Timestamps are in GMT/BST.
[0:04] <tessier_> Herkamire: How do you feel about PHP/Post nuke?
[0:05] <tessier_> Some people think they are great and others think they are hideous.
[0:05] <Herkamire> tessier_: don't know anything about it really. some kind of BB?
[0:05] <tessier_> A php based web portal system.
[0:06] <Herkamire> the term "web portal" still makes me cringe
[0:06] <Herkamire> even though I haven't heard my boss say it for a while
[0:07] * MrNaughty (MrNaughty@d199-126-252-82.abhsia.telus.net) Quit ("\(^_^)/' No Soliciting!!! Unless you have legs way, way up and really, really big tits....")
[0:08] <Herkamire> I still can't stop from thinking "what the hell is a web portal?"
[0:09] <Herkamire> seems to me it's a way, way over crowded website.
[0:12] * hirvox (~hirvox@cs181027153.pp.htv.fi) Quit (Read error: 104 (Connection reset by peer))
[0:17] <tessier_> To me it's a place to host some sort of web application
[0:17] <tessier_> Often collaboration or discussion or news. It is pretty broad.
[0:19] <Herkamire> I don't really think I know what it means.
[0:20] <Herkamire> I think it falls under what I would call a "website"
[0:26] <tessier_> Website can be just static information. A "portal" is normally something dynamic.
[0:27] <iip_i2p> <gott> A portal is a paradigm shift on the static data economy.
[0:27] <Herkamire> is that all?
[0:27] <iip_i2p> <gott> A proactive one, at that.
[0:27] <Herkamire> so my wiki is a portal?
[0:28] <Herkamire> or a simple page that shows current weather
[0:42] <tessier_> gott: What does it synergize with?
[0:42] <tessier_> It's gotta synergize.
[0:42] <tessier_> It's gotta leverage too.
[0:43] <tessier_> It's gotta leverage synergies.
[0:45] <iip_i2p> <gott> It synergises with proof-of-concept web solutions.
[0:55] <tessier_> Sounds like it leverages synergies with paradigm shifting proof-of-concept web portal solutions on a pro-active going-forward basis in the static data economy.
[0:56] <iip_i2p> <gott> Well, it's motivated by low-cost self-analysis thanks to IP datagram transfer technology, allowing a cheap and effective Internet Operating System solution.
[0:57] <iip_i2p> <gott> 'cheap' should read 'competitively priced'
[0:59] * TheSeeker (BluJaguar@h-66-167-104-40.lsanca54.dynamic.covad.net) Quit ("Wishing on a star he can no longer see. (Damn LA)")
[1:15] * Herkamire (~jason@h000094d30ba2.ne.client2.attbi.com) Quit ("bed")
[1:26] * MrNaughty (MrNaughty@d198-166-55-147.abhsia.telus.net) has joined #freenet
[1:44] * Pascal666 (Pascal@c-24-13-17-191.client.comcast.net) has joined #Freenet
[1:44] * robilad[afk] (~topic@mpiat2304.ag2.mpi-sb.mpg.de) Quit (Operation timed out)
[1:48] * Filipu (Filipu@host81-153-31-185.range81-153.btcentralplus.com) has left #freenet
[1:48] * Pascal (Pascal@c-24-13-17-191.client.comcast.net) Quit (Read error: 110 (Connection timed out))
[1:48] * Pascal666 is now known as Pascal
[2:19] * piranha_ (~piranha@192.220.223.10) Quit (Read error: 113 (No route to host))
[2:28] * piranha (~piranha@192.220.223.22) has joined #freenet
[2:51] * salahx (salahx@sc1-24.217.174.147.charter-stl.com) Quit ()
[3:51] * iip_i2p (~changate@host.teitel.net) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * Gunnar (~gunnar@34.80-203-209.nextgentel.com) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * lostlogic (~lostlogic@node-402418b2.mdw.onnet.us.uu.net) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * nextime (~nextime@danex.i-m-c.it) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * ixitxachitl_ (dn32418@d66-183-170-199.bchsia.telus.net) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * Ash-Fox (Hal-9000@aak131.neoplus.adsl.tpnet.pl) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * tav|offline (~tav@espians.com) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * Orasis (~orasis@c-66-41-30-194.mn.client2.attbi.com) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * Zorix (Brandon@fl-65-40-46-197.dyn.sprint-hsd.net) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * ShaunMacPherson (~ShaunMac@Sudbury-HSE-ppp3976660.sympatico.ca) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * sanity (~r00t@81-178-66-53.dsl.pipex.com) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * Hadaka (naked@naked.iki.fi) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * bluephile (bluephile@69-160-205-193.clvdoh.adelphia.net) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * Iakin3 (~nb@212.105.104.162) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * ejhuff (~ejhuff@dsl-096.floodcity.net) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * BrianR (brianr@2002:1893:ac33:0:0:0:0:1) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * mazzanet (~mazzanet@203-173-44-61.dyn.iinet.net.au) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * leexgx (~bbtt@cpc2-warr1-5-0-cust27.bagu.cable.ntl.com) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * KenMan (~chaziller@pcp403292pcs.mntcrm01.md.comcast.net) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * cehteh (foobar@cehteh.homeunix.org) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * ysothoth (~ysothoth@68-174-22-46.si.rr.com) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * toad_ (toad@82-32-16-91.cable.ubr03.azte.blueyonder.co.uk) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * vsalento (~vsalento@cs136227.pp.htv.fi) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * Zv2S (~rfc1413@squirrel.owl.de) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * Pascal (Pascal@c-24-13-17-191.client.comcast.net) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * MrNaughty (MrNaughty@d198-166-55-147.abhsia.telus.net) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * oierw (mathew@cpe-69-75-126-12.hawaii.rr.com) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * lolo-laptop (~lostlogic@68.251.84.226) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * mikeDOTd (~mikedotd@plutonium.nexthop.net) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * tessier_ (~treed@wsip-68-224-172-77.sd.sd.cox.net) Quit (sterling.freenode.net irc.freenode.net)
[3:51] * xena (~xena@espians.com) Quit (sterling.freenode.net irc.freenode.net)
[3:53] * sanity (~r00t@81-178-66-53.dsl.pipex.com) has joined #freenet
[3:53] * Pascal (Pascal@c-24-13-17-191.client.comcast.net) has joined #freenet
[3:53] * MrNaughty (MrNaughty@d198-166-55-147.abhsia.telus.net) has joined #freenet
[3:53] * Zorix (Brandon@fl-65-40-46-197.dyn.sprint-hsd.net) has joined #freenet
[3:53] * Gunnar (~gunnar@34.80-203-209.nextgentel.com) has joined #freenet
[3:53] * nextime (~nextime@danex.i-m-c.it) has joined #freenet
[3:53] * ysothoth (~ysothoth@68-174-22-46.si.rr.com) has joined #freenet
[3:53] * BrianR (brianr@2002:1893:ac33:0:0:0:0:1) has joined #freenet
[3:53] * ixitxachitl_ (dn32418@d66-183-170-199.bchsia.telus.net) has joined #freenet
[3:53] * ShaunMacPherson (~ShaunMac@Sudbury-HSE-ppp3976660.sympatico.ca) has joined #freenet
[3:53] * toad_ (toad@82-32-16-91.cable.ubr03.azte.blueyonder.co.uk) has joined #freenet
[3:53] * iip_i2p (~changate@host.teitel.net) has joined #freenet
[3:53] * Ash-Fox (Hal-9000@aak131.neoplus.adsl.tpnet.pl) has joined #freenet
[3:53] * oierw (mathew@cpe-69-75-126-12.hawaii.rr.com) has joined #freenet
[3:53] * mazzanet (~mazzanet@203-173-44-61.dyn.iinet.net.au) has joined #freenet
[3:53] * leexgx (~bbtt@cpc2-warr1-5-0-cust27.bagu.cable.ntl.com) has joined #freenet
[3:53] * lolo-laptop (~lostlogic@68.251.84.226) has joined #freenet
[3:53] * lostlogic (~lostlogic@node-402418b2.mdw.onnet.us.uu.net) has joined #freenet
[3:53] * Hadaka (naked@naked.iki.fi) has joined #freenet
[3:53] * bluephile (bluephile@69-160-205-193.clvdoh.adelphia.net) has joined #freenet
[3:53] * Iakin3 (~nb@212.105.104.162) has joined #freenet
[3:53] * Zv2S (~rfc1413@squirrel.owl.de) has joined #freenet
[3:53] * tav|offline (~tav@espians.com) has joined #freenet
[3:53] * xena (~xena@espians.com) has joined #freenet
[3:53] * ejhuff (~ejhuff@dsl-096.floodcity.net) has joined #freenet
[3:53] * Orasis (~orasis@c-66-41-30-194.mn.client2.attbi.com) has joined #freenet
[3:53] * KenMan (~chaziller@pcp403292pcs.mntcrm01.md.comcast.net) has joined #freenet
[3:53] * cehteh (foobar@cehteh.homeunix.org) has joined #freenet
[3:53] * vsalento (~vsalento@cs136227.pp.htv.fi) has joined #freenet
[3:53] * mikeDOTd (~mikedotd@plutonium.nexthop.net) has joined #freenet
[3:53] * tessier_ (~treed@wsip-68-224-172-77.sd.sd.cox.net) has joined #freenet
[3:53] -NickServ- This nickname is owned by someone else
[3:53] -NickServ- If this is your nickname, type /msg NickServ IDENTIFY <password>
[3:53] -ChanServ- [#freenet] This is the FreeNet channel (http://freenet.sourceforge.net)
[4:26] <Pascal> anyone have a good generic jcpuid.so ?
[4:30] <mazzanet> cat /proc/cpuinfo
[4:31] <iip_i2p> <Newsbyte> hi there
[4:31] <iip_i2p> <Newsbyte> anyone tried out the new stable?
[4:33] <iip_i2p> <Newsbyte> /me prods maza with a long stick
[4:34] <mazzanet> ow
[4:34] <mazzanet> that hurt :(
[4:36] <iip_i2p> <Newsbyte> don't worry, I softened that stick at the end
[4:36] <iip_i2p> <Newsbyte> with one of my many needles ;-)
[4:36] <iip_i2p> <Newsbyte> so..ermm...the new stable?
[4:40] <iip_i2p> <Newsbyte> /me thinks about prodding maza again...
[4:57] * Pascal (Pascal@c-24-13-17-191.client.comcast.net) Quit ()
[4:59] * Pascal (Pascal@c-24-13-17-191.client.comcast.net) has joined #Freenet
[5:04] <iip_i2p> <Newsbyte> anyone used new stable?
[5:17] <ShaunMacPherson> not me, but i heard it has the ability so you dont have to use dyndns?
[5:18] <iip_i2p> <Newsbyte> I heard it was going to make Freenet better and faster
[5:18] <iip_i2p> <Newsbyte> at least in that respect, it failed
[5:19] <iip_i2p> <Newsbyte> thusfar
[5:19] <iip_i2p> <Newsbyte> I'm, ofcourse, very nauty when posting this, like on the maillist ;-)
[5:21] <ShaunMacPherson> :)
[5:21] <ShaunMacPherson> Yes this channel is censored for off channel talk hehe
[5:26] <iip_i2p> <Newsbyte> lol
[5:26] <iip_i2p> <Newsbyte> well, one wonders sometimes, on the devl maillist
[5:26] <iip_i2p> <Newsbyte> well, one wonders sometimes, on the devl maillist
[5:27] <iip_i2p> <Newsbyte> do you think my post on the devel maillist is too pessimistic?
[5:29] <ShaunMacPherson> i didnt read it yet let me see :)
[5:30] <ShaunMacPherson> I've been engaged in Mute, it is making interesting progress you should join their mailing list just for discussion since it discusses anonymous p2p in general. Also this website is for anonymous p2p developers http://board.planetpeer.de/
[5:30] * tyesinclair (~tyesincla@82-41-215-112.cable.ubr12.edin.blueyonder.co.uk) has joined #freenet
[5:30] * tyesinclair (~tyesincla@82-41-215-112.cable.ubr12.edin.blueyonder.co.uk) has left #freenet
[5:32] <ShaunMacPherson> it is a good post
[5:33] <iip_i2p> <Newsbyte> I know; I don't make bad posts, but it's always good for my ego to hear confirmation by someone else ;-)
[5:34] <ShaunMacPherson> :)
[5:34] <ShaunMacPherson> we should discuss this, are you on iip?
[5:36] <iip_i2p> <Newsbyte> indeed
[5:36] <ShaunMacPherson> i will come there, since our talk would be off topic
[5:36] <ShaunMacPherson> at least for devl
[5:37] <iip_i2p> <Newsbyte> yes, talk about freenet would be off-topic ;-)
[6:10] * Hory (~Miranda@81.196.25.110) has joined #FreeNet
[6:10] * piranha (~piranha@192.220.223.22) Quit (Read error: 60 (Operation timed out))
[6:14] <Pascal> if I set maximumThreads=0 and leave threadFactory=Y, do I need to change targetMaxThreads or tfAbsoluteMaxThreads?
[6:15] * iip_i2p (~changate@host.teitel.net) Quit (sterling.freenode.net irc.freenode.net)
[6:25] * piranha (~piranha@192.220.223.43) has joined #freenet
[6:28] * iip_i2p (~changate@host.teitel.net) has joined #freenet
[6:39] <toad_> <iip_i2p> <Newsbyte> yes, talk about freenet would be off-topic ;-)
[6:39] <toad_> eh?
[6:40] * guido^pe (~unknown@dsl-082-082-158-159.arcor-ip.net) has joined #freenet
[6:55] * spaetz (~spaetz@217-162-197-136.dclient.hispeed.ch) has joined #freenet
[6:55] * spaetz (~spaetz@217-162-197-136.dclient.hispeed.ch) Quit (Client Quit)
[6:55] <toad_> hehe, another stable build, another disaster
[6:56] * toad_ should be used to it by now
[6:56] <iip_i2p> <Newsbyte> well, maybe I was a bit premature
[6:56] <toad_> perhaps. do you have any evidence of that?
[6:57] <toad_> i.e. is it working a little better now?
[6:57] <toad_> the RNFs would be caused just by restarting your node
[6:57] <toad_> the DNFs would be initially caused by the RT reset
[7:00] <sanity> newsbyte: premature to suggest throwing out over a year of work because a new build wihch makes massive changes doesn't work hours after its release? you don't say!
[7:01] <verl> any reports of freenet using up 100% cpu? thought maybe it was 1.4.2, so i'm giving 1.5.0 a shot, or perhaps all devs have native biginteger and therefore can't see?
[7:01] <sanity> verl: on startup or after a while of usage?
[7:01] <toad_> after startup it would be perfectly normal
[7:01] <verl> after about 40 mins i usually give up
[7:02] <toad_> however i can run two nodes here and still have 50% cpu usage free often
[7:02] <verl> cause i need to use my computer at the same time
[7:02] <toad_> and that's with crazy logging
[7:02] <verl> some recent unstables didn't have this behaviour
[7:02] <toad_> verl: okay, your computer is a pentium 1 with 128MB of RAM, and "use my computer at the same time" means "run half life 2" ?
[7:02] <verl> but the last on before merge did
[7:03] <toad_> well, run half life 1 :)
[7:03] <toad_> freenet is not going to start working in 40 minutes, sorr
[7:03] <toad_> y
[7:03] <toad_> if you want instant gratification, go get a copy of kazaa
[7:03] <verl> toad: i shall try again!
[7:04] <verl> just seemed a bit unusual so i thought maybe something was wrong
[7:04] <toad_> i didn't merge the NativeBigInt stuff btw
[7:04] <toad_> i left that out
[7:04] <toad_> will merge it when we've got the stuff fully working and in freenet-ext.jar
[7:05] <verl> seemed to make a big change? atleast on the graphs i saw
[7:07] <toad_> what did?
[7:08] <verl> native biginteger
[7:08] <toad_> I can access a fair amount of content on my stable node
[7:08] <toad_> it's not perfect, it's rather slow, but it does seem to get there, a lot of the time
[7:09] <toad_> verl: yeah
[7:09] <toad_> well
[7:09] <toad_> which graphs?
[7:09] <toad_> i know it made connection auth a LOT faster
[7:09] <verl> toad: iakins :)
[7:09] <toad_> but I didn't think we were doing much of that now?
[7:09] <toad_> yeah i mean what was it a graph of?
[7:10] <verl> cpu usuage on his box
[7:10] <verl> comparing with and without native bigint
[7:11] <toad_> over long periods?
[7:11] <toad_> do you have the graphs?
[7:11] <verl> may still be in mirc cache.. gonna have a check
[7:11] <sanity> i remember that graph, it seemed to more than half freenet's cpu usage
[7:11] <toad_> woah
[7:11] <verl> cpu usauge is *much* better now, btw, with 1.5.0
[7:11] <sanity> halve even
[7:12] <verl> ~10% with same amount of traffic
[7:12] <verl> ie. little
[7:12] <toad_> woah
[7:13] <Ash-Fox> If you ever want to learn how to bypass firefox's and mozilla's popup blocking features... just visit www.aol.com
[7:14] <toad_> lol
[7:14] <Ash-Fox> I'm being serious
[7:14] <iip_i2p> <Newsbyte> I thought with the auto-update, I would auto-update the seednodes.ref too, but it didn't
[7:14] <sanity> newsbyte: it should have
[7:15] * toad_ mails Iakin re getting this stuff into freenet-ext.jar
[7:15] <toad_> ir would be pretty useful to me, let alone to everyone else
[7:19] <sanity> toad: i wonder - would it be useful to make a DNF carry a new reference?
[7:19] <toad_> no
[7:19] <toad_> far too easily abusable
[7:19] <toad_> we've been through this
[7:22] <Pascal> the windows bigint code is already in freenet-ext.jar
[7:22] <toad_> Pascal: it is??
[7:22] <Pascal> but noone has commited the linux code yet
[7:22] <toad_> which freenet-ext.jar?
[7:22] * toad_ is attempting to compile
[7:22] <toad_> also is it used?
[7:23] <Pascal> build from cvs and it will be there
[7:23] <Pascal> should be, but I can't test
[7:23] <toad_> hmmm
[7:23] <toad_> onion FEC won't build
[7:23] <toad_> [javac] /usr/src/cvs/Contrib/fecimpl/onion/FECUtils.java:17: package freenet.support does not exist
[7:24] <Pascal> do not build just the contrib
[7:24] <toad_> right
[7:24] <Pascal> it depends on freenet.support
[7:24] <Pascal> from your freenet cvs dir do "ant distclean" then "ant jar"
[7:24] <toad_> but still, half this stuff is native, right?
[7:24] <toad_> so how can it possibly build the fec plugins?
[7:24] <Pascal> the native code is pre-build
[7:24] <Pascal> the native code is pre-built
[7:25] <Pascal> it is building the java part
[7:25] <toad_> hmmm
[7:26] <toad_> okay, it includes jbigi's, and cpuid's for windows
[7:26] <toad_> do they work?
[7:26] <toad_> just using the JAR?
[7:26] <toad_> also includes linux and windows native FEC
[7:27] <toad_> whereas the freenet-ext.jar on the site doesn't include either
[7:27] <toad_> okay
[7:27] * lostlogic_ (~lostlogic@node-402418b2.mdw.onnet.us.uu.net) has joined #freenet
[7:27] <Pascal> also building freenet-ext on windows does not include cpuid right now. someone created both a cpuinfo dir and a CPUinfo and they collide. will have to wait for sf to fix today
[7:27] * lostlogic (~lostlogic@node-402418b2.mdw.onnet.us.uu.net) Quit (Nick collision from services.)
[7:27] * lostlogic_ is now known as lostlogic
[7:27] <toad_> Pascal: I have the cpuid dll though
[7:27] <Pascal> I added the native FEC code to it months ago. e-mailed the list and mentioned it here. noone seemed to care.
[7:27] <toad_> for windows
[7:28] <toad_> Pascal: sorry, I should have chased that
[7:28] <iip_i2p> <gott> toad_
[7:28] <Pascal> if you compile on linux you will get everything
[7:28] * toad_ builds for linux...
[7:28] <toad_> of course all of this works ONLY on x86... anyone got a mac around here?
[7:28] <Pascal> but noone has commited the .so files yet
[7:28] <iip_i2p> <gott> toad_ , you do not hang out on IIP
[7:29] * guido^pe (~unknown@dsl-082-082-158-159.arcor-ip.net) Quit (Remote closed the connection)
[7:29] <toad_> Pascal: I will when I've built them
[7:29] <toad_> then I'm going to merge and release 5086, the Low CPU Usage Build
[7:29] <toad_> after replacing freenet-ext.jar
[7:30] * guido^pe (~unknown@dsl-082-082-158-159.arcor-ip.net) has joined #freenet
[7:30] <Pascal> unfortunatly most people will not get the new freenet-ext.jar automagically
[7:30] <verl> toad: :) also, tell anyone who has trouble with 1.4.2 to try switching to 1.5.0
[7:31] <iip_i2p> <gott> shouldn't they just switch to 1.4.1
[7:31] <toad_> Subject: [freenet-dev] 5085 taking forever to start (if it's going to start)
[7:31] <toad_> hrrm, interesting
[7:31] * toad_ gets similar symptoms (web interface nonresponsive) sometimes...
[7:32] <toad_> verl: NPTL causes hang even on 1.5
[7:32] <iip_i2p> <gott> toad_
[7:32] <toad_> :<
[7:32] <verl> oh look, he's using 1.4.2 :)
[7:32] <toad_> ummm hi gott
[7:32] <iip_i2p> <gott> You are doing good job.
[7:33] * lostlogic decides to switch back to stable now...
[7:33] <toad_> have fun
[7:33] <lostlogic> toad_: you know it...
[7:34] <lostlogic> toad_: when I can afford T1 or greater upstream bandwidth at home, you know I'll run both ;-)
[7:34] <iip_i2p> <gott> I would like to thank you by singing.
[7:35] <iip_i2p> <gott> lA LA lA toad is super lA LA lA toad is swell
[7:35] <iip_i2p> <gott> lA LA lA toad is super lA LA lA toad is swell
[7:35] <iip_i2p> <gott> I don't know why that repeated the text
[7:35] <toad_> hmm, it takes a while to build all the different versions...
[7:35] <iip_i2p> <gott> I only said it once.
[7:35] <toad_> a long while...
[7:36] <iip_i2p> <gott> toad_ , I think that if you came to IIP, you would increase membership because people rather _like_ you
[7:36] <lostlogic> toad_: of which? the jbigi thing?
[7:36] <toad_> lostlogic: yeah
[7:36] <lostlogic> toad_: yeah, it sure does. stupid gmp (whatever that is)...
[7:37] <Pascal> toad: can you get the jcpuid.so generic enough to run on all versions of linux?
[7:38] <lostlogic> Pascal: the jcpuid will be, the jbigi is questionable..
[7:38] <lostlogic> Pascal: jcpuid is a 15 line JNI method call
[7:38] <iip_i2p> <wart> Hi, I have a recurring problem with my node. After its been running for a while (minutes, hours, days - it varies), the mainport stops responding. Actually, it seems like the entire system stops responding. My log file fills up with messages such as: NIOOS.write(byte[],int,int) timed out.tcp/connection: ...
[7:38] <lostlogic> mostly ASM any way, no library calls
[7:38] <Pascal> jbigi is already compiled for most versions
[7:38] <Pascal> http://dev.i2p.net/~jrandom/jbigi/
[7:38] <lostlogic> Pascal: then we'll be all good...
[7:39] <iip_i2p> <wart> Restarting the node usually does the trick, though sometimes the same problem happens after consecutive restarts. Does anyone else have similar problems?
[7:39] <Pascal> lostlogic: I just don't know much about linux but seeing all the different versions of jbigi was wondering if you'd need different jcpuid's too
[7:40] <lostlogic> Pascal: the point of jcpuid is to pick the proper jbigi
[7:40] <Pascal> right, but doesn't it have to be different for x86 vs sparc vs arm vs whatever too?
[7:41] <lostlogic> Pascal: it'd actually be a different library entirely for those... currently in java, we detect if we are on x86, and if so we run jcpuid to pick a jbigi ...
[7:41] <toad_> jbigi ought to run on everything because it doesn't call libc, right?
[7:41] <lostlogic> the cpuid instruction doesn't exist except in x86
[7:41] <toad_> <iip_i2p> <wart> Hi, I have a recurring problem with my node. After its been running for a while (minutes, hours, days - it varies), the mainport stops responding. Actually, it seems like the entire system stops responding. My log file fills up with messages such as: NIOOS.write(byte[],int,int) timed out.tcp/connection: ...
[7:41] <toad_> aha
[7:41] <lostlogic> there are other similar instructions on them...
[7:42] <lostlogic> wart: we got you...
[7:42] <Pascal> oh, ok, that makes sense then. thanks
[7:42] <toad_> yeah, i think i've occasionally seen that
[7:42] <toad_> most of the time it's just sloooow
[7:42] * lostlogic gone till I'm back laters, folks
[7:42] <toad_> some of the time that's just due to cpu overload
[7:42] <iip_i2p> <wart> lostlogic, eh?
[7:42] <toad_> if the cpu is pegged for long the node generally gets problems
[7:43] <iip_i2p> <gott> wart: do you like toad ?
[7:43] <Pascal> lostlogic: he didn't repeat, toad repeated him
[7:43] <iip_i2p> <wart> say I like toad
[7:43] <iip_i2p> <gott> I like toad, too =)
[7:43] <toad_> often it appears broken that way just because you've got loads of downloads working
[7:44] <toad_> loads of fproxy pages coming down
[7:44] <toad_> the problem is they don't get killed when they lose the conn
[7:44] <toad_> need to do something about that
[7:45] <toad_> ideally i'd like to make fproxy fully nonblocking...
[7:45] <toad_> that would make things somewhat nicer...
[7:45] <toad_> lets see...
[7:46] <toad_> we essentially would make the servlet container work in non-blocking mode
[7:46] <toad_> and then when it finally decides which servlet to use, it checks whether it is actually a servlet
[7:46] <toad_> because we let it also register NIOServlet's
[7:46] <toad_> and some of the things that ARE servlets would implement QuickAndDirtyCacheMe
[7:47] <toad_> in which case we'd just spool to RAM and then send back non-blocking
[7:47] <Pascal> toad: are you planning on working on fproxy soon? I was going to start looking at it tomorrow
[7:47] * hirvox (~hirvox@cs181027153.pp.htv.fi) has joined #freenet
[7:47] <Pascal> toad: but I won't bother if you are
[7:48] * robilad[afk] (~topic@mpiat2304.ag2.mpi-sb.mpg.de) has joined #freenet
[7:48] <iip_i2p> <gott> toad_: I think foremost in development plans should be giving the nodes sentience.
[7:48] <toad_> Pascal: what were you going to do with it?
[7:49] <iip_i2p> <wart> toad_, Well, what can I do in the interim? Fuqid is probably thrashing the node. Are there common numbers that programs like fuqid and frost can use for insert/download threads or concurrent downloads that are optimal? Or would this vary based on system and connection resources?
[7:49] <toad_> okay, generated libjbigi-*
[7:49] <toad_> wart: well, I doubt FUQID has much to do with it
[7:49] <Pascal> toad: right now every single request gets written to a tempfilebucket as plain text before being sent to the requester. I'd like to bypass the buckets.
[7:50] <toad_> do you have lots of requests going THROUGH THE WEB INTERFACE ?
[7:50] <iip_i2p> <wart> toad, No, none
[7:50] <hirvox> hrm, NativeBigInteger doesn't seem to recognize plain libjbigi.so
[7:50] <lostlogic> hirvox: should be libjbigi-none.so IIRC
[7:50] <lostlogic> if you are using the jcpuid thing that is
[7:51] <toad_> libjbigi-linux-athlon.so libjbigi-linux-none.so libjbigi-linux-pentium2.so libjbigi-linux-pentium3.so libjbigi-linux-pentium4.so
[7:51] <iip_i2p> <wart> toad, But generally this doesn't start happening until I activate fuqid
[7:51] <hirvox> jbigi-linux-none.so inside classpath in net/i2p/util works, though
[7:51] <hirvox> but I'd prefer them to be outside the jar
[7:51] <toad_> wart: ooooh
[7:51] <toad_> wart: sadly, I can't run FUQID, because it's windows only
[7:51] <toad_> actually I heard it runs under wine?
[7:51] <hirvox> woah, 10kpq
[7:51] <hirvox> 10kqph
[7:51] <toad_> would Frost produce a similar effect?
[7:52] <iip_i2p> <gott> your anti-windows obsession costs the project dearly
[7:52] <toad_> lol
[7:52] <iip_i2p> <wart> toad, And again, I question whether the web interface is the only system affected by this. I usually have a busy log file, but when this problem kicks up, all log messages, save for the one I pasted for you, cease
[7:52] <Pascal> toad: the way it is now when you run an open proxy whatever files are pulled through the web interface get left on Fred's HDD as plain text and could get the node operator in trouble.
[7:52] <iip_i2p> <wart> toad It runs very well on wine
[7:52] <iip_i2p> <gott> CofE agrees with this statement. If CofE agrees, it is true.
[7:52] <toad_> okay
[7:52] <toad_> I've built the .so's for NBI
[7:52] <toad_> what about cpuid?
[7:52] <toad_> how do i build that?
[7:52] * tyesinclair (~tyesincla@170.252.64.1) has joined #freenet
[7:53] <iip_i2p> <gott> I must say, CofE attracts me. I imagine him as being a British schoolteacher that dabs in deviatory sexual practices like every Freeneter does. Oh, one of these days I shall be his lolita.
[7:53] <hirvox> run Contrib/cpuid/build.sh, and make sure that JAVA_HOME is defined
[7:53] <Pascal> once built put them all in the same place their corresponding dlls are
[7:53] <iip_i2p> <gott> Or dare I say, lolito ?
[7:53] <iip_i2p> <wart> rofl
[7:54] <Pascal> hirvox: that won't do any cpu optimizations right?
[7:54] <toad_> Pascal: well, I just have to build the freenet-ext.jar then, and it should work, right?
[7:54] <Pascal> toad: right
[7:54] <hirvox> Pascal:correct, that's why I don't use build.sh when building my RPMs
[7:54] <iip_i2p> <gott> wart: I think I may be on your ignore list, because you only seem to hear me via eyekon
[7:54] <toad_> Pascal: IMHO, jcpuid.so should be jcpuid-x86.so
[7:55] <Pascal> so change it
[7:55] <Pascal> I agree
[7:55] <Pascal> but I didn't write it
[7:55] <toad_> hmmm
[7:55] <toad_> oh well, same for the others
[7:55] <Pascal> the way I setup the build.xml anything in those dirs will be put in the jar
[7:56] <Pascal> doesn't matter what you call them
[7:56] <toad_> we can deal with that when we actually have some non-x86 users wanting to build libs :)
[7:56] <Pascal> just have to make sure the java cod looks for the right file
[7:56] <Pascal> :)
[7:57] <toad_> okay...
[7:57] <toad_> we have... all the libs, it appears
[7:57] <toad_> how do i do that performance test?
[7:58] <hirvox> java net.i2p.util.NativeBigInteger
[7:58] <iip_i2p> <wart> toad, I don't see why a similar effect wouldn't be produced by frost... I'm doing a dozen simultaneous downloads with 75 threads and two simultaneous inserts with 25 threads. So long as frost lets you control those parameters, it should be reproducable...
[7:58] <toad_> ERROR: Resource name [freenet/support/CPUInformation/jcpuid.so] was not found
[7:59] <toad_> WARN: Native CPUID library jcpuid not loaded - using pure java
[7:59] <toad_> NOTICE: Resource name [net/i2p/util/jbigi-linux-none.so] was not found
[7:59] <toad_> sh*t
[7:59] <toad_> this is running java -cp lib/freenet.jar:lib/freenet-ext.jar net.i2p.util.NativeBigInteger
[8:00] <toad_> lib/freenet/support/CPUInformation/jcpuid.so exists in freenet-ext.jar
[8:00] <Pascal> I'm wondering about the lib/
[8:00] <hirvox> ditch the lib, just freenet/support/CPUInformation
[8:00] <Pascal> I wasn't sure about that
[8:00] <toad_> okay, how to fix?
[8:00] <Pascal> I fix, 1 sec
[8:00] <toad_> the native FEC stuff has similar issues
[8:01] <Pascal> no, native fec is good
[8:01] <toad_> but it's in lib/ !
[8:01] * tyesinclair (~tyesincla@170.252.64.1) has left #freenet
[8:01] <iip_i2p> <gott> TOAD
[8:01] <Pascal> yes, thats why I put jbigi there
[8:01] <iip_i2p> <gott> <wart> Well, it seems toad doesn't care that I can't download transvestite erotica without wrecking my node
[8:01] <toad_> ah, i see
[8:01] <hirvox> that reminds me, I need to check whether I can redefine where the FEC classes look for the .so's
[8:01] <toad_> wart: sorry, I'm busy
[8:01] <iip_i2p> <wart> * wart ties a paper bag around gott's face
[8:01] <toad_> often that sort of problems are caused by CPU usage
[8:01] <iip_i2p> <wart> plastic*
[8:01] <toad_> I'm fixing the CPU usage
[8:02] <iip_i2p> <wart> toad, Cool beans. I'll do what I can to keep CPU usage down (wine is a hog). Thanks
[8:03] <toad_> wine shouldn't be a hog
[8:03] <toad_> IIRC FUQID is just badly written :)
[8:03] <Pascal> ok, cvs update
[8:03] <Pascal> contrib
[8:03] <toad_> Pascal: on which?
[8:03] <toad_> ok
[8:03] <iip_i2p> <gott> toad_: fuqid works great.
[8:03] <iip_i2p> <wart> Aye, I should say fuqid running in Wine is a hog
[8:03] <toad_> I didn't tell it to commit them as any sort of special files
[8:03] <Pascal> you did the commit already?
[8:03] <toad_> Pascal: do i need to rebuild?
[8:03] <toad_> rebuild which part?
[8:04] * greycat (~wooledg@192.35.79.70) has joined #freenet
[8:04] <Pascal> "ant distclean" "ant jar"
[8:04] <toad_> well the windows ones were committed already
[8:04] <toad_> ok
[8:04] <toad_> well, i heard something about fuqid using 100% cpu usage always from somewhere else
[8:04] <Pascal> I checked the windows ones, they were all commited correctly as binary
[8:04] <toad_> which is a little sad considering it's supposed to run BACKGROUNDED :)
[8:05] <toad_> Pascal: hmmm
[8:05] <toad_> i think CVS will autodetect
[8:05] <toad_> but I could be wrong
[8:05] <toad_> please check my binaries
[8:05] <hirvox> my Fuqid seems to behave, though
[8:05] <Pascal> checking now
[8:05] <toad_> ERROR: Resource name [freenet/support/CPUInformation/jcpuid.so] was not found
[8:05] <toad_> sh*t
[8:05] <Pascal> WTF
[8:05] <Pascal> is it in the jar correctly?
[8:05] <toad_> it's still in lib/freenet/support/CPUInformation/jcpuid.so
[8:05] <iip_i2p> <wart> It consumes 30%-40 when running under wine
[8:06] * toad_ will recompile
[8:06] <toad_> onion:
[8:06] <toad_> jcpuid:
[8:06] <toad_> bigint:
[8:06] <toad_> hmmm
[8:06] <Pascal> the did a cvs update on contrib then "ant distclean"
[8:06] <Pascal> from the freenet directory
[8:06] <toad_> aha
[8:07] <Pascal> the .so's all went in as text. fixing now
[8:07] <toad_> Pascal: will you have to rebuild them?
[8:07] <toad_> Yes, OP worries are premature. I've just upgraded an hour ago and I'm
[8:07] <toad_> getting decent performance.
[8:07] <toad_> wooh, positive feedback on 5085 !
[8:07] <Pascal> no, "cvs admin" will fix
[8:08] <toad_> Pascal: so you haven't lost the data?
[8:08] <Pascal> no
[8:08] <Pascal> if the data was lost, you would have to rebuild them :)
[8:08] <toad_> aha...
[8:08] <toad_> NOTICE: Resource name [net/i2p/util/jbigi-linux-athlon.so] was not found
[8:08] <toad_> NOTICE: Resource name [net/i2p/util/jbigi-linux-none.so] was not found
[8:08] <toad_> INFO: Native BigInteger library jbigi not loaded - using pure java
[8:08] <toad_> hmmm
[8:08] <toad_> you fixed the cpuid
[8:08] <toad_> but not the rest
[8:09] <hirvox> the build scripts don't rename libjbigi.so, you need to do it manually
[8:09] <toad_> WTF?
[8:09] <toad_> net/i2p/util/libjbigi-linux-athlon.so
[8:09] <toad_> hirvox: that's what CPUID is for !!
[8:09] <hirvox> libjbigi!=jbigi
[8:09] <toad_> I'm only interested in doing this if we can put it into the jar properly...
[8:09] <toad_> ah
[8:09] <toad_> bugger
[8:09] <toad_> why?
[8:10] <toad_> Pascal: will you deal with that too?
[8:10] <iip_i2p> <gott> toad_: I'm having problems with my node.
[8:10] <Pascal> ok, marked as binary now
[8:11] <toad_> Pascal: apparently the filenames are wrong
[8:11] <toad_> for some silly reason the java code expects jbigi-....so, whereas the build script generates libjbigi...so
[8:11] <Pascal> file renames in cvs are a bitch
[8:11] <Pascal> fix the java
[8:11] <toad_> so I'll cvs remove, rename, cvs add, cvs commit?
[8:11] <toad_> ok
[8:11] <toad_> well the problem is the windows ones already use jbigi...dll
[8:11] <iip_i2p> <wart> ugh...
[8:12] <Pascal> hahah
[8:12] <Pascal> well we need to fix either the build script or the java
[8:12] <toad_> arguably the linux ones SHOULD be lib...so and the windows ones should just be jbigi...dll
[8:12] <toad_> that's the convention
[8:12] <Pascal> ok, then fix the java :)
[8:12] <toad_> but it'd be nice to be consistent
[8:12] <toad_> I don't like special casing it...
[8:13] <Pascal> it isn't already special cased?
[8:13] <toad_> how hard is it to cvs remove the files?
[8:13] <toad_> hmmm, i'll check
[8:13] <hirvox> and while you're at it, make sure that it also recognizes libjbigi.so if it's in LD_LIBRARY_PATH or -Djava.library.path
[8:13] <Pascal> remove and re-add is easy. rename is hard
[8:13] * Pascal slaps hirvox
[8:14] * toad_ fixes the code
[8:14] <toad_> the java, that is
[8:15] * toad_ waits for disk to spin up
[8:15] <toad_> linux 2.6's swap starvation can be REALLY irritating
[8:15] <toad_> okay...
[8:15] <iip_i2p> <wart> toad, Aren't there patchsets that aim to reduce such inconveniences?
[8:16] <Pascal> no i586 or i686 libraries?
[8:17] <hirvox> Pascal:?
[8:17] <toad_> now, what was that command line?
[8:17] <toad_> Pascal: huh?
[8:17] <iip_i2p> <Newsbyte> toad, I heard you responded to my post, but I don't see it on twe web-mailistarchivethingy
[8:17] * robilad[afk] is now known as robilad
[8:17] <Pascal> I was just looking at what toad commited, and someone mentioned yesterday they were using a i586
[8:18] <toad_> ah
[8:18] <toad_> i.e. pentium1
[8:18] <Pascal> nod
[8:18] <toad_> INFO: Native CPUID library jcpuid loaded from resource
[8:18] <toad_> INFO: Optimized native BigInteger library 'libjbigi-linux-athlon.so' loaded from resource
[8:18] <toad_> yay!
[8:18] * Pascal cheers
[8:18] <toad_> how long does it take for the RNG to warm up?
[8:18] <Pascal> how many meg is the jar now?
[8:19] <toad_> 954,873 bytes
[8:19] <Pascal> hahahha
[8:19] <toad_> it'll only get big when hirvox adds his stuff to it :)
[8:19] <toad_> native run time: 4416ms (43ms each)
[8:19] <toad_> java run time: 14258ms (141ms each)
[8:19] <toad_> native = 30.972085846542292% of pure java time
[8:19] <toad_> ah, that's the second one
[8:19] <toad_> the first one:
[8:19] <toad_> native run time: 7316ms (72ms each)
[8:19] <toad_> java run time: 58090ms (575ms each)
[8:19] <toad_> native = 12.59425030125667% of pure java time
[8:20] <toad_> i presume the second one is toDouble and the first one is modPow?
[8:20] <Pascal> the one we are distibuting now is only 124k :)
[8:20] <iip_i2p> <Newsbyte> and thus, toad, it would help if the seednodes.ref would get auto-updated too
[8:20] <toad_> so I have a factor of 8 gain for modPow, and a factor of 3 for toDouble.. cool!!
[8:20] <toad_> Newsbyte: isn't it?
[8:21] <toad_> it's in dodo's crontab...
[8:21] <iip_i2p> <Newsbyte> that was the reason I couldn't get one lousy connection
[8:22] <iip_i2p> <Newsbyte> look, I did an update, and it didn't get updated, at least in my case
[8:22] <iip_i2p> <Newsbyte> I updated it yesterday
[8:22] <toad_> Newsbyte: how often??
[8:22] <toad_> it takes considerable time for it to get updated
[8:22] <toad_> because of how we get it
[8:22] <sanity> toad: almost ready to upload the new freenet-ext.jar?
[8:22] <toad_> sanity: yup
[8:22] <toad_> i'm just about to automate it on dodo
[8:22] <Pascal> toad: test the native fec for me first
[8:23] <toad_> Pascal: how?
[8:23] <toad_> okay, so i shouldn't automate it?
[8:23] <Pascal> toad: it's been so long since I added it I'm not sure about it anymore
[8:23] <Pascal> I don't remember how
[8:23] <iip_i2p> <gott> toad_: if I tell you something, will you listen ?
[8:24] <toad_> gott: I will listen, it's a matter of whether I will do anything about it
[8:24] <iip_i2p> <Newsbyte> what you mean: 'how often'?
[8:24] <iip_i2p> <Newsbyte> how often should I update to the next version? ;-)
[8:24] <toad_> Newsbyte: how old is the seednodes.ref now?
[8:24] <iip_i2p> <gott> toad_: I think I should tell you something for better understanding between us
[8:24] * toad_ restarts node... it'll probably mumble something about native FEC
[8:24] <toad_> on startup
[8:25] <iip_i2p> <Newsbyte> I just downloaded it an hour ago, now
[8:25] <toad_> Newsbyte: and? what's the problem?
[8:25] <hirvox> toad:it mumbles about native FEC only when it needs it
[8:25] <toad_> hirvox: hrrm
[8:25] <Pascal> toad: only when you use it
[8:25] <toad_> it doesn't if it loads it successfully?
[8:25] <toad_> okay, so i'll use it
[8:25] <Pascal> it loads only when used
[8:25] <iip_i2p> <Newsbyte> but, besides that, and though Ian will crusify me, I don't think my criticism was totally unwarranted
[8:25] <Pascal> the first time you use it it loads
[8:25] <toad_> right but it'll say something on stderr when I do use it?
[8:26] <hirvox> yes
[8:26] <Pascal> yes
[8:26] <toad_> Newsbyte: that's your problem
[8:26] <toad_> tell me what you would have me do
[8:27] <toad_> how come there's no lib optimized for pentium 1?
[8:27] <iip_i2p> <Newsbyte> problem? ermm...now I DO get at least two activelinks, but with my former update I didn't get one iota
[8:27] <toad_> all our users on old firewalls won't benefit!
[8:28] <iip_i2p> <Newsbyte> what you mean, my problem? I think most would expect, with an auto-update, that the seednodes get updated too; I had to install it afterwards, thnks to the recommandation of some dude on IIP
[8:29] <toad_> Newsbyte: they do
[8:29] <toad_> they're just not used unless it's absolutely vital
[8:29] <iip_i2p> <Newsbyte> well, aparently they don't do it always
[8:30] <toad_> Newsbyte: so?
[8:30] <toad_> INFO: Native CPUID library jcpuid loaded from resource
[8:30] <toad_> INFO: Optimized native BigInteger library 'libjbigi-linux-athlon.so' loaded from
[8:30] <toad_> resource
[8:30] <toad_> hmmm
[8:30] <toad_> it doesn't say anything about native FEC
[8:30] <iip_i2p> <Newsbyte> if you get absolutely no contact with any other nodes, I would call it absolutely vital, but I still had to install it again afterwards
[8:31] <toad_> hmmm
[8:31] <toad_> it seemed to encode pretty fast...
[8:31] <iip_i2p> <Newsbyte> look, toad, so, so....one would better check if the update works like it should, instead of soing ;-)
[8:31] <toad_> Newsbyte: I see no reason to make every stable build automatically reseed
[8:31] <toad_> and plenty not to
[8:32] <Pascal> toad: if you did a fec operation and it didn't say the library wasn't found then it was
[8:32] <iip_i2p> <Newsbyte> no, but one could reasonably expect it to do so with new builds that have major differences that may cause networkproblems
[8:33] <toad_> Newsbyte: why? there are no new nodes to seed FROM
[8:33] <toad_> what you want is a network reset
[8:33] <toad_> I'm sorry I didn't do that, it would have pissed off those users that didn't download the new seednodes automatically
[8:33] <toad_> i.e. most of them
[8:33] <iip_i2p> <Newsbyte> basically, what I want is an update that actually updates, so that it works....call me crazy ;-)
[8:33] <toad_> Pascal: you sure?
[8:34] <Pascal> bbiab
[8:34] <toad_> Newsbyte: you're crazy
[8:34] <toad_> emergent systems never instantly work
[8:34] <Pascal> toad: put the old jar back and do the same test
[8:34] <toad_> I could have reset the network
[8:35] <iip_i2p> <Newsbyte> hehe...you're neuty, toad. But still, with such a major revision (as you have said yourself on the maillist) it would have been rather logical to update the seednodes too
[8:36] <toad_> nno, it wouldn't
[8:36] <toad_> because the nodes HAVE NOT CHANGED
[8:36] <Pascal> toad: if it doesn't work, i'll be back in 1/2 hour
[8:37] <toad_> it didn't show anything
[8:37] <iip_i2p> <Newsbyte> nodes have not cyhanged! pfff! How do you explain it turned totally into crap after my update, then?
[8:37] <Pascal> toad: of course you still have to test everything on windows too
[8:37] <toad_> Pascal: it didn't show anything either way re FEC
[8:37] <hirvox> Newsbyte:because the estimator format was changed, and thus all estimators were reset
[8:37] <Pascal> didn't show anything? your now redirecting stdout to null or anything are you?
[8:38] <hirvox> thus the network suffered complete amnesia
[8:38] <toad_> Newsbyte:1)) you restarted your node. that mmeans the node turns into crap for hours as it reestablishes conns
[8:38] <Pascal> s/now/not
[8:38] <toad_> Pascal: I see nothing on stderr, and nothing in the logfile
[8:38] <Pascal> I have to go..bbiab
[8:40] <iip_i2p> <gott> toad
[8:40] <iip_i2p> <Newsbyte> I find this explanation doubtfull. After running 10 hours and restrting it several times, I just didn't get anything. Nothing at all. then some guy suggest to reseed, I do so, and immediately afterwards I get at least some activelinks. That doesn't seem to be in accordance with what you say is the cause.
[8:41] <toad_> /ignore Newsbyte... owait, that doesn't work here...
[8:41] <toad_> :)
[8:41] <toad_> if your node stops working then logically reseeding may help
[8:41] <iip_i2p> <Newsbyte> You should be thankfull that I call your work crap!!
[8:41] <hirvox> lol
[8:42] <iip_i2p> <Newsbyte> ;-)
[8:43] <iip_i2p> <gott> toad, I want you to know something. Are you listening ?
[8:45] <toad_> Unable to find native library for fec8
[8:45] <toad_> Unable to find native library for fec16
[8:45] <toad_> aha
[8:45] <toad_> it was my fault...
[8:45] <iip_i2p> <gott> god dammit, toad
[8:45] <iip_i2p> <gott> this involves us here
[8:45] <iip_i2p> <gott> be christian and listen !
[8:45] * nicktastic (~nicktasti@c-67-163-227-74.client.comcast.net) has joined #freenet
[8:46] <toad_> i am (regretfully!) listening, gott, go ahead and say whatever it was
[8:46] <iip_i2p> <Newsbyte> I concur; everything is your fault, toad! :-)
[8:47] <lolo-laptop> toad_: you didn't merge the jcpuid stuff to stable in the big merge?
[8:47] <toad_> lolo-laptop: not yet
[8:47] <iip_i2p> <gott> toad, I have said many negative things to you and joked with you quite a bit
[8:47] <lolo-laptop> nod.
[8:47] <toad_> I'm about to up the freenet-ext.jar
[8:47] <toad_> after I finish testing native FEC
[8:47] <iip_i2p> <gott> I'm not sure how you take it, but I have the feel you think I don't like you
[8:47] <toad_> later I will put out 5086 with the new code
[8:47] <iip_i2p> <gott> not that you'd care, but I should set the record straight
[8:47] <lolo-laptop> toad_: nod.
[8:48] <toad_> I have to go soon
[8:48] <iip_i2p> <gott> I think you are a super person and what you are doing for the world is splendid !
[8:48] <iip_i2p> <gott> *hug!*
[8:48] * lolo-laptop gets a tear in his eye
[8:48] <lolo-laptop> wasn't that sweet!
[8:49] <toad_> Made 7 262144 byte check blocks in 484ms.
[8:49] <toad_> aha
[8:49] <toad_> that means it's using native, right?
[8:49] <toad_> so why is the client still ENCODING ?
[8:50] <hirvox> toad:no, it displays the same message without native fec
[8:51] <toad_> SEC Jul 23, 2004 2:06:14 PM (freenet.client.FECTools, main, DEBUG): Loaded FECEn
[8:51] <toad_> coder [OnionFECEncoder]
[8:51] <toad_> SEC Jul 23, 2004 2:06:14 PM (freenet.client.FECTools, main, DEBUG): Loaded FECDe
[8:51] <toad_> coder [OnionFECDecoder]
[8:51] <iip_i2p> <Newsbyte> and where is your response on my post, toad?
[8:53] <toad_> feck
[8:53] <toad_> so how am i supposed to tell the difference?!?!?
[8:55] <Pascal> bak
[8:55] <iip_i2p> <Newsbyte> by looking at it a bit closer?
[8:55] <toad_> Pascal: doees it show "Made 7 262144 byte check blocks in 484ms." either way?
[8:56] <Pascal> yes
[8:56] <toad_> aaargg
[8:56] <toad_> so how am I supposed to tell the difference?
[8:56] <Pascal> Unable to find native library for fec16
[8:56] <toad_> especially now that FCP is playing up
[8:56] <Pascal> Unable to find native library for fec8
[8:56] <Pascal> should not show if it found the native libs
[8:56] <toad_> hmmm
[8:57] <toad_> i wonder if that's the bug people were talkign about...
[8:57] <toad_> maybe WSL gets stuck...
[8:57] <Pascal> it should show those two lines with the old jar but not the new one. is that what it is doing?
[8:58] <toad_> it just gets stuck encoding
[8:58] <toad_> it doesn't repond at all
[8:58] <Pascal> hmmm
[8:58] <toad_> it gets to encoding and then stops
[8:58] <toad_> even though it says it's encoded them
[8:58] <toad_> this probably isn't related to freenet-ext.jar
[8:58] <Pascal> so it gets stuck on insert?
[8:59] <toad_> but i should probably find out before rolling out the new freenet-ext.jar?
[8:59] <iip_i2p> <Newsbyte> indeed. But now, let's concentrate on the testnetwork :-) (me thinks toad will set me on ignore, now)
[8:59] <toad_> ah, it's gone through
[8:59] <Pascal> cool
[8:59] <toad_> iit jusok aaaaaaaaaaaaaaaages
[8:59] <Pascal> you do the tests on windows yet?
[8:59] <toad_> it just took ages
[9:00] <toad_> ah, i've still got stable node running...
[9:01] <Pascal> what was the command to test bigint again?
[9:01] <hirvox> java net.i2p.util.NativeBigInteger
[9:02] <Pascal> ok, confirmed working on windows
[9:02] <toad_> ugh
[9:02] <toad_> latest idiocy on tech:
[9:02] <toad_> Subject: Fighting bandwidth hogs (Re: [Tech] Split File (or Chunk) Size
[9:02] <toad_> (was: Incompatible changes?)
[9:03] <iip_i2p> <Newsbyte> So, you deem my problems was a sign of a freak occurence, probably a punishement of God, that will not cause any problems to anyone else?
[9:03] <Pascal> toad: you know, you'd like longer if you didn't read tech
[9:03] * thelema|away is now known as thelema
[9:03] * thelema has 5 minutes. What's up
[9:04] <toad_> Pascal: Ian insists that anything that isn't absolutely totally strictly on topic for devl goes on tech
[9:04] <toad_> so many of my iddeas have to go to tech
[9:04] <toad_> let alone everyone else
[9:04] <toad_> okay
[9:04] <toad_> uploaded freenet-ext.jar
[9:04] <Pascal> cool
[9:04] <toad_> bbl
[9:04] <toad_> later I will merge the code to use it properly
[9:04] <iip_i2p> <Newsbyte> well, everyone know Ian is
[9:05] <iip_i2p> <Newsbyte> wait a sec...is he online, there?
[9:05] <Pascal> toad: what?
[9:05] <toad_> Pascal: nativebigint support
[9:05] <toad_> not in stable yet
[9:05] <toad_> bbl
[9:05] <Pascal> oh merge, thought you said commit :) sorry
[9:06] <iip_i2p> <Newsbyte> Ian means well...but it's best to ignore him when he becomes belittling ;-)
[9:08] <Pascal> we should prob move freenet\src\net\i2p\util\NativeBigInteger.java and freenet\src\EDU\oswego\cs\dl\util\concurrent\* into contrib as well before telling people to update freenet-ext.jar
[9:08] <Pascal> toad: were you planning on working on fproxy soon or not?
[9:11] * thelema is now known as thelema|DTT
[9:11] <hirvox> Pascal:I agree, then I can take the concurrent classes away altogether and just use the jpackage concurrent rpm :)
[9:12] <Pascal> hirvox: I have absolutly no idea what you just said
[9:12] <Pascal> I just meant that they are contributed code and thus should be in contrib :)
[9:13] <hirvox> EDU\oswego\cs\dl\util\concurrent\* is available as an rpm, so I can just use the rpm instead of bundling them in freenet-ext.jar
[9:13] <Pascal> but most people couldn't use the rpm, so we would put them in freenet-ext.jar, so why would you prefer to use the rpm?
[9:15] <hirvox> Pascal:just because they are in contrib doesn't mean that *I* have to include them in *my* freenet-ext.jars. other people can use the one in contrib
[9:15] <iip_i2p> <Newsbyte> yes, well...can someone send me the response of toad to my post on devl?
[9:15] <iip_i2p> <Newsbyte> yes, well...can someone send me the response of toad to my post on devl?
[9:16] <hirvox> I prefer cleaner RPMs with proper dependencies to make sure that there's no duplicated code
[9:16] <hirvox> AFAIR onion fec also uses concurrent
[9:17] <iip_i2p> <Newsbyte> Newsbyte@freenethelp.org
[9:17] <iip_i2p> <Newsbyte> pls
[9:17] <iip_i2p> <Newsbyte> pretty pls
[9:17] <iip_i2p> <Newsbyte> with suggar on top
[9:18] * Ash-Fox (Hal-9000@aak131.neoplus.adsl.tpnet.pl) Quit (Read error: 110 (Connection timed out))
[9:18] * robilad is now known as robilad[afk]
[9:18] <hirvox> one of my long-term projects is to dissect onion fec into proper RPMs. unfortunately jpackage doesn't support nested classpaths (as in: freenet requires onion-fec, and onion-fec requires log4j, and log4j is included automagically in freenet's classpath)
[9:19] * Ash-Fox (Hal-9000@aad77.neoplus.adsl.tpnet.pl) has joined #FREENET
[9:19] <Pascal> Newsbyte: why can't you just read it on the website like evryone else?
[9:24] <Pascal> hirvox: so you'll be working on jpackage for a bit then?
[9:25] <hirvox> well, not really. my RPMs are jpackage-compatible, but that's about it
[9:25] <hirvox> using jpackage's tools saves me a lot of headaches (like locating the JVM)
[9:26] <iip_i2p> <Newsbyte> pascal: that's what I would want to know too
[9:26] <iip_i2p> <Newsbyte> I don't see it anywhere, while ppl tell me he responded
[9:28] <iip_i2p> <Newsbyte> Is it on devel?
[9:29] <hirvox> yes
[9:29] <hirvox> try gmane
[9:30] <iip_i2p> <Newsbyte> http://news.gmane.org/gmane.network.freenet.devel ... nothing to see
[9:32] <hirvox> hrm, apparently not
[9:32] <hirvox> I'll forward it
[9:32] <iip_i2p> <Newsbyte> woohoo!
[9:33] <iip_i2p> <Newsbyte> you're an angel hirvox (sort of)
[9:34] <iip_i2p> <Newsbyte> a fallen angel, no doubt, but still...
[9:37] <Pascal> Now that we have good JNI examples in place, I wonder if we should add code to detect disk space on install set set storesize accordingly
[9:37] <Pascal> s/set set/and set
[9:58] * greycat (~wooledg@192.35.79.70) Quit (Remote closed the connection)
[10:06] * greycat (~wooledg@192.35.79.70) has joined #freenet
[10:12] * piranha (~piranha@192.220.223.43) Quit (Read error: 60 (Operation timed out))
[10:26] * piranha (~piranha@192.220.223.37) has joined #freenet
[10:49] <KenMan> so we only have ~60 nodes on the new unstable, thus far ?? why don't i get more peers ? toad said he had 200 or more...
[10:51] * KenMan has reproduced toad's 80% routing success results with testnet ...
[10:52] <hirvox> 4.7% max pSuccess on unstable
[10:55] <Pascal> Number of distinct nodes connected 86
[10:56] <KenMan> max pSuccess=3.9% max here, yet requestSuccessRatio/hr (for 10 hours) average is 6.1% ...
[10:56] <KenMan> rather, that was averaged over about 20 hours... :)
[10:57] <KenMan> i think pSuccess counts all requests ever, so hourly success seems to be improving here
[10:59] <KenMan> the hourly volumes of incoming requests is bouncing up and down a bit more then it used to... this is also true when you plot queries/minute :o
[10:59] * Pascal (Pascal@c-24-13-17-191.client.comcast.net) Quit (Read error: 104 (Connection reset by peer))
[11:07] * Zorix (Brandon@fl-65-40-46-197.dyn.sprint-hsd.net) Quit (Read error: 110 (Connection timed out))
[11:11] * MikeW (Mike@159-134-233-14.as1.cbr.castlebar.eircom.net) has joined #freenet
[11:12] <MikeW> is stable 5085 basically unstable 60167?
[11:16] <KenMan> i believe so, the differences should be minor
[11:20] * Pascal (Pascal@c-24-13-17-191.client.comcast.net) has joined #Freenet
[11:24] * Sugadude (~Sugadude@darkridge.com) has joined #freenet
[11:32] * MikeW (Mike@159-134-233-14.as1.cbr.castlebar.eircom.net) Quit ()
[11:45] * TLF (francisco@174.Red-81-40-116.pooles.rima-tde.net) has joined #freenet
[11:52] * jay (jcl@ool-18bf6dac.dyn.optonline.net) has joined #freenet
[12:01] <toad_> hi ppl
[12:01] <toad_> hi jay
[12:01] <toad_> hi Pascal
[12:01] <jay> ello
[12:02] <jay> toad_: how do you trace a sig 11 error from the jvm?
[12:02] <jay> i actually got that once about a week ago and haven't seen it again
[12:03] <jay> and i saw a post on Dev i think about it
[12:05] <jay> oh yeah and we have to rewrite freenet in python
[12:06] <toad_> you can't debug the sun jvm
[12:06] <toad_> not easily
[12:07] <jay> im hardcore with gdb
[12:07] <jay> but i've never tried it
[12:07] <jay> if i ran the vm within gdb i could get the stacktrace i guess
[12:07] <jay> that sounds like it won't work
[12:09] <jay> it sucks because the node has crashed already by the time i see the sig 11 message in the log
[12:09] <toad_> well sure but then you'd have to get the source
[12:10] <toad_> and even if you can get the source you're probably not allowed to hack it
[12:10] <jay> i dont suppose there's a debug version of sun's linux vm?
[12:10] <iip_i2p> <gott> GOOD AFTERNOON TOAD_ !
[12:10] <toad_> hi gott
[12:11] <iip_i2p> <gott> IT IS AFTERNOON IN THE UK, NO ?
[12:11] <greycat> late afternoon, I'd imagine
[12:11] <toad_> early evening almost - 17:29
[12:11] <iip_i2p> <gott> I see !
[12:11] <iip_i2p> <gott> are you busy ? would you like to read something intriguing ?
[12:12] <jay> ARE WE NOW CHATTING IN ALL CAPS ???
[12:12] <jay> eheh
[12:12] <greycat> no, we are not.
[12:13] <toad_> jay: that's just gott
[12:13] <jay> IT'S SO ANNOYING
[12:13] <jay> greycat: OK
[12:13] <toad_> there are freesites about him (two??)
[12:13] <jay> greycat: heh xchat lowercased your nick
[12:13] <jay> toad_: i know gott.. the destroyer of worlds i believe
[12:13] <greycat> or of words, at any rate
[12:14] <jay> i defer to others
[12:14] <iip_i2p> <gott> toad ! you have not answered my question !
[12:15] <toad_> gott: what question? oh, yeah, fire away
[12:16] <jay> damn mirc codes
[12:16] <iip_i2p> <gott> http://en.wikipedia.org/wiki/Gimli_Glider
[12:17] <toad_> gott: what's that?
[12:17] <toad_> is it freenet-related?
[12:17] <jay> no it's aviation related
[12:18] <toad_> and tolkien related?
[12:18] <jay> doesn't appear to be
[12:18] <jay> u haven't clicked it yet :)
[12:20] <toad_> okay, I don't get it
[12:20] <toad_> where do we use NativeBigInteger.modPow() ?
[12:20] <toad_> we DONT use it in DLES !!
[12:20] <toad_> if we used it anywhere it would be in DLES
[12:21] <toad_> but we don't!
[12:22] <jay> [5:27] <iip_i2p> <Newsbyte> do you think my post on the devel maillist is too pessimistic?
[12:22] <jay> hmm.. yes
[12:22] <toad_> hehe
[12:22] <jay> the log says all
[12:22] <jay> [6:56] <iip_i2p> <Newsbyte> well, maybe I was a bit premature
[12:22] <jay> heh
[12:23] <iip_i2p> <gott> toad_: I think you should be more interested in aviation
[12:23] <iip_i2p> <gott> port freenet to work on aircrafts.
[12:23] * Hory (~Miranda@81.196.25.110) Quit (Read error: 54 (Connection reset by peer))
[12:25] <toad_> i think... i know... you are a troll
[12:25] <jay> trolls are so helpful for conversation
[12:26] <iip_i2p> <gott> toad: what is a troll ?
[12:26] <iip_i2p> <gott> I want to hear your definition first
[12:26] <jay> heh
[12:26] * jay looks at toad
[12:26] <KenMan> well, try to combine them to produce something useful. Let's see... most aircraft fly above the clouds. Consider his words as clouds, and just 'fly above them'
[12:27] * KenMan looks at NO ONE
[12:27] <jay> KenMan: oh go back to your l33t custom linux d1str0 ;)
[12:27] <KenMan> haha
[12:27] <jay> eheh
[12:27] * toad_ looks it up... woah
[12:27] <jay> Kenman Linux my a**
[12:28] <KenMan> i would, but KenMan linux doesn't seem to want me to use it. Keeps having network hiccups due to NAT and some screw-up somewhere
[12:28] <jay> KenMan: you should contact the Kenman Dev List
[12:28] <iip_i2p> <gott> toad: I am not your enemy
[12:28] <iip_i2p> <gott> anymore
[12:28] <jay> echo "problem" | mail localhost
[12:28] <jay> well user@lo
[12:29] <KenMan> I am considering just switching to fedora , and twisting it into what I want it to be.
[12:29] <jay> KenMan: go debian
[12:29] <iip_i2p> <gott> in fact, toad, I am here for you
[12:29] <KenMan> KenMan Linux has served me well for several years...
[12:29] <iip_i2p> <gott> jay: are you here for toad ?
[12:29] <jay> KenMan: hack the crap out of it
[12:29] * KenMan would probably just hack it into crap :o
[12:29] <jay> gott: im here for freenet in general
[12:30] <iip_i2p> <gott> jay: I think it is time to give toad a BIG hug
[12:31] <iip_i2p> <gott> *HUG* !!!!
[12:31] <KenMan> quit coming on to toad... it makes him uncomfortable !!
[12:31] <toad_> hrrm
[12:31] <toad_> why do we use NativeBigInteger some places and not others?
[12:31] <jay> frogs don't like to be hugged
[12:32] <iip_i2p> <gott> toad_: do you appreciate my hugs ?
[12:32] <jay> there's nude pictures of a frog on Frost
[12:32] <toad_> really? are there pictures of partially clothed frogs on Frost?
[12:33] <iip_i2p> <gott> there are pictures of partially clothed you on frost
[12:33] <iip_i2p> <gott> in fact, toad
[12:33] <iip_i2p> <gott> I think you should come to IIP at once
[12:35] <jay> what's with all the Oprah posts on Frost ?
[12:36] <iip_i2p> <gott> Let's discuss that on IIP
[12:36] <iip_i2p> <gott> IT IS OFF TOPIC HERE
[12:36] <KenMan> probably someone trying to compensate for the lack of women using freenet...
[12:38] <iip_i2p> <gott> tooooooooooooooooooooooooooaaaaaaaaaaaaaaaaaaaddddddddddddddddddddddddddddddddddddd
[12:39] <iip_i2p> <gott> toad is christian.
[12:39] <iip_i2p> <gott> toad is christian.
[12:39] * KenMan thinks gott is lonely on IIP, without any other participants
[12:39] <iip_i2p> <gott> I want my toad :(
[12:39] <jay> gott is ?
[12:40] <KenMan> hey, he is MY toad. You get near him and I'll crack your skull. Keep away from MY toad. (rowwwrr)
[12:40] <jay> The Destroyer of Worlds (TDW)
[12:40] <jay> To deceive is human. To destroy, divine.
[12:41] * toad_ writes to thelema, Pascal and Iakin demanding answers on whether we can use NativeBigInteger _everywhere_ in Fred
[12:41] <toad_> there's a lot of modPow()ing going on that isn't using the optimized code
[12:42] <iip_i2p> <gott> toad, why don't you use frost ?
[12:42] <iip_i2p> <gott> for that matter, why don't you do more freesite stuff
[12:42] <toad_> good question, it's a great Freenet stress tester
[12:42] <toad_> as far as freesites go, mostly because I regard my job as getting Freenet working
[12:43] <KenMan> i think iakin just profiled to find the best places for maximum improvement
[12:43] <toad_> KenMan: and DLES didn't show up? that seems rather peculiar...
[12:43] <iip_i2p> <gott> I would cherish frost posts with you
[12:43] <iip_i2p> <gott> discussion and conversation via frost
[12:43] * toad_ thinks Freemail would be useful
[12:43] <iip_i2p> <gott> freemail breaks on freenet last time I tried
[12:44] <toad_> unfortunately I can't run it on both networks simultaneously because I can't run two nodes constantly
[12:44] <toad_> also I'd have to hack it to deal with both networks
[12:44] <toad_> KenMan: you're sure he didn't say anything about it not working in some places, perhaps because of signedness issues?
[12:44] * Hory (~Miranda@81.196.25.110) has joined #FreeNet
[12:44] <KenMan> i don't remember now :(
[12:44] * TLF (francisco@174.Red-81-40-116.pooles.rima-tde.net) Quit ("http://www.vitaliano.esp.cc || Adi?s a todos y a todas y que les vaya bien.")
[12:44] <iip_i2p> <gott> we need your lifeblood.
[12:45] <iip_i2p> <gott> it's also the christian thing to do
[12:48] <toad_> oooh
[12:48] <toad_> here's an interesting one
[12:49] <toad_> crypt.Util.readMPI
[12:49] <toad_> aren't they always positive?
[12:49] <toad_> if so, why is it new NativeBigInteger(<bytes>) and not new NativeBigInteger(1, <bytes> ) ??
[12:50] <iip_i2p> <gott> toad is doing a good job
[12:50] <iip_i2p> <gott> despite being british
[12:51] * robilad[afk] is now known as robilad
[12:55] <toad_> hey, anyone on i2p know anything about DSA?
[12:56] <toad_> p, q, and r should ALWAYS be positive, right?
[12:56] <toad_> ditto on signatures, and everywhere else?
[12:58] <KenMan> toad: most definitely.
[12:59] <iip_i2p> <gott> I would like toad to know he is good.
[13:13] <KenMan> perhaps i do not understand how the sliding sectors should be working... i thought they were supposed to balance the number of reports in each.
[13:13] <KenMan> therefore, when doing testnet with prespecialized hockey pucks (remember the barcode graphic) , the dividing keys should cluster something like this :
[13:14] <KenMan> http://mywebpages.comcast.net/jkcorson/FreeNet/sectors.png - no ?
[13:14] <KenMan> but they don't seem to do that :(
[13:14] <iip_i2p> <gott> If you have more money in your pocket, you buy more things, which encourages more production. Consumer demand is stimulated by tax cuts. And the great thing about our society is that with consumer demand, there is generally somebody there to produce. And so, more jobs and more production takes place.
[13:15] <iip_i2p> <gott> God. Bless. The. United. States. Of. America.
[13:17] * BCoates (~coates@adsl-66-159-197-3.dslextreme.com) has joined #freenet
[13:17] <toad_> gott: and all her victims
[13:17] <BCoates> does anyone remember the name of the eternity-service like thingy where you exchange data with other people to store for fixed periods of time?
[13:19] <jay> ah consumerism
[13:20] <jay> the human rat race
[13:20] <jay> work harder, get more
[13:20] <KenMan> eat harder, weigh more ??
[13:20] <jay> dangle the prize on a string just out of reach
[13:22] <toad_> ummm
[13:22] <toad_> the global groups aren't NativeBigIntegers?!?!
[13:23] <toad_> surely THEY get modPow()ed?
[13:23] <toad_> BCoates: if you find out, tell me
[13:25] <jay> Los Alamos put 19 on leave in the inquiry on missing computer disks. The nuclear lab may be headed for a change in management.
[13:25] <jay> uh oh
[13:29] * Hory (~Miranda@81.196.25.110) Quit (Read error: 60 (Operation timed out))
[13:32] * tyesinclair (~tyesincla@82-41-215-112.cable.ubr12.edin.blueyonder.co.uk) has joined #freenet
[13:32] * tyesinclair (~tyesincla@82-41-215-112.cable.ubr12.edin.blueyonder.co.uk) has left #freenet
[13:35] * tyesinclair (~tyesincla@82-41-215-112.cable.ubr12.edin.blueyonder.co.uk) has joined #freenet
[13:37] * tyesinclair (~tyesincla@82-41-215-112.cable.ubr12.edin.blueyonder.co.uk) has left #freenet
[13:40] <BCoates> freehaven was what i was thinking of
[13:41] <KenMan> okay, so the next two tests for testnet go like this : 1) just scatter the keys randomly and see what success levels you get - don't expect too much emergence to occur 2) pre-specialize the nodes with a gaussian distribution model
[13:42] <KenMan> HTL=2 rather than 1 would probably be appropriate...
[13:46] * Hory (~Miranda@81.196.25.110) has joined #FreeNet
[13:47] <KenMan> testnet datastore barcodes : http://mywebpages.comcast.net/jkcorson/FreeNet/60167barcodes.png
[13:47] <KenMan> this is after ~50,000 requests...
[13:54] <toad_> KenMan: that's with prespecialization?
[13:54] <toad_> pretty serious prespecialization?
[13:54] <KenMan> that's with "hockey-puck" perfect specialization, yeah
[13:55] * toad_ restarts unstable node with just about everything using NativeBigInteger... lots of stuff that other people missed that does need it, and probably lots of stuff that doesn't, resulting in a slightly increased memory overhead :(
[13:55] <toad_> but probably the cpu gain outweighs that
[14:07] <KenMan> i think it may be a problem that NGR rolls time-to-service and likelihood-of-DF into one measurement...
[14:08] <KenMan> perhaps we could measure them separately, and combine the results into one number... it gives NGR more flexibility in whether it is trying hardest to find-fast or find-accurate
[14:09] <KenMan> the question is 'how well does typical-estimate reflect the reality?'
[14:11] <KenMan> is it too simplistic a measurement to be useful ??
[14:13] <toad_> woah
[14:13] <toad_> i just started unstable...
[14:14] <toad_> more or less constantly 50%+ idle
[14:14] <toad_> with an uptime of ~ 2 minutes !
[14:15] <KenMan> well, how many people out of 100 are actually using the network ??
[14:16] <KenMan> oh, you are proclaiming happiness with jbigi ... gotcha ;)
[14:16] <toad_> Number of known routing nodes 252
[14:16] <toad_> Number of node references 251
[14:16] <toad_> Number of newbie nodes 9
[14:16] <toad_> Number of uncontactable nodes 7
[14:16] <toad_> Contacted and attempted to contact node references 174
[14:16] <toad_> Contacted node references 34
[14:16] <toad_> Contacted newbie node references 9
[14:17] <KenMan> 250 unstable nodes that are 60150+ ??? wow!
[14:17] <KenMan> of course, 150 of those are probably RIAA / MPAA mole-nodes :o
[14:18] <toad_> KenMan: why bother?
[14:18] <toad_> they'd do unstable
[14:18] <KenMan> :)
[14:18] <toad_> err stable
[14:19] <toad_> 9000 of the 10,000 stable nodes are run by the RIAA, MPAA and NSA collectively
[14:19] <KenMan> shhh
[14:20] <toad_> I may be able to permanently run 2 nodes, one on each network, with THIS code...
[14:20] <KenMan> and now testnet nodes can crank 5x the volume through, in the same time period :)
[14:20] <toad_> memory usage isn't too bad either
[14:21] <hirvox> word, I've been toying with -mx128m for the last week
[14:22] <KenMan> it seems to me the greatest factor which determines mem usage is number of peers...
[14:22] <toad_> well I have 250 peers, and 58MB memory used according to Env
[14:22] <toad_> otoh, according to top: 23953 freenetu 30 15 771m 91m 65m S 3.0 6.0 0:02.57 java
[14:22] <KenMan> let it run for a day or a week then...
[14:22] <toad_> :)
[14:22] * piranha (~piranha@192.220.223.37) Quit (Read error: 113 (No route to host))
[14:22] <toad_> well if I can run both branches at once, I can run the stable node long-term
[14:23] * KenMan wishes toad good fortune
[14:23] * piranha (~piranha@192.220.223.39) has joined #freenet
[14:25] <toad_> SEC Jul 23, 2004 7:37:21 PM (freenet.support.io.NIOInputStream, YThread-7, NORMAL): waited more than 120000ms in NIOIS.read() tcp/connection: 38459>209.210.55.3:15151,freenet.transport.tcpConnection@1c0974:freenet.support.io.NIOInputStream@bbd886- closing
[14:25] <toad_> and another 5
[14:25] <toad_> what happened there?
[14:25] <toad_> hmm, they're not close together in time..
[14:25] <toad_> but even so, a little disturbing
[14:27] <hirvox> btw, when the available bandwidth drops quickly due to other usage, which is better from the node's point of view: a) aggressively begin dropping packats or b) queueing everything
[14:28] <KenMan> if we dropped max filesize to 256K, do you believe that would cause a proportional drop in MRIs ??
[14:28] <KenMan> dropping packets is bad... queueing packets is bad. Perhaps a balance of the two ?
[14:29] <KenMan> you are referring to outgoing bandwidth, i assume. So dropping is no good, queueing is approximated by bw limiting...
[14:30] <hirvox> I mean that the bandwidth available for the node might drop from 20kB/s to 1kB/s and rise back up later. currently, my node gets any and all unused bandwidth
[14:31] <KenMan> i guess the greatest impact is on existing transfers... so we would not want to take on (m)any new ones, if this was a protracted event
[14:32] <KenMan> the average user won't think in these terms, but how much netbytes must one provide in order to anonymously get X netbytes...
[14:33] <hirvox> for example, my uplink is currently sending 34 packets per second, and the backlog for freenet's traffic class is ~550 packets, so even if Freenet suddently got all the bandwidth, it would take ~15 seconds to clear out the backlog
[14:34] <KenMan> nasty
[14:36] <hirvox> I know, so it's really about selecting the lesser evil so that the node would slow itself down whenever bandwidth is scarce
[14:37] <hirvox> afk
[14:43] <toad_> hmmm
[14:43] <toad_> i wonder if we could detect the current BW requirement?
[14:51] * lolo-laptop (~lostlogic@68.251.84.226) Quit ("Client exiting")
[14:59] * TLF (francisco@174.Red-81-40-116.pooles.rima-tde.net) has joined #freenet
[15:15] <toad_> okay, any reason for me not to increase the default Fproxy request HTL to 20?
[15:17] <toad_> hmmm
[15:17] <toad_> are we sure that native FEC is working?
[15:21] * lostlogic (~lostlogic@node-402418b2.mdw.onnet.us.uu.net) Quit (sterling.freenode.net irc.freenode.net)
[15:21] * Gunnar (~gunnar@34.80-203-209.nextgentel.com) Quit (sterling.freenode.net irc.freenode.net)
[15:47] * piranha (~piranha@192.220.223.39) Quit (Read error: 60 (Operation timed out))
[15:49] * TLF (francisco@174.Red-81-40-116.pooles.rima-tde.net) Quit ("http://www.vitaliano.esp.cc || Adi?s a todos y a todas y que les vaya bien.")
[15:49] * TLF (francisco@174.Red-81-40-116.pooles.rima-tde.net) has joined #freenet
[15:51] * toad_ learns to use tcpdump to traffic analyze his own node
[15:52] * ixitxachitl_ (dn32418@d66-183-170-199.bchsia.telus.net) Quit (Remote closed the connection)
[15:52] * jay (jcl@ool-18bf6dac.dyn.optonline.net) Quit ("Client exiting")
[16:00] * piranha (~piranha@192.220.223.25) has joined #freenet
[16:05] * BCoates (~coates@adsl-66-159-197-3.dslextreme.com) Quit (""gone for a week"")
[16:10] * lostlogic (~lostlogic@node-402418b2.mdw.onnet.us.uu.net) has joined #freenet
[16:10] * Gunnar (~gunnar@34.80-203-209.nextgentel.com) has joined #freenet
[16:11] * greycat (~wooledg@192.35.79.70) Quit ("Lick Bush in '04")
[16:14] * Hory (~Miranda@81.196.25.110) Quit ("Miranda IM! Smaller, Faster, Easier. http://miranda-im.org")
[16:31] * Hory (~Miranda@81.196.25.110) has joined #FreeNet
[16:32] * Hory (~Miranda@81.196.25.110) Quit (Remote closed the connection)
[16:33] <toad_> woah
[16:33] <toad_> I'm pretty sure an attacker can identify what message types are being transferred
[16:33] <toad_> sometimes he can see the source of trailerpacketmessages by their unique lengths
[16:34] <toad_> and he can also use the UniqueID's to control packet lengths and therefore attempt to track a state chain across the network
[16:34] <toad_> bbiab :)
[16:34] <toad_> have a nice day ;)
[16:34] <toad_> bbiab, seriously
[16:50] <ShaunMacPherson> hi
[16:50] <ShaunMacPherson> thats no good
[16:50] <ShaunMacPherson> :>
[16:54] <toad_> :)
[16:55] <toad_> okay, here's the major messages:
[16:55] <toad_> DataNotFound: ~ 94-104 bytes
[16:55] <toad_> Accepted: ~ 74-83 bytes
[16:56] <toad_> well
[16:57] <toad_> first, we remove RequestInterval from FNP messages
[16:57] <toad_> well no not first
[16:57] <toad_> first we pad
[16:57] <toad_> later we remove RequestInterval from FNP messages
[16:57] <iip_i2p> <gott> toad
[16:57] <toad_> the question is how much to pad?
[16:58] <toad_> hmm
[16:58] <iip_i2p> <gott> did you ever see my interview with CofE ?
[16:58] <toad_> this WOULD be easier without RequestInterval.. ok
[16:58] * jay (jcl@ool-18bf6dac.dyn.optonline.net) has joined #freenet
[16:59] <toad_> okay, how do we fix that?
[16:59] <toad_> we remove RequestInterval from FNP messages
[16:59] <iip_i2p> <gott> he said somethings that you might want to take into consideration
[17:00] <toad_> and then we make sure outgoing packets always have an MRIMessage on them
[17:00] <toad_> gott: I'm trying to give the NSA a headache, would you mind STFUing for a few hours?
[17:00] <iip_i2p> <gott> toad: really, toad, you have to trust me. I'm not trying to troll with you.
[17:01] * Hory (~Miranda@81.196.25.110) has joined #FreeNet
[17:02] * Hory (~Miranda@81.196.25.110) Quit (Read error: 104 (Connection reset by peer))
[17:02] <iip_i2p> <gott> until you start to trust me, we can never go above the bond of hatred which feels rather unfortunate for me
[17:02] <iip_i2p> <gott> not to mention you are doing a poor job of being a christian
[17:03] <iip_i2p> <gott> perhaps you will trust me one of these days, but until then, you are very stubborn
[17:03] <iip_i2p> <gott> *hug*
[17:03] * toad_ ignores iip_i2p
[17:03] * Hory (~Miranda@81.196.25.110) has joined #FreeNet
[17:03] <toad_> sorry to everyone else on #freenet on the other side
[17:05] <toad_> okay, RequestInterval is 17+8 to 17 = 23 to 34 bytes of every FNP message
[17:06] <toad_> therefore if we can get rid of it, we're looking at:
[17:06] <toad_> Accepted: ~~ 48 bytes
[17:06] <toad_> QueryRejected: ~~ 100 bytes
[17:07] <toad_> QueryRestarted: ~~ 64 bytes
[17:07] <toad_> DataNotFound: ~~ 76 bytes
[17:07] <toad_> DataReply: ~~ 300 bytes (!!)
[17:07] <toad_> InsertRequest: ~~ 120 bytes
[17:07] <toad_> StoreData: ~~ 34kB (!)
[17:08] <toad_> any observant individuals here will notice that you can easily identify packets with such sizes :|
[17:08] <toad_> so what would the appropriate padded sizes be?
[17:08] <toad_> I don't think we can do much with the StoreData :|
[17:09] <toad_> after 1kB we're looking at padding to a whole kilobyte
[17:09] <toad_> but we can pad to less than that below it
[17:09] <toad_> the above numbers suggest 128 bytes might be cutting it a little bit too fine
[17:10] <toad_> how about padding up to the next 256 byte boundary?
[17:10] <toad_> slightly wasteful...
[17:10] <toad_> depending on how many messages are involved...
[17:10] * toad_ thinks he'll do it that way anyway for now
[17:11] <toad_> now, how to implement? PeerPacket needs to check its length in the constructor and add a void at the end to bring it up appropriately if needed
[17:12] * ysothoth (~ysothoth@68-174-22-46.si.rr.com) Quit ("using sirc version 2.211+KSIRC/1.3.10")
[17:17] * ysothoth (~ysothoth@68-174-22-46.si.rr.com) has joined #freenet
[17:20] <Pascal> have the packet types all been switched to binary instead of text yet?
[17:21] <Pascal> <toad_> i wonder if we could detect the current BW requirement?
[17:22] <Pascal> isn't that what messageSendTimeRequest is for?
[17:23] <Pascal> when doing padding, it would be best to do it in the multiplexer rather than the individual packets
[17:23] <toad_> Pascal: of course
[17:23] <toad_> that's what I'm doing
[17:23] <toad_> the multiplexer being PeerPacket
[17:24] <toad_> which packs a load of messages into a single transit packet
[17:24] <Pascal> padding to next log2 would make sense
[17:25] <Pascal> though watch MSS
[17:26] <toad_> well it's not a panacea
[17:26] <toad_> some of our packet sizes straddle the boundary
[17:26] <Pascal> you will have to detect MSS for padding to avoid fragmentation
[17:31] <toad_> hmmm?
[17:31] <toad_> you mean MTU?
[17:31] <toad_> I'm just going to round up to the next multiple of 256 bytes for now
[17:32] <Pascal> MTU is the packet including IP headers. MSS is without
[17:33] <Pascal> IIRC for IPv4 you have 40bytes of headers
[17:33] <Pascal> so default MSS is 1460
[17:35] <toad_> okay, that's the field bandage
[17:35] <toad_> lets see if it compiles
[17:35] * toad_ plans to improve on it, but lets get something out first
[17:36] * sylvinus (~sylvinus@ca-metz-4-201.w80-8.abo.wanadoo.fr) has joined #freenet
[17:37] <sylvinus> hello all
[17:37] <toad_> hi
[17:37] <sylvinus> do you guys know about fairshare ?
[17:37] <toad_> not really
[17:37] <toad_> i remember there was a freesite
[17:37] <sylvinus> http://freenetproject.org/index.php?page=fairshare
[17:38] <toad_> ah, THAT fairShare
[17:38] <sylvinus> I'm implementing it.
[17:39] <sylvinus> Is Ian clarcke or Steven Star or Rob Kramer here ?
[17:39] <toad_> Ian MIGHT be here
[17:39] <toad_> try privmsg'ing him, he's sanity
[17:40] <sylvinus> thanks
[17:41] <sylvinus> I would appreciate to have your feedback on my implementation
[17:41] <sylvinus> http://www.peermajor.org/index.php?p=static&s=peerlabel
[17:41] * TLF (francisco@174.Red-81-40-116.pooles.rima-tde.net) Quit ("http://www.vitaliano.esp.cc || Adi?s a todos y a todas y que les vaya bien.")
[17:42] * toad_ is just testing the changes before committing...
[17:47] <Pascal> my inbox got corrupted :(
[17:47] <Pascal> anyone know of a repair tool for netscape?
[17:50] <Iakin> re: jcpuid should be jcpuid-x86..
[17:50] <Iakin> Per definition CPUID is x86-only
[17:52] <Iakin> toad: you know that you have update a jar with gzip, dont you?
[17:52] <Pascal> the CPUID instruction is x86 only but programs for IDing a CPU exist for most (all?) procs
[17:55] <Iakin> toad: the gain factor for native doubleValue() is in practice even better than those 3x...
[17:56] <Iakin> due to the elimination of that lock contention situation
[17:56] <Iakin> (this only shows up when multithreading though)
[17:57] <Iakin> ll: given my last commits you have to rename your jcpuid.so to libjcpuid-linux.se
[17:57] <Iakin> ... or well.. .so
[17:58] <Pascal> freenet-ext.jar should now build fine on windows too. The extra cpuInformation is now gone.
[17:59] * Hory (~Miranda@81.196.25.110) Quit ("Miranda IM! Smaller, Faster, Easier. http://miranda-im.org")
[18:01] <Iakin> Heh.. I thought toad/thelema had done the DLES NativeBigInteger stuff when they initially worked on that class
[18:02] <toad_> <Iakin> toad: you know that you have update a jar with gzip, dont you?
[18:02] <toad_> eh?
[18:05] * Sugadude (~Sugadude@darkridge.com) Quit ()
[18:05] <Iakin> re. you rebuilding the jar when stuff accidentially appeared in /lib/XXXX
[18:06] <Iakin> So... have I destroyed your freenet-ext.jar now that I changed the name of jcpuid.so to jcpuid-linux.so
[18:08] <toad_> hmm?
[18:09] <toad_> just fix the codee.
[18:09] <toad_> anyway it should be jcpuid-linux-x86.so
[18:11] <Iakin3> yup, I just read that..
[18:11] <Iakin3> I'll fix
[18:17] * Ash-Fox (Hal-9000@aad77.neoplus.adsl.tpnet.pl) Quit ("Leaving")
[18:19] * toad_ commits 60168
[18:21] * Ash-Fox (Hal-9000@aad77.neoplus.adsl.tpnet.pl) has joined #FREENET
[18:28] * toad_ changes topic to 'Upgrade to 5085 | Unstable: Upgrade to 60168 (60160: network level IP detection should work now); network reset at 60151 so 60151 is mandatory | Unstable nodes must be updated daily | Stable nodes should generally be updated weekly | #freenet-chat and #freenet-politics are open for business! Off topic discussions may be directed there. | Logs: http://newton.matcmp.ncc.edu/~lockej/freenet/chanlog/'
[18:28] <toad_> hmmm
[18:28] <toad_> I wonder if we'll get slashdotted... my announce mail was a little alarmist...
[18:35] * toad_ compiles up merged stable with jbigi and the padding
[18:39] * Iakin3 committed the updated names for the jcpuid dll's/so's
[18:40] <toad_> hmmm
[18:40] <Iakin3> (so you might consider updating your source :) )
[18:40] <toad_> I was trying to get the native stuff into 5086...
[18:41] <Iakin3> Sounds like a good idea
[18:41] <Iakin3> But you need to remerge CPUID.java...
[18:42] <toad_> and regenerate and reupload freenet-ext.jar
[18:42] <toad_> unless it's compatible with the old name as well?
[18:44] <toad_> half of me wants to ditch the native stuff and just merge the padding...
[18:44] <Pascal> since both stable and unstable share a common freenet-ext.jar, when you regen it with the new names you will have to do a new release of both.
[18:45] <toad_> yeah, Iakin3 has just done the latter
[18:45] <toad_> for unstable that is
[18:46] <Iakin3> [01:00] <toad_> and regenerate and reupload freenet-ext.jar
[18:46] <Iakin3> Right
[18:46] <toad_> ERROR: Resource name [freenet/support/CPUInformation/libjcpuid-x86-linux.so] was
[18:46] <toad_> not found
[18:46] <toad_> hmmm
[18:46] <toad_> Iakin3: did you move the files?
[18:46] <Iakin3> just modify the jar.. no need to regen it
[18:46] <toad_> in Contrib?
[18:46] <toad_> you didn't, did you?!
[18:46] <Iakin3> I renamed them
[18:46] <toad_> grrrrrr
[18:47] <toad_> ok thanks
[18:47] <Iakin3> ?
[18:47] <toad_> aieeeeee
[18:47] <toad_> AIEEEEEEEEEEEEEEEEe
[18:47] <toad_> I can't get to CVS now
[18:47] <Iakin3> Possibly build.xml needs a touch
[18:47] <Pascal> make sure you do a distclean so freenet-ext.jar gets rebuilt too
[18:47] <toad_> The DSA host key for cvs.freenet.sf.net has changed,
[18:47] <toad_> and the key for the according IP address 66.35.250.209
[18:47] <toad_> is unchanged. This could either mean that
[18:47] <toad_> DNS SPOOFING is happening or the IP address for the host
[18:47] <toad_> and its host key have changed at the same time.
[18:47] <toad_> Offending key for IP in /root/.ssh/known_hosts:7
[18:48] <Pascal> Iakin3: no, it looks at dirs not files
[18:48] <Iakin3> I have never used ant..
[18:48] <toad_> now what?!
[18:48] <Iakin3> I say again.. just update the jar instead of rebuilding it
[18:48] <toad_> Iakin3: irrelevant
[18:48] <toad_> I WANT TO RELEASE A STABLE BUILD
[18:48] <toad_> how can I do that if sf fucks us around like this?
[18:49] <Iakin3> Ahhh..
[18:49] <toad_> and it works from dodo...
[18:49] <toad_> something seriously wierd is going on here
[18:50] <toad_> especially as I have been working on Contrib TODAY
[18:50] <Pascal> is there a way I can run frost on my linux box and display the frontend on windows?
[18:52] <mikeDOTd> Pascal: you'll need an X11 implementation for windows
[18:52] * jay (jcl@ool-18bf6dac.dyn.optonline.net) Quit ("Client exiting")
[18:52] <Iakin3> like hummingbird xceed
[18:52] <mikeDOTd> exceed is a commercial offering. cygwin is open source, as is weird-x
[18:52] <Pascal> thanks
[18:52] <Iakin3> or xwin32
[18:53] <Iakin3> or ?stardocks?
[18:53] <Pascal> toad: is your padding backward compatible?
[18:54] <toad_> Pascal: yes
[18:55] <Pascal> cool
[18:58] * toad_ waits...
[18:58] <toad_> it can take 10 seconds to open a file here when the system seems otherwise responsive
[18:58] <toad_> something wierd with the kernel...
[18:59] <Pascal> how are you set for ram
[19:00] <toad_> oh come on, I'm not falling for THAT one...
[19:00] <toad_> not only have they changed their key, but they now don't recognize my key
[19:00] <toad_> forcing me to enter my password
[19:00] <toad_> yeah, I'm really that stupid, come on guys...
[19:00] <toad_> this is absurd..
[19:01] <toad_> we are dependant on idiots at sourceforge
[19:01] <toad_> and when they fuck up we think we're being MITM'd
[19:01] <toad_> as if anyone capable of MITM'ing us would bother to.. they'd just root our machines, right?
[19:01] <toad_> actually DNS spoofing is easier than MITM...
[19:04] <Iakin3> Hmm.. so, the top-end nvidia graphics cards coming out this fall will feature 512 megs of memory.. ridiculous I say
[19:04] * toad_ heard that some other cards will have a gig
[19:05] <toad_> okay, until we get this CVS mess sorted out...
[19:05] <toad_> I'm going to ditch the nativebigint changes and just put out 5086 with the security fix
[19:05] <toad_> which is what I should have done in the first place
[19:06] <KenMan> cvs.sf.net with SSH working fine here :(
[19:06] <toad_> KenMan: you're scaring me
[19:07] <mazzanet> !
[19:07] <toad_> do you have unusually permissive cvs options?
[19:07] <mazzanet> why are you sshing into a semi-non-sshable server?
[19:07] <toad_> in any case, we are going to get our own version control system, and that's it
[19:07] <KenMan> dig cvs.sf.net gives me 66.35.250.207
[19:08] <mazzanet> toad_: so what's bodgered up with cvs?
[19:08] <toad_> hmmm
[19:08] <toad_> now Eclipse is playing up
[19:08] <toad_> it's managed to delete PeerPacket.java
[19:08] <toad_> and Storables
[19:08] <toad_> shit
[19:09] * KenMan thinks that witch-cult who worships Stonehenge is to blame for hexing the toad
[19:09] <mazzanet> speaking of toads....
[19:09] <toad_> yikes
[19:09] <mazzanet> did you know you can whack a toad in a freezer and freeze it
[19:09] <toad_> this is nuts
[19:10] <mazzanet> come back a few days later, thaw it and it'll be still alive
[19:10] <toad_> yeah, either I'm hexed or I'm making real progress with freenet and They are trying to discourage me
[19:10] <mazzanet> and hop around doing random stuff that toads do
[19:10] * toad_ wonders if Eclipse just gave away my password...
[19:11] <toad_> doesn't matter, we're getting our own vc server soon...
[19:12] <KenMan> mazzanet - that's what it means to be cold-blooded ...
[19:12] * ysothoth (~ysothoth@68-174-22-46.si.rr.com) Quit ("using sirc version 2.211+KSIRC/1.3.10")
[19:12] * ysothoth (~ysothoth@68-174-22-46.si.rr.com) has joined #freenet
[19:15] <mazzanet> so whats this about cvs being broken?
[19:16] <mikeDOTd> toad: is there a pre-built freenet-ext.jar that i could use with stable?
[19:16] <mikeDOTd> (with the snazzy optimizations)
[19:17] <toad_> mikeDOTd: no
[19:17] <toad_> because that code isn't in stable yet
[19:18] <toad_> however, you will need to upgrade to 5086
[19:18] <toad_> 5087 may have the native optimizations
[19:19] <toad_> 5086 will have a MAJOR bugfix in it
[19:19] <toad_> and by bug I mean vulnerability
[19:19] <toad_> so it won't Fix The Network
[19:19] <toad_> but it will make it a LOT harder to do that attack
[19:21] <toad_> okay, merged stable node starting...
[19:36] <toad_> okay, bbl zzz
[19:44] <KenMan> night toad_
[19:49] * piranha (~piranha@192.220.223.25) Quit (Read error: 60 (Operation timed out))
[19:53] <Pascal> how do I extract a zip on linux?
[19:53] <nicktastic> unzip
[19:54] <Pascal> cool. thanks
[19:55] * mazzanet slaps KenMan around a bit with a frozen toad
[19:55] <KenMan> ouch !
[19:56] <mazzanet> :P
[20:03] * piranha (~piranha@192.220.223.29) has joined #freenet
[20:09] <Pascal> in the command "java -jar frost.jar "$@"" what does "$@" do?
[20:13] <mazzanet> any command line arugments?
[20:14] <Pascal> oh, it passes the command line arguments passed to the .sh into the jar? that makes sense. thanks.
[20:15] * JustMe_ (JustMe_@cs66253-140.houston.rr.com) has joined #freenet
[20:16] * JustMe_ (JustMe_@cs66253-140.houston.rr.com) Quit (Client Quit)
[20:18] <Pascal> toad: looks like a problem with the padding
[20:19] * Ash-Fox (Hal-9000@aad77.neoplus.adsl.tpnet.pl) Quit (Remote closed the connection)
[20:19] <Pascal> toad: Caught java.lang.NullPointerException notifying freenet.MuxConnect
[20:19] <Pascal> ionHandler
[20:20] <toad_> Pascal: stack trace?
[20:21] <Pascal> http://sial.org/pbot/3625
[20:22] <toad_> strangely enough that doesn't happen here
[20:23] <toad_> hmmm
[20:23] <toad_> L60 is "Core.logger.log(this, "Didn't need to pad "+this, Logger.NORMAL);"
[20:23] <toad_> I don't see how that could cause an NPE
[20:23] <toad_> will look at it tomorrow
[20:23] <toad_> but I suspect it's something on your end
[20:25] <Pascal> the NPE was on line 24
[20:25] <Pascal> prob messages.length
[20:28] * Ash-Fox (Hal-9000@aad77.neoplus.adsl.tpnet.pl) has joined #FREENET
[20:32] <Pascal> toad: though why messages would be null I have no clue. fixed now.
[20:32] <toad_> Pascal: fixed how?
[20:33] <toad_> messages was null?
[20:33] <toad_> not some element of messages?
[20:33] <toad_> strange
[20:33] <toad_> bbl zzz
[20:34] <Pascal> well if indeed messages was null then fixed. it is the only thing on line 24 that could have caused a NPE.
[20:36] * guido^pe (~unknown@dsl-082-082-158-159.arcor-ip.net) Quit (Remote closed the connection)
[20:57] <toad_> Pascal: move the log down to after the assignment
[20:57] <toad_> and it will fix it
[20:59] <Pascal> hahahahhaah
[20:59] <Pascal> ya, that would do it :)
[21:00] <Pascal> I never even looked at L60 :)
[21:00] <Pascal> obviously neither did you :)
[21:01] <Pascal> committed. thanks.
[21:03] <Pascal> toad: I wonder if L66-L72 are also in the wrong order
[21:13] * Ash-Fox (Hal-9000@aad77.neoplus.adsl.tpnet.pl) Quit (Remote closed the connection)
[21:14] * Ash-Fox (Hal-9000@aad77.neoplus.adsl.tpnet.pl) has joined #FREENET
[21:47] * Ash-Fox reads the huge... almost flamewar like stuff... of kde users vs gnome users comparing the GUI
[21:47] <Ash-Fox> on slashdot
[21:51] <Ash-Fox> It's kind of funny... the gnome users are terrified of KDE, because it's default theme settings are ugly, and looks a little bloated... some of them get so angry that some apps have the letter K in the middle of them instead of C, that they won't use them (Couldn't they edit the menu name of the program)
[21:53] <Ash-Fox> Lol... someone commented that it makes the press releases look like it was written by script kiddies (because of the K thing)
[23:19] * robilad is now known as robilad[afk]
[23:26] <lostlogic> why is freenet stable using like 100% cpu to do nothing? I thought we were past that?
[23:31] <lostlogic> why on earth are 35 threads opening connection?
[23:31] * lostlogic mutters profanity
[23:41] * Gunnar (~gunnar@34.80-203-209.nextgentel.com) Quit (Remote closed the connection)
[23:51] <mikeDOTd> RNF: 58 backed off nodes, yay!
[23:51] * piranha (~piranha@192.220.223.29) Quit (Operation timed out)
[23:53] * TLF (francisco@137.Red-81-40-113.pooles.rima-tde.net) has joined #freenet
[23:59] * TLF (francisco@137.Red-81-40-113.pooles.rima-tde.net) Quit ("http://www.vitaliano.esp.cc || Adi?s a todos y a todas y que les vaya bien.")
These logs were automatically created by Jay Oliveri with his gimp hapi on irc.freenode.net.
