mikegrb>whom started the naval nuclear power program
20:23<@mikegrb>best part was having water cleaner then the tap water and not being able to discharge it into the river due to enviromental regulations
20:23<gpd>I have now switched to using fluorophores instead... much prettier... less hot...
20:24<gpd>yup... we are not allowed to dispose of NaCl solutions...
20:24<gpd>they are 'hazardous waste'
20:24<@mikegrb>so instead of dumping clean non radioactive water, hundreds of thousands of dollars spent to to put it in a tanker truck marked radioactive material and send it elsewhere to make it disapear
20:25<@mikegrb>once every other month or so
20:25<@mikegrb>don't see the point in the billion dollar waste water purification and treatment system if they are just going to pay the same cost to get rid of it had they done nothing
20:25|-|tierra [~tierra@c-24-10-173-249.hsd1.ut.comcast.net] has quit [Quit: Don't hate yourself in the morning -- sleep til noon.]
20:25[~]mikegrb shrugs
20:26[~]gpd awaits next nuclear reactor meltdown... :)
20:26<gpd>some of the British ones are getting very old
20:27|-|tompuppy [~tom@proxy.tom.kao.name] has joined #linode
20:27<@mikegrb>nah
20:27<@mikegrb>the problem was faulty design
20:27<gpd>this one has a visitor center... pretty strange concept...
20:27<gpd>http://www.visitcumbria.com/wc/svc.htm
20:27<@mikegrb>you need a moderator to continue the reaction, then a coolant to cary the heat away to do work
20:28<@mikegrb>they used seperate moderator and coolant
20:28<@mikegrb>coolant went away, moderator stuck around
20:28<@mikegrb>water is about the best moderator, and makes a good coolant too
20:28<@mikegrb>bonus, if the coolant skips town, there is no moderator and the reaction stops
20:29<@mikegrb>though there was the army reactor in iowa
20:29<@mikegrb>they were calibrating the control rods
20:29<@mikegrb>were supposed to pull one rod out 2 inches, it was written as 2" person reading directions thought that meant 2 ft
20:29<@mikegrb>pulled rod out 2 ft, boom
20:29<@mikegrb>death followed
20:30<gpd>hah... sounds like spinal tap...
20:30<@mikegrb>so now things aren't done like that
20:30<@mikegrb>rods are in groups
20:30<gpd>imperial measurements... in a nuclear plant!!! wtf?
20:30<@mikegrb>and one rod can't bring the reactor critical, let alone super critical to begin starting up
20:30<@mikegrb>it was the army
20:30<@mikegrb>needless to say, the army isn't alowed to play with nuclear power any more
20:31<gpd>they still have nuclear subs don't they?
20:31<@mikegrb>all of the subs are nuclear
20:31<gpd>s/?/./
20:31<@mikegrb>well, I believe we have one diesel electric for research
20:31<@mikegrb>but that's the navy, whom haven't had any reactor accidents
20:32<@mikegrb>it's amazing how inheritly safe nuclear reactors are (when properly designed)
20:32<@mikegrb>and how things "just work"
20:33<@mikegrb>you pull out more steam to generate more electricity, the coolant gets colder therefor more dense, therefor moderates better, therefor reactor power goes up
20:33<@mikegrb>automagically
20:33<@mikegrb>control rods don't even need to be moved to increase output
20:33<@mikegrb>they are used primarily just to turn it on and off
20:35<honey>night everyone, thanks for the help
20:35|-|honey changed nick to honeyzzz
20:43<gpd>sorry, had to deal with my HPLC purifcation stuff...
20:44<gpd>so how much *hot* waste does it produce per year? and where does it all go? back in the ground?
20:48<@mikegrb>no clue where it goes
20:48<@mikegrb>but a tanker truck every other month
20:49<@mikegrb>water has a very short half life
20:49<@mikegrb>like < a minute
20:49<gpd>water?
20:50|-|FireSlash [Liger@36.207-191-238.gbrdialin.gbronline.com] has joined #linode
20:50<gpd>The average nuclear power station produces 20-30 tonnes of spent fuel each year.
20:51<@mikegrb>google isn't helpful
20:51<gpd>http://en.wikipedia.org/wiki/Nuclear_power
20:51<@mikegrb>seem to think it was 12 seconds or such
20:51<@mikegrb>yeah, water is what as I was complaining about
20:52<@mikegrb>the water that was taken off by tanker truck was always 0 counts above background
20:52<@mikegrb>counts per minute
20:52<gpd>can't be the water... must be something in it...
20:52<@mikegrb>well, the issue is metal
20:52<@mikegrb>generally cobalt
20:52<@mikegrb>from pipes
20:52<@mikegrb>but that is all removed from the water
20:52<@mikegrb>on site
20:53<@mikegrb>this wasn't a commercial plant
20:53<@mikegrb>was navy
20:53<@mikegrb>solid waste was quite low
20:54<@mikegrb>mostly incidental waste
20:54<@mikegrb>clothing, test stuffs, etc
20:54<gpd>wow... 238U half life is 4.5 billion years... wtf?
20:55<@mikegrb>actual waste from the plant was how ever much every 10 years or so when they replace resin in primary coolant filter and fuel
20:55<gpd>so did you work on the computer side of it? or did you have a funky suit?
20:56<@mikegrb>two rolls, reactor operator, actually sit at the control panel and switch pumps, shim rods, etc
20:56<@mikegrb>and second was maintainance on the electronics and periodic inspections
20:57<gpd>less IRC in those days then!
20:57<@mikegrb>heh none
20:57<@mikegrb>12 hour shifts
20:57<@mikegrb>on site for a good 13 hours or so
20:57<@mikegrb>add 8 hours for sleep
20:57<@mikegrb>not much time left
20:57<gpd>I would hvae thought the opposite... short shifts to prevent tiredness causing accidents...
20:57<@mikegrb>hah
20:58<@mikegrb>well, the shifts were long
20:58<@mikegrb>actual shifts at the control panel were 4 hours
20:58<@mikegrb>but no bathroom breaks
20:58<@mikegrb>rest of the 12 hours spent on maintenance
20:58<gpd>crazy
20:59<@mikegrb>PM
20:59<@mikegrb>er
20:59<@mikegrb>PMS
20:59<@mikegrb>periodic maintenance schedule
20:59<gpd>I think linbot would disagree
20:59<@mikegrb>item x must be inspected and tested daily
20:59<gpd>!acronym pms
20:59<@mikegrb>item y, weekly
20:59<@linbot>gpd: pms could be Premenstrual Syndrome, or Martinique Socialist Party, or NAVSEA Program Manager, or PAC II Project Management System, or Pacific Masters Swimming, or Pack My Suitcase, or Package Management System (computer software), or Pacoima Middle School, or Paint Mixing System(s), or Palms Middle School, or Pamper Me Softly, Inc., or Pantone Matching System (color matching for (1 more message)
20:59<@mikegrb>item z, each startup
21:00<@caker>linbot: tell gpd more
21:00<gpd>!more
21:00<@linbot>gpd: publishing), or Pardon My Stupidity, or Parked Motorcycle Syndrome (winter blues), or Parras Middle School, or Particle Measurement System, or Pass My Shotgun, or Passe Musikalske Studiner, or Passport Management System, or Password Management System
21:00<@mikegrb>startups took days
21:00<@mikegrb> Parked Motorcycle Syndrome (winter blues) <-- caker
21:01<@caker>hah
21:02<@mikegrb>so I finished this game the other day
21:02<@mikegrb>Smugglers Run -- Warzone
21:02<@mikegrb>got to the last level a while back and got stuck
21:02<@mikegrb>finished it and have been on a quest since last night to unlock the last vehicle
21:03<@mikegrb>it's teh hard and agrevating
21:03<@mikegrb>you get a ratting on each mission, OK, Good, Great
21:03<@mikegrb>have to have great on all of them
21:03<@mikegrb>I have Great on 95%
21:03<@mikegrb>the last 5% is angering me :<
21:04<@linbot>pats mikegrb
21:04<@mikegrb>they aren't even levels at the end
21:04<@mikegrb>some are towards the beginning :<
21:05<@caker>meaning you have to start over?
21:05<@mikegrb>nah
21:05<@mikegrb>meaning they should be easier
21:13<gpd>gamecube?
21:13<@mikegrb>yes
21:14|-|tompuppy [~tom@proxy.tom.kao.name] has quit [Quit: Leaving]
21:15<gpd>friend of mine just got one of those Mammoth Toys C64 plug-into-tv things...
21:15<gpd>paradroid - now there is a game!
21:16<gpd>http://www6.tomshardware.com/game/200502191/
21:17<@mikegrb>!!!
21:17<@mikegrb>just got one of the worst done
21:26<thelsdj>can usb power a 3.5" hard drive or do all the ide drive enclosures available require external ac power?
21:38|-|FireSlash [Liger@36.207-191-238.gbrdialin.gbronline.com] has quit [Quit: http://www.FireSlash.com -- Your #1 source for cool, yet totally useless PHP scripts!]
21:44<gpd>thelsdj: I think LaCie pocket drives can use usb2 / firewire power
21:45<gpd>http://www.lacie.com/products/product.htm?pid=10400
21:45<thelsdj>i know there are drives that can use usb2/firewire power i'm just running if there are external enclosures that can run any 3.5" drive off usb2/firewire power
21:45<thelsdj>er
21:45<thelsdj>running = wondering
21:45<thelsdj>not sure how that happened
21:47<warewolf>usually only 2.5" drives can run *solely* off usb2/firewire power
21:47<warewolf>3.5" drives require too much current
21:48<thelsdj>ya ok thats what i was wondering
21:51<gpd>does firewire800 have higher voltage / current spec?
21:53<gpd>http://www.dansdata.com/usbadapt.htm
21:54<gpd>not what you want but quite novel
21:54<thinfu>it appears firewire 800 provides 45 watts
21:54<thinfu>usb 2.0 only provides 2.5 watts
21:54<gpd>wow...roll on usb 3.0
21:55<thinfu>hmm, i guess firewire is gonna become more common for the PC now that apple is moving over to PC..
21:56<fuzzie>roll on people just using the superior spec.
21:56<fuzzie>please.
21:58<thinfu>erm, firewire 800 provides same watts as firewire 400
21:58<thelsdj>so firewire 400 also provides 45 watts? or 45 is wrong?
21:58<thinfu>also provides 45
21:58<thinfu>usb 1 and 2 both provide 2.5 watts
21:59<thelsdj>yea death to usb i say
21:59<thelsdj>but i don't think its going to happen anytime soon
21:59<thelsdj>dug in too deep
22:00<thinfu>give it a few years :P
22:00<thelsdj>well how long did it take to get rid of isa?
22:00<thinfu>i'm still on isa
22:00<thinfu>heh
22:01<thelsdj>what are you using thats isa?
22:01<thinfu>i mean i don't have usb at all
22:01<thelsdj>hah
22:01<thinfu>my keyboard isn't even ps/2
22:01<thinfu>its ATA
22:01<thinfu>woot
22:02<thinfu>ironically i upgraded last year ;P
22:02<@mikegrb>my hard drive is mfm
22:02<thelsdj>shit, i remember days without usb and still with isa but i don't remember ever having a non ps2 keyboard
22:02[~]mikegrb does
22:02<Battousai>me too
22:02<thelsdj>well i bet i did at one point
22:02<Battousai>those crazy thick pins
22:02<thelsdj>but was before my real geek days
22:02<thelsdj>ah
22:02<@mikegrb>my kaypro had telco handset connectors
22:02<thelsdj>like twice as big as ps2?
22:02<@mikegrb>lolz
22:02<thinfu>lol
22:03<thinfu>yeah
22:03<thelsdj>i guess i remember those keyboards
22:03<Battousai>AT keyboards
22:03<@mikegrb>my wyse terminals did too
22:03<thinfu>er yeah i meant AT
22:03<thinfu>not ATA
22:03<thinfu>doh
22:03<thinfu>my comp is fast enough for i686 optimized linux ;P
22:03<thinfu>but just barely..
22:04<thinfu>s/fast/new
22:04<thinfu>being behind the curve is fun!
22:04<thinfu>probably saves money too
22:04<thelsdj>not much...
22:05<thelsdj>you could get a computer better than that for free these days
22:05<fuzzie>something which would likely use less power
22:27<thinfu>at any rate, i fantasize about the day when i upgrade to have a faster computer that everyone else, mwahahah :P
22:27<gpd>thinfu: err... newegg.com... cc out... stop messing about...
22:28<thinfu>tigerdirect for me cuz i'm canadian & cuz its cheaper on the things that matter to me
22:28<gpd>or mwave.com ... one of those 3 are good
22:28<thinfu>hmm, never heard of mwave.com before
22:28<gpd>meanwhile... mine still boots up with /dev/ errors :(
22:29<gpd>mwave.com++ as they accept purchase orders... :)
22:29<thinfu>?
22:29<gpd>so I can order plenty o' stuff for work
22:29<thinfu>ah so they invoice you basically?
22:29<gpd>UC system uses purchase orders instead of real money
22:29<thinfu>UC ?
22:29<gpd>University of California
22:30<gpd>meanwhile... must fix my /dev/ errors on boot...
22:31<gpd>not entirely sure where to start with that problem though...
22:33<@caker>gpd: mwave rocks
22:33<@caker>every single Linode host has come via them
22:33<@caker>always on time, always the best prices
22:34<@caker>Often I'll place an order at 4:00PM, and it's here next day by noon
22:53|-|dddd44 [~dhb55@lostpackets.de] has joined #linode
22:59|-|VS_ChanLog [~stats@ns.theshore.net] has left #linode [Rotating Logs]
22:59|-|VS_ChanLog [~stats@ns.theshore.net] has joined #linode
22:59<dddd44>excuse me .. why the proxy server divided into 2 type: http and socks?? why not use http proxy for every application?? can't i use http for irc?? why i must use sock in irc??anyone??
23:01<@linbot>New news from forums: Mysql development libraries downloads in Linux, Apache, Mysql and PHP (LAMP) Forum <http://www.linode.com/forums/viewtopic.php?t=1842>
23:01<SupaZubon>dddd44: dude, irc is not a MIME protocol
23:01<SupaZubon>dddd44: HTTP is session layer, and socks plugs in almost at the transport layer
23:02<SupaZubon>dddd44: but this isn't a general networking support channel. This is for customers and staff of linode.com
23:07<dddd44>sorry
23:29|-|ElectricElf [~david@electricelf.noc.oftc.net] has quit [Quit: ]
23:29|-|ElectricElf [tor@electricelf.chair.oftc.net] has joined #linode
---Logclosed Sun Aug 28 00:00:24 2005