14:03 < jrandom> 1) 0.4.1.3 14:03 < jrandom> 2) Tunnel test time, and send processing time 14:03 < jrandom> 3) Streaming lib 14:03 < jrandom> 4) files.i2p 14:03 < jrandom> 5) ??? 14:03 < jrandom> 0) hi 14:03 * jrandom waves 14:04 < modulus> hi hi 14:04 < jrandom> weekly status notes posted up @ http://dev.i2p.net/pipermail/i2p/2004-October/000469.html 14:04 < deer_> howdy 14:04 < jrandom> i didn't spend much time on the notes, so they're pretty brief 14:05 < jrandom> but, c'est la vie 14:05 < jrandom> moving on to 1) 0.4.1.3 14:05 < jrandom> the release came out the other day and its been.. well... largely like before 14:05 < jrandom> working good enough for most things, but not as reliable as we'd like 14:06 < jrandom> throughput is still low, but thats a know issue to be dealt with in 0.4.2 14:06 < jrandom> as mentioned in the email, I dont expect there to be any more 0.4.1.* releases 14:07 < jrandom> I dont have much more to say on that - anyone have any comments / concerns? 14:07 < deer_> yes: what about the freeze-up? 14:09 < jrandom> I'm not going to discount the possibility that your machine hung due to I2P, but I severely doubt it 14:09 < jrandom> no one else has ever reported that happening on any platform 14:09 < deer_> well...it must be related to it somehow, if not directly, IMHO 14:09 < deer_> maybe the java? 14:10 < jrandom> you're on 1.5 on w2k? 14:10 < jrandom> or 1.4.2_05? 14:10 < deer_> nope, 1.5 14:10 < jrandom> ok 14:10 < deer_> I can't exclude it's something else, ofcourse 14:11 < deer_> could be coincidence it happend two times 14:11 < jrandom> well, we can discuss further how to find out the cause after the meeting if you'd like 14:11 < deer_> but the last time..I dunno...nothing much else was running, then 14:11 < deer_> 1.5 on w2k works good for me :) 14:11 < deer_> indeed, though 14:11 < deer_> isn't there a simple debug log or something? 14:11 < jrandom> if it happens again, please send me wrapper.log and logs/log-router-*.txt 14:11 < deer_> that might be usefull when it freezes 14:11 < jrandom> there are more logs than dirt ;) 14:12 < jrandom> ok cool dinoman 14:12 < jrandom> perhaps it was some interaction with your software firewall 14:12 < deer_> maybe 14:12 < jrandom> but, yeah,bounce me logs if it happens again 14:12 < jrandom> (please :) 14:12 < deer_> well, that it would get blocked, I would understand 14:12 < deer_> but a total freeze...dunno...was creepy 14:13 < deer_> on the bright side: I've 27/63 now 14:13 < jrandom> great 14:13 < jrandom> ok, anyone else have any questions/comments/concerns with 0.4.1.3? 14:13 < deer_> I'll guees I'll ask Whoo to guide my through the eep thingy 14:13 < deer_> just don't use it with Sygate Personal Firewall bad bad 14:13 < deer_> why? 14:14 < deer_> crash 14:14 < deer_> yes; you forgot 6) profit!! 14:14 < deer_> ;-) 14:14 < deer_> crash? 14:14 < deer_> ermm 14:14 < jrandom> dinoman: it crashes your OS? the firewall? I2P? 14:14 < deer_> well, wouldn't that explain it, then? ;-) 14:15 < jrandom> newsbyte: are you running Sygate Personal Firewall? 14:15 < deer_> indeed 14:15 < deer_> well, not on my router 14:15 < deer_> but on the puter, yes 14:15 < deer_> seems we're on to something 14:16 < deer_> newsbyte: /join #i2p-chat so jrandom can get through his meeting 14:16 < deer_> though it doesn't crash/freeze immediately, apperently 14:16 < deer_> os it crashes windows 14:16 < deer_> ? 14:16 < deer_> jrand is already here 14:16 < deer_> sorry looked away 14:16 < jrandom> ok, perhaps we can look into what SPF is b0rking on 14:16 < jrandom> if there's nothing else on 0.4.1.3, moving on to 2) Tunnel test time, and send processing time 14:17 < jrandom> there was some discussion yesterday exploring some of the timeouts, and basically things just occationally take too long 14:17 < jrandom> i dont think the spikes you can see in http://dev.i2p.net/~jrandom/processingTime.png are legitimate though 14:18 < jrandom> well, they're real - it really does take that long 14:18 < jrandom> what i mean is, we should be able to get rid of them 14:18 < jrandom> some queueing is going to happen, but if we are more careful with what we accept, we should be able to reduce it 14:19 < jrandom> the delays are also likely due to some occational spikes in job processing time, which we can tune the fsck out of 14:20 < jrandom> in general though, the message queueing seems all right, even if it spikes up some tunnel tests 14:20 < deer_> darn..I wish freenet and i2p could really merge...seems like progress would be a lot faster, possibly beneficial to both 14:20 < deer_> yeah, I don't see why fsck would be useful for jon processing :) 14:20 < deer_> s/jon/job/ 14:21 < jrandom> there is much potential for collaboration, but the two projects have very different aims 14:21 < jrandom> !thwap Ragnarok 14:21 < deer_> ermm 14:21 < jrandom> oh, one thing I mentioned yesterday 14:21 < deer_> I don't think the projects' goals, however, are all that different... 14:22 < deer_> jrandom: technical goals 14:22 < jrandom> newsbyte: we can discuss that in 5) ??? or later if you prefer, we're on 2) right now 14:22 < deer_> oops newsbyte: technical goals 14:22 < deer_> hehe 14:22 < deer_> yes, and 3)profit! according to /. traditions! 14:22 < deer_> :-) 14:22 < deer_> I can't believe Tor is not backwards compatible from 0.0.8 to 0.0.8.1 14:23 < jrandom> with the tunnel testing, there is a floor to the test period - currently set to 5 seconds by default 14:23 < jrandom> the previous release had a hard limit of 30 seconds, but you can configure your own tunnel test time by updating http://localhost:7657/configadvanced.jsp and adding "router.tunnelTestMinimum=10000" (or whatever - that value is in milliseconds) 14:23 < deer_> those seconds, are they alchimagical? 14:24 < jrandom> the 5s default should be fine though 14:24 < deer_> I actually upgraded Tor the day before yesterday because it stopped working, and now the network is telling me again, I have a non compatible version... what the.. 14:24 < deer_> oh... hello everyone :) 14:24 < jrandom> newsbyte: the tunnel test time is MAX(avgTunnelTestTime*2, minTunnelTestTime) 14:25 < jrandom> (we have the minTunnelTestTime because otherwise a series of fast tests could cause a cascading failure) 14:26 < jrandom> more details can be found in http://dev.i2p.net/cgi-bin/cvsweb.cgi/i2p/history.txt?rev=HEAD 14:26 < deer_> hmm 14:26 < deer_> this is really funny... a job agency wants me to use Internet Explorer, otherwise I'm not able to register an application 14:27 < jrandom> *cough* y'all realize these meeting logs go on the web, right? :) 14:27 < deer_> <-- not too good in english 14:27 < deer_> they do?! 14:27 < deer_> Hi mum! 14:27 < deer_> ;-) 14:27 < deer_> um, sorry. .I'm disturbing the meeting.. I'm off 14:28 < jrandom> naw, please stay, but discuss i2p stuff ;) 14:28 < deer_> don't worry; disturbing is an art, just keep an eye on me, and you'll learn 14:28 < deer_> ;-) 14:28 < jrandom> ok, anything else on 2) Tunnel test time, and send processing time ? 14:28 < deer_> focus people 14:29 -!- znation [~znation@ip68-226-31-250.tc.ph.cox.net] has quit [Read error: 60 (Operation timed out)] 14:29 < jrandom> if not, moving on to 3) Streaming lib 14:29 < jrandom> as mentioned in the status notes, lots of progress 14:29 -!- znation [~znation@ip68-226-31-250.tc.ph.cox.net] has joined #i2p 14:29 < deer_> done by you? 14:29 < jrandom> still not there yet, but I hope to be doing some live tests in the next week 14:30 < jrandom> i've been working on the streaming lib, yeah 14:30 < jrandom> i finally got it ping()ing earlier today ;) 14:30 < deer_> nice :) 14:31 < jrandom> ok, i dont really have anything else to add about that 14:31 < jrandom> anyone have any questions / comments / concerns? 14:31 < deer_> ermm...speed? 14:31 < jrandom> speed is fine 14:31 < deer_> what type of speed up/through put do you expect? 14:31 < jrandom> i expect significant throughput improvements 14:32 < deer_> he expects a fine, he said 14:32 < deer_> for speeding 14:32 < deer_> ;-) 14:32 < jrandom> in addition, for small request/response connections, the latency will be dramatically reduced 14:32 < jrandom> (cut in half) 14:32 < deer_> wow 14:32 < deer_> is that using udp? 14:33 < jrandom> the new lib exposes all the neat tunable parameters for normal TCP stacks too, so apps will be able to tweak out their own setup 14:33 < jrandom> no dinoman, this works on top of i2p's I2CP 14:33 < deer_> wow x2 14:33 < jrandom> (though we'll be writing similar code in a month or so to get the UDP transport..) 14:34 < jrandom> but, well, we'll see. 14:34 < deer_> because...? 14:34 < jrandom> there's still a lot of work to do 14:34 < jrandom> because what? 14:34 < deer_> well, can't tcp do it as well? 14:35 < jrandom> oh, why we're going to go UDP? http://www.i2p.net/todo#transport 14:35 < deer_> I remember the same discussion on freenet too, but they sticked to tcp as yet 14:35 < jrandom> plus TCP is a general purpose streaming transport - we can dramatically simplify it, since we can put up with a certain degree of out of order 14:35 < deer_> not that all decisions they make are good ;-) 14:36 < jrandom> newsbyte: i've followed those discussions and we're going to go udp 14:36 < jrandom> (that doesnt mean freenet is wrong - they've got different constraints) 14:37 < deer_> i2p should not be compared too closely to freenet. They're very different technically. 14:37 < deer_> (or: they ARE wrong ;-) 14:37 < jrandom> i dont think their use of TCP right now is wrong, just as I dont think I2P's previous use of TCP is wrong. progress requires small steps 14:38 < deer_> newsbyte makes sure the meetings don't get too short 14:38 < jrandom> heh 14:38 < deer_> yeah, nothing worse then short meetings 14:38 < deer_> you can't eat all the popcorn and drink all the beer, then 14:38 < jrandom> ok, anything else on 3) Streaming lib ? 14:39 < jrandom> if not, 4) files.i2p 14:39 < deer_> I think we're cool 14:39 < deer_> well, I know I am 14:39 < deer_> ;-) 14:39 < deer_> and funny too 14:39 < deer_> most of the time 14:39 < deer_> and also annoying 14:39 < deer_> ;-) 14:39 < jrandom> well, i just wanted to point out files.i2p - a new search engine on i2p 14:40 < deer_> ah, I see 14:40 < deer_> I was hoping it would be about putting eepsites up 14:40 < jrandom> one interesting thing to note is that you can reach eepsites that aren't up anymore with it, since it caches 14:41 < deer_> does it cache everything? 14:41 < deer_> all searchengines thusfar are server-side? 14:41 < deer_> interesting. Shouldn't be too hard, these days :). 14:41 < jrandom> baffled: caches text/html from what i can tell 14:42 < deer_> at least it has limits on file size and types, so won't cache movies 14:42 < deer_> Auh, that's what I thought not binary. 14:42 < deer_> I mean, they are not in js, I suppose? 14:43 < jrandom> it uses nutch if anyone wants to look into it further. or i'm sure we'll get the site author to put up a feedback form or something ;) 14:43 < jrandom> newsbyte: correct, this is just a normal website hosted anonymously 14:43 < jrandom> the site contains a search engine (like google) 14:44 < jrandom> anyway, i just wanted to mention it 14:44 < jrandom> there have also been a lot of blogs popping up lately, which imho is really cool 14:44 < jrandom> my 'eep' bookmark folder almost fills a screen :) 14:44 < deer_> hehe, myi2p is happening all by itself :) 14:45 < jrandom> you just have to bring up the sore points, dont ya ragnarok? ;) 14:45 < deer_> sorry :) 14:46 < jrandom> ok, anyone have any questions/comments/concerns wrt files.i2p? 14:46 < jrandom> if not, let me move on to 4.1) biff 14:46 * jrandom almost forgot biff 14:46 < jrandom> postman, you arond? 14:47 < deer_> I think he's biffed up 14:47 < jrandom> well, if not, biff is this new kickass mail notification bot 14:47 < jrandom> if you've got an email acct at mail.i2p, you can tell biff to notify you when you get new mail 14:47 < deer_> does it has archives? 14:48 < jrandom> newsbyte: biff is just a notification bot, the mail is stored on the mail server (and accessed with your normal mail reader - kmail, etc) 14:48 < jrandom> see http://www.postman.i2p/ 14:49 < jrandom> ok, so, yeah, go to the eepsite or check out #mail.i2p over there 14:49 < deer_> I will, as soon as I get my eepsite on 14:49 * jrandom doesnt really know much more wrt biff - redirect any questions to postman 14:50 < jrandom> instead, we can move on to 5) ??? 14:50 < deer_> indeed 14:50 < jrandom> does anyone have anything else they want to bring up? 14:50 < deer_> * mule_iip raising hand to get voice: would like to recall my persistent FCP over I2P problems. but probably that can wait and will automagically be solved by 0.4.2. 14:50 < deer_> yes, and the freeze 14:50 < jrandom> i hope so mule_iip 14:50 < deer_> ok, will be your test platform :) 14:50 < jrandom> newsbyte: is there anything we need to discuss about it? could you just email me your logs if it happens again? 14:51 < jrandom> ooh mule, that'd rule 14:51 * jrandom will definitelytake you up on that 14:51 < deer_> well...can i still send those, if everything is frozen? 14:51 < jrandom> the files are written to disk. 14:51 < jrandom> when you restart, send me the logs 14:51 < deer_> I mean, in that case, I could send it now, since they should be somewhere 14:51 < jrandom> (please) 14:51 < deer_> i was in the forum and see that the jabber service is gone. was thaat of us to anyone if it was i would like to run one if it would be cool? 14:51 < jrandom> the files rotate though newsbyte 14:52 < jrandom> duck and demonic_1 have had jabber servers at various times, but it seems most of the i2p IM activity has been on irc 14:52 < deer_> the files rotate? surely it stores quite some data before it starts deleting? 14:53 < jrandom> newsbyte: ok, send me your logs, maybe it has something in it 14:53 < deer_> good 14:53 < deer_> ermm 14:54 < deer_> darn 14:54 < deer_> a lot of .logs 14:54 < deer_> ok 14:54 < deer_> a noob is never gonna follow this 14:54 < deer_> I guess you're right in not making a /. article yet 14:55 < jrandom> we're in no rush 14:55 < deer_> log-router.txt? 14:55 < jrandom> wrapper.log and logs/log-router-*.txt 14:56 < deer_> and the mailaddy to use would be...? 14:56 < deer_> dinoman, a jabber server would be cool imo 14:56 < jrandom> jrandom@i2p.net 14:56 < deer_> accessible by i2p, I hope? 14:56 < deer_> ;-) 14:56 < jrandom> newsbyte: you can put your logs on your eepsite and msg me the url 14:57 < jrandom> or you can send mail to jrandom@mail.i2p 14:57 < deer_> indeed! 14:57 < deer_> a good idea! 14:57 < deer_> there is only one little problem with it: It's not up yet 14:57 < jrandom> ok, anyone else have anything they want to bring up? 14:57 < jrandom> well, we can work on that newsbyte 14:57 < jrandom> (after the meeting) 14:59 < deer_> thnks, but whoo is already helping 14:59 < jrandom> if there's nothing else... 14:59 < deer_> we need a detailed howto/wiki/helpsite/something, though 14:59 * jrandom winds up 14:59 < deer_> i'd like to say, for the meeting, if a public release of i2p can be made before the u.s. election on november 2nd, this would go a long way to helping ensure a stable democracy 14:59 < deer_> what about 6)? 14:59 < jrandom> newsbyte: would you like to work on that? 15:00 < jrandom> newsbyte: i do agree it'd be great to get some more howtos and help info 15:00 < deer_> 6) There is no.... number 6 15:00 < deer_> well, yeah, sort of, but it's a strange thing, with me 15:00 < deer_> I'm pro-wiki and public thingy and free for everyone and all that 15:00 < deer_> but my ego protests and wants minimal control 15:00 < jrandom> great 15:00 < deer_> go figger 15:00 < jrandom> heh 15:01 < jrandom> well, if you'd like to make your own eepsite into a wiki you control, that'd be great too 15:01 < deer_> indeed 15:01 < jrandom> though ugha.i2p has a pretty good uptime 15:01 < deer_> I'll think about it 15:01 < jrandom> cool 15:02 < deer_> 6 would be the freenet-i2p thingy 15:02 * jrandom winds up 15:02 * jrandom *baf*s the meeting closed {% endblock %}