Timestamps are in GMT/BST.
[0:03] * MrNaughty (MrNaughty@d198-166-55-147.abhsia.telus.net) Quit ("\(^_^)/' No Soliciting!!! Unless you have legs way, way up and really, really big tits....")
[0:03] * shendaras (~shendaras@ip68-97-113-208.ok.ok.cox.net) Quit (Read error: 104 (Connection reset by peer))
[0:11] * robilad (~topic@mpiat2304.ag2.mpi-sb.mpg.de) has joined #freenet
[0:22] * KenMan (~chaziller@pcp403292pcs.mntcrm01.md.comcast.net) has joined #freenet
[0:26] <Seed212> kenman: did that file ever download?
[0:26] <KenMan> which one ??
[0:26] <Seed212> gnu
[0:26] <KenMan> gnu.ogg , yeah. It took about 30 minutes though...
[0:26] * Seed212 is now known as wind789
[0:27] <wind789> there we go, my other identity hehe
[0:27] * KenMan is now known as frog_
[0:27] <frog_> heehee, got to keep things interesting around here ;)
[0:27] <wind789> lol
[0:27] <frog_> How does stable treat you today/night... ?
[0:28] <wind789> not sure, I havn't been messing with it much
[0:28] <wind789> I just turned my node on
[0:28] <frog_> i forget , you use windows or linux ?
[0:28] <wind789> can't play games with freenet on....
[0:28] <wind789> windows
[0:29] <frog_> ah, okay. I am trying to get my linux system under control - I caught it being very naughty, and have been trying to reprimand the penguin.
[0:29] <wind789> one day I'll break away from the M$ mindcontrol
[0:29] <wind789> lol
[0:30] <frog_> everything used to be so simple. For many applications, there only was 1 choice. Now things are so mixed up and confusing, it is hard to keep track.
[0:30] <frog_> this works with that, but NO , not THOSE. And just the kernel (proper) has so many darned new features, it makes my head spin.
[0:30] <wind789> darn
[0:31] <frog_> and to top it all off, I rarely use windows XP, so I have gotten rather out of touch on that side of the fence :(
[0:31] <wind789> if I had a second computer I'd have linux for sure
[0:31] <wind789> as it is I don't feel like dual botting at the moment
[0:31] <frog_> yes, i find it easier to segregate them that way. But lately I'm using the other box for linux and Freenet testing...
[0:32] * MrNaughty (MrNaughty@d199-126-215-93.abhsia.telus.net) has joined #freenet
[0:32] <wind789> aahh
[0:32] <frog_> there is very little that I *need* windows for. Action games mainly... and I'm not a big video game freak, so...
[0:33] <wind789> the following files are avalible on stable:
[0:33] <wind789> CHK@bL35x8zPm6Arup2hbeMoD1sRjygTAwI,uaQeOnu1lA0SQV4hROhnpg/Richard_Stallman-Free_Software.mp3
[0:33] <wind789> CHK@EtRYbTd71THRf-ElPO-BW4uFCRQMAwI,d3KCuwzSgC02n8~SwqNJ2Q/gnu.ogg
[0:33] <wind789> CHK@FwDwUWZKHDd3IeAzR95gCSifKVwMAwI,URWySMGDJK2BUbFK4s1JkQ/Richard_Stallman-Free_Software(remix).mp3
[0:33] <wind789> CHK@UGqWl--XXSk7g1lFTUWXD9hLsvgLAwI,A5I3Vy4JaAbcLvlJ2pu3qQ/Fenster-Free-Software-Song.ogg
[0:33] <wind789> I play some games myself, mostly tribes and a little startcraft at the moment
[0:34] <wind789> an open gl game called erternal lands works on both
[0:35] <frog_> my node is too backed off to use right now :( It is having teething troubles, but soon it should come under control (?) . This is the normal 'life-spiral' (strikingly similar to a death spiral) of FreeNet, upon start up.
[0:35] <wind789> how long does it take to correct?
[0:35] <frog_> I play 'Battlefield 1942 and Battlefield Vietnam' mainly, but not too often.
[0:36] <frog_> Sometimes a few hours, sometimes more than a day...
[0:36] <wind789> sheesh
[0:36] <frog_> yeah, well, there are 'issues' to be worked out. Lots and lots of 'issues' .
[0:37] <wind789> what page do you check backoff status on? Node Status Interface?
[0:37] <frog_> The routing table page, it is about the 10th line down.
[0:38] <wind789> Backed off nodes 24 , isn't that for other peoples nodes?
[0:38] <frog_> yes, Node Status Interface, then Node Reference Status
[0:38] <frog_> well, they are YOUR routes :(
[0:38] <wind789> hehe
[0:38] <frog_> in fact, it really means, 24 nodes have backed YOU off...
[0:38] <wind789> lol
[0:39] <frog_> things were real bad with 5087, fortunately today we have a 5088 to solve all the troubles and make Freenet really zoom !!
[0:39] <frog_> or at least make them a little better, anyway.
[0:39] <wind789> what's a newbie node ?
[0:40] <wind789> is that just a node that is new to me?
[0:40] <frog_> yes
[0:41] <frog_> they are given special rights and powers, for an initial period of time.
[0:41] <frog_> it is supposed to be like a trial period. Then they either make the cut, or they don't ...
[0:42] <wind789> lol cool
[0:43] <frog_> node management is more important than most people realized, until very recently.
[0:43] <frog_> and that is a big part of what NGR is supposed to be all about
[0:44] <wind789> I'm sure you've heard the saying " the simplest solution is usaly the best one"
[0:45] <wind789> when freenet 1.0 comes how I wonder how much more complicated or simplified it will be
[0:45] <frog_> not said very often around here :o
[0:45] <wind789> hehe
[0:45] <frog_> I shudder to imagine such a thing. Just 0.6 must have some lofty goals.
[0:46] <wind789> are they published some place?
[0:46] <frog_> but what is in a name (other than 242K new lines of code) ? a jar by any other name is still...
[0:47] <frog_> um, i think toad has set some goals in several posts. But mainly he carries it around in his head, like so many other things.
[0:47] <wind789> aahh hehe
[0:47] <frog_> i think the last one was "we must make routing work first" - which might be fair !!
[0:48] <frog_> but we are rather far from doing that, i think.
[0:50] <wind789> does routing simpy refer to choosing the correct node to send a requst to?
[0:51] <frog_> that's not been specified. Basically, yes. Rate limiting is a major impediment to routing, even *if* NGR was perfect.
[0:51] <frog_> so, once we see rate limiting get out of the way, it mainly comes down to success rates.
[0:52] <wind789> rate is the number of request over a period of time?
[0:53] <frog_> yes. NGR takes a single request, and orders the routes from most likely to worst. Then rate limiting prevents the request from going to any but the 2nd or 3rd worst... :(
[0:53] <frog_> the reason is that too many requests are being allowed to enter the network at any given moment.
[0:54] <frog_> the receiving end of a route says how fast a requestor can talk, and it varies that rate quite frequently.
[0:54] <wind789> is rate measured in byte per time unit or request?
[0:55] <TheSeeker> can someone try http://theseeker.bounceme.net:8888/ ?
[0:55] <frog_> there are still bottlenecks everywhere, but toad is currently focused on that problem, i think. Rate of requests / unit of time.
[0:56] <TheSeeker> oops *updates his dyndns*
[0:57] <frog_> in fact, toad implements it as individual intervals , not a rate, but they are of course the same meaning. Except that a rate has multiple samples or events, whereas intervals can be treated as distinctly independent units :o
[0:57] <wind789> I see
[0:57] <wind789> is there any way for me to find the tranfer rate of a single transer?
[0:58] <frog_> sort of... look for a diagnostic called successTransferRate, and view the raw data
[0:58] <Iakin> [01:23] <lolo-laptop> io.ch.SelectionKeyImpl@15b8c38 - AND IT WROTE SUCCESSFULLY! This can happen occasionally but if it happens a lot, it could indicate a broken NIO implementation
[0:58] <frog_> you can't associate it with a particular transfer, but you can see the individual rates.
[0:59] * frog_ is really FrogMan ...
[0:59] <frog_> hi Iakin !
[0:59] <Iakin> Toad, you should also add an 'unwrite' call and then see if you get that key with the next select call.. :)
[1:02] * salahx (salahx@sc1-24.217.174.147.charter-stl.com) has joined #freenet
[1:03] <Iakin> hello frog
[1:03] <frog_> are you running 5088 yet ??
[1:04] <wind789> is it possible to manual set the rate limiting for your own node?
[1:04] <frog_> wind789 - hell no !
[1:04] <wind789> hehe
[1:05] <frog_> wait, you can set your bandwidth limits, but you can't really control your query rate(s) ...
[1:05] <salahx> lok like 60175 isn't in the snapshots yet
[1:06] <wind789> so there no way for me to get fewer simultaneous transfer that each use more bandwith?
[1:06] <salahx> no
[1:06] <salahx> you woudlnt WANT that
[1:06] <salahx> the nubmer of tranfers is unlimeted. It has to be that way
[1:06] * frog_ wonders why stable take sooo long to get some peers... 2.5 hours only 32 peers (31 backed off)
[1:07] <wind789> doens't that really thin out the bandwidth?
[1:07] <salahx> it can
[1:07] <salahx> but if the nubmer of trnafers is limited, its an easy low-bandwidth DoS
[1:07] <wind789> I see
[1:08] <frog_> we could detect and abort on piss-poor-performance / DoS , but that is extra work and logic...
[1:08] <salahx> yeah
[1:08] <frog_> but, in theory, it could be done (guaranteed minimum xfer rates)
[1:10] <frog_> well... not until Freenet 0.9 , slated for a 2008 release
[1:10] <wind789> hehe
[1:10] <salahx> YOur such an optimist frog_
[1:10] <frog_> :o lol
[1:15] <wind789> have there been any surveys on the differnt preformance levels of differnt htls?
[1:17] <wind789> for instance what is the liklyhood of a htl 20 request actualy making it though 20 nodes sucessfuly?
[1:18] <wind789> will it take about 20 times longer than htl 1 or 100 times longer?
[1:18] * frog_ leaps up on his lilypad to cover his eyes for a few hours... zzz
[1:18] <wind789> cya
[1:19] <salahx> ok
[1:20] <salahx> well tehre an easy way to check that
[1:20] <salahx> if the request can make it the end of the HTL, you get an RNF
[1:20] <salahx> otherwise, you either get teh data or a DNF
[1:20] <salahx> Note that the HTL is fudged, so ti works otu to more the 25 hops
[1:21] <salahx> if bleeive there is jsut a meaturement on the diagnotic pages...
[1:22] <wind789> cool
[1:23] <wind789> if the request can make it the end of the HTL, you get an RNF? I though it was the other way around
[1:23] <salahx> no
[1:24] <salahx> when HTL reach 0 and the data is not found, ou get a DNG
[1:24] <salahx> err DNF
[1:24] <salahx> I think that stat you;re looking for is routingFailRNFRatio
[1:24] <wind789> thanks
[1:26] <wind789> "The number of requests that failed by routing" what would that be? DNF?
[1:27] <salahx> i think so
[1:27] <salahx> I',m not sure if DNF is consieres a routing failure or not
[1:28] <wind789> hehe
[1:29] <wind789> 25435 18901 0.7431098879496757
[1:29] <wind789> does 74% sound decent?
[1:31] <salahx> not sure
[1:32] <wind789> cool, well I'm off to bed, thanks for answering my newbie questions. ttyl
[1:34] <salahx> ok
[1:57] <mazzanet> salahx
[1:58] <mazzanet> why can't i compile latest unstable from cvs?
[1:58] <salahx> ?
[1:58] <salahx> i udunno, I ahve'nt tries it yet
[1:58] <mazzanet> mega compile errors
[1:58] <mazzanet> all with "NativeBigInteger"
[2:01] <salahx> works here
[2:01] <salahx> try this:
[2:01] <mazzanet> it cannot resolve the symbol
[2:01] <mazzanet> on nativebigint
[2:02] <salahx> cvs -d:pserver:anonymous@cvs.sourceforge.net/cvsroot/freenet -z3 co -d freenet Contrib/jcpuid Contrib/freenet_ext Contrib/javax Contrib/fecimpl Contrib/NativeBigInteger
[2:03] <mazzanet> oooh extra stuff
[2:05] <mazzanet> actually
[2:05] <mazzanet> just ticking "Create missing directories that exist in repository" fixed it :P
[2:05] <salahx> he :)
[2:09] <salahx> you'll probably wabnt the other stuff though
[2:10] <mazzanet> i have the entire contrib module already there
[2:10] <salahx> ok!
[2:13] <mazzanet> hrm
[2:13] <mazzanet> whats this new jcpuid thingo
[2:14] <mazzanet> Jul 30, 2004 4:32:35 PM (freenet.node.http.DistributionServlet$DistributionRequest, YThread-22, NORMAL): RouteNotFound Fetching (running) freenet:CHK@c~fw58SjxQ~5TppuciSXd1OZRcARAwI,~MBTCdzHObqXmf8mUplXFg as freenet-webinstall.exe for DistributionServlet
[2:14] <mazzanet> :o
[2:15] <salahx> CPU detect to laod the correct optized GMP libraries
[2:15] <mazzanet> oh
[2:16] <mazzanet> umm
[2:16] <mazzanet> why on earth is my fresh cvs build, build 60105
[2:17] <salahx> Typo ?
[2:18] * Connelly (Default@c-24-21-147-63.client.comcast.net) Quit ("Leaving")
[2:18] <mazzanet> everythings all broken too
[2:18] <mazzanet> oh whoops
[2:18] <mazzanet> i forgot to 'ant clean'
[2:18] <mazzanet> before buildinh
[2:18] <mazzanet> g
[2:18] <salahx> shoud be 'ant cleanbuild' actually
[2:20] <mazzanet> Load due to messageSendTimeRequest = 221.9%
[2:20] <mazzanet> ouch
[3:04] * salahx (salahx@sc1-24.217.174.147.charter-stl.com) Quit ()
[3:23] * Janete (~jano@155.210.155.101) has joined #freenet
[4:09] * hirvox-work (~hirvox@cs181027153.pp.htv.fi) has joined #freenet
[4:10] <hirvox-work> hrm, can anyone verify that inserting splitfiles via the web interface and freenet.client.cli.Main works?
[4:10] <hirvox-work> latest unstable (build 60175) of course
[4:18] <Pascal> CANNOT RENAME NODE FILE node-temp TO node
[4:19] <Pascal> why is it trying to update my node file?!?!?
[4:22] <hirvox-work> (freenet.support.io.NIOInputStream, YThread-2, NORMAL): waited more than 120000ms in NIOIS.read() tcp/connection: 8481>127.0.0.1:44184,freenet.tra
[4:22] <hirvox-work> nsport.tcpConnection@140c281:freenet.support.io.NIOInputStream@c9131c- closing
[4:23] <hirvox-work> hrm, forgot lib/native.properties from my fec rpm, that's why it wouldn't work
[4:27] * kers (~kers@59.ppp135.rsd.worldonline.se) has joined #freenet
[4:33] <Pascal> in there an undelete comand for ext3?
[4:33] <Pascal> is there an undelete comand for ext3?
[4:54] <hirvox-work> that's better..
[4:54] * kers (~kers@59.ppp135.rsd.worldonline.se) Quit (Read error: 104 (Connection reset by peer))
[4:54] <hirvox-work> now I have pure java and native classes separated into different rpms
[5:41] * Sugadude (~Sugadude@defiant.circainformation.com) has joined #freenet
[6:20] * hirvox-work (~hirvox@cs181027153.pp.htv.fi) Quit ("Leaving")
[7:02] <iip_i2p> <Newsbyte> anyone here familiar with IIP?
[7:09] * Pascal (Pascal@c-24-13-17-191.client.comcast.net) Quit (Read error: 104 (Connection reset by peer))
[7:17] * robilad is now known as robilad[afk]
[7:28] * Hory (~Miranda@81.196.25.110) has joined #FreeNet
[7:29] <iip_i2p> <Newsbyte> anyone here familiar with IIP?
[7:44] <hirvox> hmm, 8kqph
[7:44] <hirvox> pSuccess still sucks, though
[7:48] * tyesinclair (~tyesincla@170.252.64.1) has joined #freenet
[7:50] * tyesinclair (~tyesincla@170.252.64.1) has left #freenet
[8:14] * Janete (~jano@155.210.155.101) has left #freenet
[8:28] * greycat (~wooledg@192.35.79.70) has joined #freenet
[9:04] * nmk_ (~nmk@power205.adsl.netsonic.fi) has joined #freenet
[9:04] * nmk_ (~nmk@power205.adsl.netsonic.fi) Quit (Client Quit)
[9:36] * Ash-Fox (Hal-9000@aae168.neoplus.adsl.tpnet.pl) Quit (Client Quit)
[9:40] * Ash-Fox (Hal-9000@aae168.neoplus.adsl.tpnet.pl) has joined #FreeNET
[9:42] <frog_> ribbit
[9:49] <frog_> "The Bandwidth Graph" -> http://mywebpages.comcast.net/jkcorson/FreeNet/5088bw.png
[9:50] <frog_> "The Other Graph" -> http://mywebpages.comcast.net/jkcorson/FreeNet/5088c.png
[9:52] <frog_> the peak outgoing KB/s gets interesting towards the end of first graph. The rateLimitingLoad seems to have some sort of max cap, somehow, around 1.25 ...
[9:53] * lolo-laptop (~lostlogic@68.251.84.226) has joined #freenet
[9:53] * frog_ likes the plot of queries/minute a whole lot more than yesterday's plot
[10:10] * lostlogic_ (~lostlogic@node-402418b2.mdw.onnet.us.uu.net) has joined #freenet
[10:11] <frog_> okay, i have to leave just as things are getting interesting :(
[10:11] <frog_> somebody tell the toad_ that http://mywebpages.comcast.net/jkcorson/FreeNet/5088c.png will auto-update every 10 minutes...
[10:30] * lostlogic (~lostlogic@node-402418b2.mdw.onnet.us.uu.net) Quit (Read error: 110 (Connection timed out))
[10:31] * robilad[afk] is now known as robilad
[10:32] * guido^pe (~unknown@dsl-213-023-233-132.arcor-ip.net) has joined #freenet
[10:56] * MikeW (Mike@159-134-232-105.as1.cbr.castlebar.eircom.net) has joined #freenet
[10:58] <toad_> hi
[10:58] <toad_> frog_: here?
[10:59] <toad_> cool, globalQuotaPerHour tracks received requests pretty closely
[10:59] <toad_> that means that that part of rate limiting works as intended
[11:00] <toad_> <Connelly> http://en.wikipedia.org/wiki/Freenet#Scalability
[11:00] <toad_> <Connelly> u guys might also want to edit that and make it more NPOV
[11:00] <toad_> <Connelly> it reads like half-propaganda and half-encyclopedia right now
[11:00] <toad_> hmmm
[11:00] <toad_> afaics we didn't write it... but i don't know what NPOV means :)
[11:00] <lolo-laptop> Neutral Point of View ... or were you joking?
[11:02] <toad_> lostlogic using 220kB/s total bandwidth on a 1.5m/768k DSL yay for getting everything out of my service.
[11:02] <toad_> woah
[11:03] <toad_> Fred,0.5,STABLE-1.50,5084 104
[11:03] <toad_> Fred,0.5,STABLE-1.50,5085 17
[11:03] <toad_> Fred,0.5,STABLE-1.50,5086 65
[11:03] <toad_> Fred,0.5,STABLE-1.50,5087 94
[11:03] <toad_> Fred,0.5,STABLE-1.50,5088 119
[11:03] <toad_> hmm, improving...
[11:03] <toad_> give it another day...
[11:04] <greycat> I still have more 5084 than any other
[11:04] <toad_> <greycat> I'm also curious why freenet.client.cli.Main put is throwing stack traces at me.
[11:04] <toad_> <greycat> INSERTING_BLOCKS (20/36):[1/1] queued: 0 running: 4
[11:04] <toad_> <greycat> java.io.EOFException: Could not parse message from stream
[11:04] <toad_> <greycat> at freenet.presentation.FCPRawMessage.<init>(FCPRawMessage.java:76)
[11:04] <toad_> <greycat> at freenet.presentation.ClientProtocol.readMessage(ClientProtocol.java:63)
[11:04] <toad_> <greycat> and so on
[11:04] <toad_> GAH
[11:04] <toad_> that's yet another grrring timeout
[11:05] <toad_> <KenMan> SOB - stable/5088 Unexpected Signal : 11 occurred at PC=0x4030CC9A - you could try using runfreenet to keep the node up, but if it reliably faults at 45 minutes, you need a better JVM...
[11:05] <toad_> Transfers active (Transmitting/Receiving) 35 (13/22) - my sending xfers isn't crazy-high...
[11:06] <iip_i2p> <oOo> 5088 now = a lot of transfer "actives" but a very small BW use, weird
[11:07] <toad_> <KenMan> which one ??
[11:07] <toad_> <Seed212> gnu
[11:07] <toad_> <KenMan> gnu.ogg , yeah. It took about 30 minutes though...
[11:07] <toad_> CHK? i didn't get the remix...
[11:07] <wind789> hhmm
[11:07] <wind789> CHK@FwDwUWZKHDd3IeAzR95gCSifKVwMAwI,URWySMGDJK2BUbFK4s1JkQ/Richard_Stallman-Free_Software(remix).mp3
[11:07] <wind789> I'll post them all, these are on stable
[11:07] <wind789> CHK@bL35x8zPm6Arup2hbeMoD1sRjygTAwI,uaQeOnu1lA0SQV4hROhnpg/Richard_Stallman-Free_Software.mp3
[11:07] <wind789> CHK@EtRYbTd71THRf-ElPO-BW4uFCRQMAwI,d3KCuwzSgC02n8~SwqNJ2Q/gnu.ogg
[11:07] <wind789> CHK@FwDwUWZKHDd3IeAzR95gCSifKVwMAwI,URWySMGDJK2BUbFK4s1JkQ/Richard_Stallman-Free_Software(remix).mp3
[11:07] <wind789> CHK@UGqWl--XXSk7g1lFTUWXD9hLsvgLAwI,A5I3Vy4JaAbcLvlJ2pu3qQ/Fenster-Free-Software-Song.ogg
[11:09] <toad_> <wind789> as it is I don't feel like dual botting at the moment - :(. I use linux exclusively, and use games under a commercial and semi-proprietary emulator (winex)... a lot doesn't run on winex, but a lot does. I don't however game right up to the minute... mostly i play warcraft 3, starcraft, and half life
[11:09] <wind789> so directX runs under winex?
[11:10] <toad_> <wind789> I play some games myself, mostly tribes and a little startcraft at the moment - starcraft can be made to run on regular (free) Wine, and runs well on WineX (except for major graphical issues in battle.net). No idea re Tribes.
[11:11] <toad_> wind789: some dX games run under Wine. Some run under WineX/Cedega
[11:11] <toad_> some don't run under either
[11:11] <wind789> cool
[11:11] <wind789> does linux have a built in NAT ?
[11:11] <toad_> <wind789> an open gl game called erternal lands works on both - is that the overhead-view open source one?
[11:11] <wind789> yeah
[11:12] <wind789> an MMORPG
[11:12] <toad_> <frog_> I play 'Battlefield 1942 and Battlefield Vietnam' mainly, but not too often. - BF1942 definitely runs on WineX, not sure re BV (probably not, but may be being worked on)
[11:13] <toad_> <frog_> but what is in a name (other than 242K new lines of code) ? a jar by any other name is still... - LOL! That would nearly triple our current size.. we'll see :)
[11:15] <toad_> <Iakin> Toad, you should also add an 'unwrite' call and then see if you get that key with the next select call.. :) - add a what?
[11:16] <toad_> <frog_> well... not until Freenet 0.9 , slated for a 2008 release - :). Well, that IS in fact optimistic, by my standards! ;)
[11:17] <toad_> <Pascal> CANNOT RENAME NODE FILE node-temp TO node
[11:17] <toad_> <Pascal> why is it trying to update my node file?!?!?
[11:17] <toad_> iakin told me why. i will fix when i get around to it.
[11:18] <toad_> <frog_> "The Bandwidth Graph" -> http://mywebpages.comcast.net/jkcorson/FreeNet/5088bw.png - hrrrm... increasing transfers...
[11:19] <toad_> <frog_> "The Other Graph" -> http://mywebpages.comcast.net/jkcorson/FreeNet/5088c.png - HRRM. Looks like it takes a ludicrously long time to recover from the initial spike... still oscillating but very slowly... hrrm
[11:20] <toad_> <frog_> somebody tell the toad_ that http://mywebpages.comcast.net/jkcorson/FreeNet/5088c.png will auto-update every 10 minutes... - confirmed
[11:20] <toad_> <iip_i2p> <oOo> 5088 now = a lot of transfer "actives" but a very small BW use, weird - hmmm...
[11:21] <toad_> <wind789> does linux have a built in NAT ? - yup, linux has built in serious router support (in the kernel)
[11:22] <toad_> all sorts of wierd networking stuff
[11:22] <toad_> frog_: evidently not here? :(
[11:23] <toad_> i don't think oscillation is necessarily the enemy
[11:23] <toad_> i think excessive oscillation is a problem
[11:23] <toad_> the graph suggests that the startup spike is a MAJOR problem
[11:24] <toad_> but there will be an interaction with # connected peers, even if we store the data on globalQuota average and incoming requests...
[11:25] <wind789> gotta run bbl
[11:25] <toad_> seeya
[11:41] * TLF (francisco@133.Red-81-40-113.pooles.rima-tde.net) has joined #freenet
[11:43] * MikeW (Mike@159-134-232-105.as1.cbr.castlebar.eircom.net) Quit ()
[11:45] * toad_ just withdrawn next month's paycheck, yay
[11:53] <hirvox> hmm, my node's handling a lot more traffic than it used to
[11:54] <hirvox> 9kqph currently
[11:54] <greycat> I still get RNFs when inserting an FEC splitfile
[11:55] <toad_> hmmm
[11:56] <toad_> it looks to me based on my local node that there ARE oscillations...
[11:56] <Sugadude> Can I get an opinion from someone? When using RDF/RSS/Atom feeds on Freenet, how should I represent the URI? As freenet:<KEY> or as http://127.0.0.1:8888/<KEY> ?
[11:57] <toad_> Sugadude: /freenet:<KEY> is the convention inside freenet (relative URL)
[11:57] <toad_> or /<KEY>
[11:57] <toad_> more likely
[11:57] <toad_> brb
[11:58] <Sugadude> toad: I'm talking about RDF-like, which require a fully-qualified URI, i.e. not-relative.
[11:59] <hirvox> freenet:<KEY> would be the "right" way, but there's no handlers for freenet: urls, and 127.0.0.1:8888 will break if the node isn't running on localhost at port 8888
[11:59] <Sugadude> hirvox: I know all that, but which one do you think I should choose?
[12:01] <hirvox> I'd have to say http://127.0.0.1:8888, freenet:<KEY> url won't probably work
[12:01] <hirvox> urls, that is
[12:02] <Sugadude> *URI, actually, but never mind. ;) Thanks.
[12:02] <Sugadude> (Actually, I'm pretty sure that freenet keys are URNs, but again, never mind.)
[12:06] <toad_> well, ian has always opposed any attempt to make freenet:<key> work in browsers
[12:06] <toad_> Sugadude: URN is AKA URI
[12:06] <toad_> URL is also aka URI
[12:06] <toad_> freenet URLs are actually URIs
[12:06] <toad_> :)
[12:07] * interrupt (~chatzilla@64.122.23.213) has joined #freenet
[12:07] <toad_> URNs are something else... maybe freenet URIs could be URNs, I dunno
[12:08] <Sugadude> URL/URN are derived from URI, AFAIK. URL points to data and where to get it (Yahoo/index), URN points only to what the data is (i.e. SHA1/MD5, anything global, really).
[12:08] <hirvox> freenet: URIs would be fine if we had a little BHO or .xpi that would transparently rewrite them to http://yournode:yourport/<key>
[12:08] * Ash-Fox (Hal-9000@aae168.neoplus.adsl.tpnet.pl) Quit (Client Quit)
[12:09] <Sugadude> Why is Ian oppososed to using to freenet: in browsers? I think it's a Good Thing.
[12:10] * Ash-Fox (Hal-9000@aae168.neoplus.adsl.tpnet.pl) has joined #FreeNET
[12:12] <toad_> Sugadude: because he thinks we'd have to maintain plugins for every browser
[12:12] <toad_> hirvox: what's BHO?
[12:13] <Sugadude> toad: Browser Helper Object. An IE plugin.
[12:13] <hirvox> Browser Helper Object. like Mozilla's .xpis, but for IE
[12:13] <toad_> hmmm
[12:13] <toad_> this is a little concerning...
[12:13] <hirvox> actually, I think that windows has a way of registering a program to handle specific protocols
[12:13] <toad_> rateLimitingLoad has only once been under 1.0 in the last hour
[12:14] <toad_> ah no
[12:14] <toad_> 13 minutes of 60
[12:14] <toad_> hrrm
[12:14] <hirvox> so you could have irc: urls automagically starting your irc application, telnet: urls to start your terminal program, news: to start your newsreader etc
[12:14] <Sugadude> & Mozilla (and Firefox?) already have a plugin for redirecting freenet: links to http://node:port/%s. With a bit of registery hacking the same can be done for IE.
[12:14] <toad_> hirvox: and shell: urls to mess up your system :)
[12:14] <hirvox> ;-)
[12:15] <Sugadude> What's shell: URIs?
[12:15] <hirvox> exactly what it sounds like: a way to call commands via urls
[12:15] <toad_> gah
[12:16] <toad_> http://mywebpages.comcast.net/jkcorson/FreeNet/5088c.png
[12:16] <toad_> we STILL oscillate...
[12:16] <hirvox> it caused the recent mozilla vulnerablity. mozilla was passing urls with unknown protocols straight to the OS
[12:16] <toad_> just over a longer time period
[12:16] <toad_> but quite seriously
[12:17] <Sugadude> An idea surely by the last living brain-cell in some MS employee. ;)
[12:17] <toad_> oscillation goes down to 20 qpm and up to 40
[12:17] <toad_> so fairly major oscillation...
[12:18] <toad_> thelema here? frog_ here? KenMan here? frogman here? anyone here?
[12:18] <toad_> does ANYONE have any opinions on the graph?
[12:19] <toad_> we need to do something about the startup spike, but that's not going to be enough...
[12:19] <toad_> as it doesn't seem to settle...
[12:24] * toad_ wonders if there's a fundamental issue with bandwidth limiting here...
[12:25] <toad_> if load is mostly below 1, the globalQuota increases
[12:25] <toad_> if load is mostly above 1, the globalQuota falls
[12:28] <toad_> between 8AM and 10AM, most of the time the link is saturated
[12:28] <toad_> so mostly the globalQuota falls
[12:28] <toad_> ditto between 10 and 12 although it's not so clear-cut
[12:33] * toad_ WOAH re http://127.0.0.1:8888/servlet/nodestatus/diagnostics/normalizedSuccessTime/hour
[12:33] <toad_> there are some absurd times in my copy...
[12:34] <toad_> this is probably what causes the feedback...
[12:34] <toad_> 7/30/04 11:00:00 AM BST 3941605.0 1.1595988286701234E7 10973.0 804068.0 19
[12:34] <toad_> 7/30/04 12:00:00 PM BST 2570557.409090909 6318440.618887276 488798.0 488798.0 22
[12:34] <toad_> 7/30/04 1:00:00 PM BST 1320314.5333333334 3643122.4899148936 NaN NaN 15
[12:34] <toad_> 7/30/04 2:00:00 PM BST 3257258.466666667 7229015.9792066235 538432.0 538432.0 15
[12:34] <toad_> 7/30/04 3:00:00 PM BST 4084507.117647059 9428853.619267037 944594.0 944594.0 17
[12:34] <toad_> 7/30/04 4:00:00 PM BST 2086977.3333333333 2557106.937305088 NaN NaN 15
[12:34] <toad_> 7/30/04 5:00:00 PM BST 916325.1333333333 1857271.0561057741 NaN NaN 30
[12:35] <toad_> these are AVERAGES
[12:35] <interrupt> lol
[12:35] <toad_> for the time it takes for a request to succeed, including downloading
[12:35] * robilad is now known as robilad[afk]
[12:35] <toad_> and they're ludicrous...
[12:36] <toad_> hmmm
[12:36] <toad_> these are normalized...
[12:42] <verl> toad: the same stat on my node is diffrent, would you like to see?
[12:42] * TLF (francisco@133.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.")
[12:42] <toad_> successSearchTime is low, 30 seconds maybe
[12:42] <toad_> verl: sure
[12:43] <verl> running the patched jar you sent me yesterday:
[12:43] * nicktastic (~nicktasti@c-67-163-227-74.client.comcast.net) Quit (Remote closed the connection)
[12:43] <verl> 2004-07-30 17:00:00 CEST 6784820.3965517245 1.6384307071205597E7 1396855.0 1396855.0 58
[12:43] <verl> 2004-07-30 18:00:00 CEST 6108893.387096774 1.7768048445106663E7 NaN NaN 31
[12:43] <verl> 2004-07-30 19:00:00 CEST 2912808.3870967743 1.0689176054107465E7 NaN NaN 31
[12:47] <toad_> verl: yikes
[12:47] * nicktastic (~nicktasti@c-67-163-227-74.client.comcast.net) has joined #freenet
[12:47] <verl> tell me what it means!
[12:47] <toad_> even 3M is bad (it's an entire hour!)
[12:47] <toad_> 6M is nearly two hours
[12:47] <toad_> one way or another, not good...
[12:48] <interrupt> lol, mines 9.8M
[12:48] <toad_> interrupt: what is?
[12:48] <toad_> normalizedSuccessTime?
[12:49] <interrupt> aye
[12:49] <toad_> granted it'll be pulled up by some really bad stuff at the top end...
[12:55] <toad_> hmmm
[12:55] <toad_> this looks bad...
[12:56] <toad_> rate limiting oscillates because transfers often take totally absurd periods of time...
[12:58] <toad_> interrupt: stable or unstable?
[12:58] <interrupt> stable
[13:00] <toad_> fullSearchTime can get highish... but I doubt it's over 200 seconds much...
[13:00] <toad_> if ever
[13:00] <toad_> the average is pretty low - 10 seconds, 5 seconds, 30 seconds
[13:01] <toad_> not low enough but too low to cause any of this...
[13:01] * verl_ (~verlverl@h165n2fls33o877.telia.com) has joined #freenet
[13:02] <toad_> wb verl_
[13:02] <verl_> thank you :)
[13:03] * verl (~verlverl@h165n2fls33o877.telia.com) Quit (Read error: 54 (Connection reset by peer))
[13:04] <toad_> these stats have to be bogus...
[13:05] <hirvox> hmm, fixKeys added <something> - AND IT WROTE SUCCESSFULLY! This can happen occasionally but if it happens a lot, it could indicate a broken NIO implementa
[13:05] <hirvox> tion
[13:06] <toad_> hmmm
[13:06] <toad_> it might all be small files like this one:
[13:06] <toad_> PRI Jul 30, 2004 5:44:59 PM (freenet.node.rt.NGRouting, YThread-1060, MINOR): Not logging transfer rate becau
[13:06] <toad_> se size=4097, transferTime=5094
[13:06] * toad_ restarts with better logging/diags
[13:10] * nicktastic (~nicktasti@c-67-163-227-74.client.comcast.net) Quit (Remote closed the connection)
[13:12] * toad_ waits for restarted node to become responsive... :|
[13:16] * kers (~kers@52.ppp138.rsd.worldonline.se) has joined #freenet
[13:20] * jay (jcl@ool-18bf6dac.dyn.optonline.net) has joined #freenet
[13:26] <toad_> hmmm
[13:27] <toad_> hmmm
[13:28] <toad_> the lack of completed requests by transfer so far... suggests that maybe reqs DO take aaaages on average...
[13:28] <toad_> otoh it could still be a stats bug...
[13:30] <toad_> bbiab
[13:38] <toad_> hmmm
[13:39] <toad_> request completion times so far:
[13:39] <toad_> 3196ms
[13:39] <toad_> 261,185ms
[13:39] <toad_> 517,018ms
[13:39] <toad_> 186,675ms
[13:39] <toad_> the overwhelming majority of the higher figures is the transfer
[13:41] <toad_> also 500 seconds is 8 minutes... the node has only been up for 25 minutes...
[13:41] <toad_> hmmm
[13:41] <toad_> the 500 seconds was at 500 bytes/sec, so that was a 256kB chunk...
[13:42] <toad_> that rate would make a 1MB chunk take 2000 seconds...
[13:42] <toad_> since most transfers are 1MB chunks, that explains why I haven't seen any big xfers yet...
[13:43] <toad_> the 256kB chunks are from my own local splitfile request
[13:43] <toad_> for the stallman oggs/mp3s
[13:43] * Sugadude (~Sugadude@defiant.circainformation.com) Quit (Remote closed the connection)
[13:46] * TLF (francisco@133.Red-81-40-113.pooles.rima-tde.net) has joined #freenet
[13:50] * ShaunMacPherson (~ShaunMac@Sudbury-HSE-ppp3981095.sympatico.ca) has joined #freenet
[13:52] <hirvox> 16kqph, a new record
[13:59] <greycat> Fred,0.5,STABLE-1.50,5084 137
[13:59] <greycat> Fred,0.5,STABLE-1.50,5085 23
[13:59] <greycat> Fred,0.5,STABLE-1.50,5086 57
[13:59] <greycat> Fred,0.5,STABLE-1.50,5087 82
[13:59] <greycat> Fred,0.5,STABLE-1.50,5088 97
[13:59] <greycat> most of the 5087 adopters are moving to 5088; 5084 users seem to be staying there.
[14:04] <toad_> not surprising really
[14:05] <toad_> woah
[14:05] <toad_> just got two requests completed successfully
[14:05] <toad_> one did it in 131456.0ms
[14:05] <toad_> the other in 1,354,582.0
[14:05] <toad_> the first was a 4kB file, the second was a 256kB file
[14:06] <toad_> that is 197 bytes per second...
[14:09] * jay (jcl@ool-18bf6dac.dyn.optonline.net) Quit (".")
[14:09] * toad_ opens routing table... maybe 5088 nodes have really really slow transfer rates...
[14:09] <toad_> maybe it's just a bug in 5088
[14:09] <toad_> in other words
[14:10] * Pascal (Pascal@c-24-13-17-191.client.comcast.net) has joined #Freenet
[14:10] <toad_> hmmm
[14:10] <toad_> perhaps they are a bit lower...
[14:10] <toad_> hmmm
[14:10] <toad_> hi Pascal
[14:11] <Pascal> howdy
[14:11] * toad_ is about to mail devl recommending we go to a fixed small file size, binary presentation, transparent FEC in fproxy, and network content reset
[14:12] <toad_> on the grounds mainly that I don't see how rate limiting can work if a large number of transfers take 1000 seconds+ to complete
[14:12] <greycat> Would that also be "transparent FEC in FCP"?
[14:12] * kers (~kers@52.ppp138.rsd.worldonline.se) Quit (Remote closed the connection)
[14:12] <toad_> greycat: I don't know, maybe we could add an API for it, but what's wrong with the existing FEC support in FCP
[14:12] <toad_> ?
[14:13] <greycat> It *has* FEC support?
[14:13] <hirvox> obviously the unstable network is going to act as the guinea pig?
[14:13] <toad_> greycat: I know
[14:13] <toad_> hirvox: possibly
[14:13] <toad_> possibly we'd continue with stable as sort-of 0.5, unstable with reset is the 0.6 network
[14:13] <toad_> which eventually becomes stable 0.6
[14:14] <toad_> i.e. it would be protocol compatible with 0.6, at least for a while
[14:14] <greycat> searching for FEC on http://freenetproject.org/index.php?page=fcp gives no matches other than "affects"
[14:14] <toad_> greycat: we have FEC support in FCP
[14:14] <greycat> then it's not documented
[14:14] <toad_> ummm
[14:14] <toad_> it is somewhere :)
[14:15] * kers (~kers@52.ppp138.rsd.worldonline.se) has joined #freenet
[14:16] <toad_> hmmm
[14:16] <toad_> possible alternativE:
[14:16] <toad_> track requests that are "in flight"
[14:16] <toad_> use that to moderate rate limiting?
[14:17] <toad_> i.e. like the expected inbound transfers?
[14:17] <toad_> except for outbound transfers, and more accurate?
[14:19] * toad_ hmmm
[14:20] <toad_> so we track how many requests are in what state?
[14:21] <toad_> we have N requests transferring data for a total of M bytes
[14:21] <toad_> L requests awaiting store data
[14:21] * TLF (francisco@133.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.")
[14:21] <toad_> P requests awaiting initial Accepted
[14:22] <toad_> Q requests in actual routing, with a total of K bytes?
[14:22] <toad_> hmmm
[14:24] <toad_> AHA
[14:24] <toad_> what we want to do is not to include bandwidth usage in rate limiting (well, not primarily - maybe we will anyway)
[14:24] <toad_> but to include EXPECTED bandwidth usage
[14:25] <toad_> we want to accept queries so that the expected transfers over the next <long time period> will add up to lets say 80% of our bandwidth limit
[14:25] <toad_> which leaves us some slacxk
[14:25] <toad_> slack
[14:31] * toad_ found a really quick fix :)
[14:36] <hirvox> what is it?
[14:40] * hirvox reads the post on dev
[14:41] <hirvox> it sounds too simple :)
[14:41] * TLF (francisco@15.Red-81-40-113.pooles.rima-tde.net) has joined #freenet
[14:46] <toad_> hirvox: well, i'll probably have to make it more accurate
[14:46] <toad_> i.e. more complex
[14:46] <toad_> but i can implement a first draft quickly
[14:46] <hirvox> go for it
[14:59] <hirvox> hmm, I wonder how that would react to my situation, where the node doesn't always have the bandwidth it was allocated?
[15:09] <toad_> your situation will be largely controlled by messageSendTimeRequest probably
[15:10] * dystopia_ (admin@gaia-wan0.ipv6.freeshell.bofx.net) has joined #freenet
[15:11] * piranha (piranha@3ffe:b80:1ca1:0:0:0:deca:fbad) Quit (zelazny.freenode.net irc.freenode.net)
[15:11] * dystopia (admin@gaia-wan0.ipv6.freeshell.bofx.net) Quit (zelazny.freenode.net irc.freenode.net)
[15:11] * dystopia_ is now known as dystopia
[15:11] <hirvox> ah, ok
[15:12] * toad_ had to add code to track probability of transfer given an incoming request...
[15:15] <toad_> amazingly we didn't track that..
[15:15] * toad_ also turned off conventional high level bandwidth limiting completely
[15:15] <toad_> another way to deal with it would be to record overhead somehow
[15:16] * piranha (piranha@3ffe:b80:1ca1:0:0:0:deca:fbad) has joined #freenet
[15:17] * robilad[afk] is now known as robilad
[15:19] * Gunnar (~gunnar@34.80-203-209.nextgentel.com) has joined #freenet
[15:20] <toad_> hi robilad
[15:20] <robilad> hi toad_
[15:24] * Ash-Fox (Hal-9000@aae168.neoplus.adsl.tpnet.pl) Quit (Client Quit)
[15:24] <toad_> how's kaffe?
[15:26] <robilad> growing fast ;
[15:26] <robilad> 0
[15:26] <toad_> :|
[15:26] * TLF (francisco@15.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.")
[15:26] <robilad> guilhem's hacking on the gc, i'm going thourgh the patch queue
[15:27] <toad_> cool
[15:27] <toad_> what's up with the GC?
[15:28] <robilad> swutching to boehm weiser to make it faster
[15:28] <toad_> boehm what?
[15:28] <robilad> http://www.hpl.hp.com/personal/Hans_Boehm/gc/
[15:28] <toad_> what's weiser?
[15:28] <robilad> one of the authors
[15:28] * Ash-Fox (Hal-9000@aae168.neoplus.adsl.tpnet.pl) has joined #FreeNET
[15:29] <toad_> i always knew it as boehm conservative GC :)
[15:29] * interrupt (~chatzilla@64.122.23.213) Quit ("Chatzilla 0.9.64b [Mozilla rv:1.7/20040707]")
[15:38] * _bio (~cyrus@botanix.wu-wien.ac.at) has joined #freenet
[15:38] <_bio> hi
[15:39] <toad_> hi
[15:41] * lolo-laptop (~lostlogic@68.251.84.226) Quit ("Client exiting")
[16:01] <Pascal> what program is used to write cds on linux?
[16:02] <greycat> cdrecord, usually
[16:02] <Pascal> thanks
[16:02] <greycat> there are umpteen different GUI front-ends to it
[16:30] * TLF (francisco@15.Red-81-40-113.pooles.rima-tde.net) has joined #freenet
[16:36] * greycat (~wooledg@192.35.79.70) Quit ("Lick Bush in '04")
[17:01] * TLF (francisco@15.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.")
[17:01] * TLF (francisco@15.Red-81-40-113.pooles.rima-tde.net) has joined #freenet
[17:17] * sanity (~r00t@lv-67-76-71-152.sta.sprint-hsd.net) has joined #freenet
[17:17] * ChanServ sets mode +o sanity
[17:20] <kers> happy sysadmin day everyone! - http://www.sysadminday.com/
[17:21] * TLF (francisco@15.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.")
[17:21] * TLF (francisco@15.Red-81-40-113.pooles.rima-tde.net) has joined #freenet
[17:23] * sanity (~r00t@lv-67-76-71-152.sta.sprint-hsd.net) Quit ("Leaving")
[17:23] * sanity (~r00t@lv-67-76-71-152.sta.sprint-hsd.net) has joined #freenet
[17:23] * ChanServ sets mode +o sanity
[17:26] * toad_ (toad@82-32-16-91.cable.ubr03.azte.blueyonder.co.uk) Quit ("using sirc version 2.211+KSIRC/1.3.10")
[17:27] * toad_ (toad@82-32-16-91.cable.ubr03.azte.blueyonder.co.uk) has joined #freenet
[17:28] <toad_> hi ppl
[17:28] <kers> whats up toad
[17:29] * toad_ is trying to debug rate limiting
[17:31] <kers> ok, good luck :) happy sysadmin day (in lack of "freenet developer day")
[17:33] * kers (~kers@52.ppp138.rsd.worldonline.se) Quit ("Leaving")
[17:33] <_bio> hmm, does it really take so long to get my node running? runs now for approx 8 hours and cant get 90% of the pages :/
[17:34] * sanity (~r00t@lv-67-76-71-152.sta.sprint-hsd.net) Quit (Read error: 60 (Operation timed out))
[17:36] <toad_> woah
[17:37] <toad_> outputBytes:
[17:37] <toad_> 7/30/04 10:50:00 PM BST 1245 628049 5.971650694962998E7 48.189558232931724
[17:37] <toad_> 7/30/04 10:51:00 PM BST 1254 628001 6.034446169463423E7 47.854864433811805
[17:37] <toad_> 7/30/04 10:52:00 PM BST 1149 617500 6.096513572161599E7 52.24107919930374
[17:37] <toad_> 7/30/04 10:53:00 PM BST 917 599070 6.1587792333198994E7 64.94002181025081
[17:37] <toad_> 7/30/04 10:54:00 PM BST 1249 632803 6.2191757776365444E7 48.37469975980785
[17:37] <toad_> 7/30/04 10:55:00 PM BST 1225 617876 6.281181704334061E7 48.990204081632655
[17:37] <toad_> outputBytesTrailerChunks:
[17:37] <toad_> 7/30/04 10:51:00 PM BST 2612 490636 4.288671679547727E7 22.973966309341503
[17:37] <toad_> 7/30/04 10:52:00 PM BST 2324 436772 4.3358578745721474E7 25.821428571428573
[17:37] <toad_> 7/30/04 10:53:00 PM BST 1894 356072 4.375949941604271E7 31.449841605068638
[17:37] <toad_> 7/30/04 10:54:00 PM BST 2607 489845 4.416896464886697E7 23.17337936325278
[17:37] <toad_> 7/30/04 10:55:00 PM BST 2440 458422 4.4640114743197985E7 24.597950819672132
[17:38] <toad_> conclusion? it probably
[17:38] <toad_> WOULD work if it wasn't for the increased overhead...
[17:38] <toad_> err i mean
[17:38] <toad_> the new limiting algo probably would work if it wasn't for the increased overheads
[17:43] <oierw> hey - did you ever figure out what was causing the connection problems?
[17:46] * toad_ adds a diagnostic for padding overhead
[17:46] <toad_> oierw: yes
[17:46] <toad_> bug in 5087
[17:47] * Ash-Fox (Hal-9000@aae168.neoplus.adsl.tpnet.pl) Quit (Client Quit)
[17:47] <oierw> ah, okay then
[17:47] <oierw> no weird crypto bug then. that's good
[17:49] * phony (~resu@d463c26d.datahighways.de) has joined #freenet
[17:49] <toad_> ahhh
[17:49] <toad_> there was a wierd crypto bug on unstable
[17:49] <toad_> but i fixed it before merging
[17:49] <phony> What does it mean, if you send out a request and the other clients "back off"?
[17:50] <toad_> phony: it's not the clients...
[17:50] <toad_> you got a message saying route not found, lots of nodes backed off, right?
[17:50] <toad_> it means that the nodes are too busy for you to send them a request
[17:50] <toad_> nodes tell each other how often to send requests
[17:51] <phony> i've been running freenet for almost 4 hours, 60MB of data have been transferred, yet I'm still not able to open a single webpage
[17:52] <toad_> hmmm
[17:52] <toad_> how many connections?
[17:52] <toad_> stable or unstable?
[17:52] <phony> about 40 I'd guess
[17:53] <phony> STABLE-1.50
[17:53] <toad_> from the open connections page?
[17:53] <phony> yep
[17:53] <toad_> any incoming?
[17:54] <phony> 3 inbound, the others outbound
[17:54] <phony> but there's only one connection that's not idle, and that one is outbound
[17:55] <phony> now they're all idle...
[17:56] <toad_> hmmm
[17:56] <toad_> don't know
[17:56] <toad_> 40 conns should be enough
[17:56] <toad_> but your node is still quite inexperienced...
[18:01] * robilad is now known as robilad[afk]
[18:03] * lostlogic_ is now known as lostlogic
[18:37] * Ash-Fox (Hal-9000@aay216.neoplus.adsl.tpnet.pl) has joined #FreeNET
[18:38] * pupok (~janie@lv-69-69-100-62.sta.sprint-hsd.net) has joined #freenet
[18:42] <toad_> hmmm
[18:43] <toad_> i think there may be some breakage in serialization...
[18:52] <toad_> splat!
[18:52] * toad_ munches another bug
[18:58] * TLF (francisco@15.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.")
[19:00] <mazzanet> woo
[19:00] <mazzanet> i can finally browse
[19:00] * mazzanet dances around
[19:18] * toad_ hmz at paypal's rather paranoid money laundering verification system
[19:19] <toad_> verify by telephone (they phone you, enter PIN), letter (they send you something, you send it back to web site), AND bank transfer
[19:20] <toad_> all 3 are required
[19:20] <mazzanet> !
[19:21] <mazzanet> mmm nice fast browsing
[19:22] <toad_> mazzanet: hrrm
[19:22] <toad_> ?
[19:22] <toad_> of freenet?
[19:22] <mazzanet> ya
[19:22] <mazzanet> 60175
[19:22] <toad_> cool
[19:23] <mazzanet> nice pretty activelinks for all the bookmarks
[19:23] <mazzanet> yoyo loaded in about 7 secs
[19:23] <mazzanet> :D
[19:37] <toad_> bbl
[19:40] * Ash-Fox (Hal-9000@aay216.neoplus.adsl.tpnet.pl) Quit (Client Quit)
[19:43] * Ash-Fox (Hal-9000@aay216.neoplus.adsl.tpnet.pl) has joined #FreeNET
[19:54] * verl_ changes topic to 'Freenet: Taming the World's Largest Tamagotchi | Upgrade to 5088 | Unstable: Upgrade to 60176 | Logs: http://newton.matcmp.ncc.edu/~lockej/freenet/chanlog/'
[19:59] * greycat (rfc1413@wooledge.org) has joined #freenet
[20:14] * pupok (~janie@lv-69-69-100-62.sta.sprint-hsd.net) Quit (Read error: 110 (Connection timed out))
[20:27] * Hory (~Miranda@81.196.25.110) Quit (Read error: 104 (Connection reset by peer))
[20:29] * Tailchaser (~tailchase@pool-68-162-16-41.nwrk.east.verizon.net) has joined #freenet
These logs were automatically created by Jay Oliveri with his gimp hapi on irc.freenode.net.
