Timestamps are in GMT/BST.
[0:05] * IMCensored1 (n=KMIntern@) Quit ()
[0:16] * NullAcht15 (n=NullAcht@) Quit (Remote closed the connection)
[0:19] * IMCensored1 (n=KMIntern@) has joined #freenet
[0:19] * DebolazX (n=DebolazX@) has joined #freenet
[0:20] <CIA-14> mrogers * r11039 /trunk/apps/load-balancing-sims/phase7/sim/ (Node.java Peer.java Sim.java handlers/MessageHandler.java): Move token/backoff/throttle switches to the command line
[0:39] * PraiseChaos (n=kcecil@) has joined #freenet
[0:47] * Iceman_B (n=Ice@) Quit (Read error: 104 (Connection reset by peer))
[0:47] * Iceman_B (n=Ice@) has joined #freenet
[0:57] * Iceman_B (n=Ice@) Quit (Read error: 104 (Connection reset by peer))
[0:57] * Iceman_B (n=Ice@) has joined #freenet
[1:06] * nextgens sets mode -b *!*@83.71.214.43
[1:06] * nextgens sets mode -b ian*!*@83.71.214.43
[1:08] * drduck_ (n=drduck@) has joined #freenet
[1:10] * drduck (n=drduck@) Quit (Read error: 110 (Connection timed out))
[1:11] * PraiseChaos (n=kcecil@) Quit (Read error: 104 (Connection reset by peer))
[1:13] * PraiseChaos (n=kcecil@) has joined #freenet
[1:58] * mozillaman (n=chatzill@) Quit (Read error: 104 (Connection reset by peer))
[2:12] * OctobersDark_ (n=October@) has joined #freenet
[2:13] * OctobersDark (n=October@) Quit (Read error: 110 (Connection timed out))
[2:16] * toad_ (n=toad@) Quit (Remote closed the connection)
[2:33] * FrankBA_ (i=fbas@) Quit (Remote closed the connection)
[2:37] * FrankBA_ (i=fbas@) has joined #freenet
[2:38] * OctobersDark_ (n=October@) Quit (Read error: 110 (Connection timed out))
[2:39] * OctobersDark_ (n=October@) has joined #freenet
[2:53] * Siluro (n=PurplePu@) has joined #freenet
[2:53] * Siluro (n=PurplePu@) has left #freenet
[3:09] * OctobersDark_ (n=October@) Quit (Read error: 110 (Connection timed out))
[3:09] * OctobersDark_ (n=October@) has joined #freenet
[4:24] <sandos> nextgens, hehe, I see something very very similar to that on my general bandwidth graphs
[4:24] <sandos> (freenet is the only app Ive got running that uses any bw)
[4:24] <sandos> stopped at 4 this morning though and went down to low levels
[4:35] <sandos> went up to 1.2Mbit incoming/0.5 outgoing and then down to normal <0.2Mbit levels
[4:36] <sandos> maybe backoff started happening or something?
[4:43] * OctobersDark_ (n=October@) Quit (Read error: 110 (Connection timed out))
[4:44] * OctobersDark_ (n=October@) has joined #freenet
[4:45] * OctobersDark_ (n=October@) Quit (Client Quit)
[4:54] <sandos> Running threads: 504
[4:54] <sandos> alot of threads...
[4:56] <sandos> cpu-usage is way down though, for the period since my latest upgrad 1 day ago, and during the high bandwidth-usae
[5:03] * JustMe (i=JustMe_@) Quit ()
[5:06] * JaboH (n=Jabo@) Quit (Read error: 110 (Connection timed out))
[5:44] * christl_s (i=christin@) Quit (Read error: 110 (Connection timed out))
[5:46] * PraiseChaos (n=kcecil@) Quit ("Leaving")
[5:57] * freenut (n=freenut@) Quit ("Ex-Chat")
[6:31] * FrankBA_ (i=fbas@) Quit (Client Quit)
[6:36] * Q-collective (n=Q-collec@) has joined #freenet
[7:05] * Apophis2_ (n=Apophis@) Quit (Read error: 104 (Connection reset by peer))
[7:05] * xxxxx (n=Apophis@) has joined #freenet
[7:07] * _ph00 (n=z@) Quit ("Leaving")
[7:07] * railk (n=railk@) has joined #freenet
[7:14] * betatester (n=WeisBj02@) has joined #freenet
[7:15] <betatester> what do you guys think at what pace is freenet growing?
[7:16] <betatester> does anybody know if the freenet software is allowed in china or burma?
[7:16] <betatester> ur cuba?
[7:17] <betatester> i mean the use
[7:18] <betatester> anybody knows what the future planf for freenet are? is there anything to look forward to?
[7:19] <DebolazX> I don't think it would matter much in Cuba, since Internet isn't allowed there for the majority of the population.
[7:19] <betatester> some major change or big feature add ?
[7:19] <betatester> ok
[7:19] <betatester> question answered
[7:21] <betatester> is the development of freenet slowing down? the latest news are from august
[7:26] * christl_s (i=christin@) has joined #freenet
[7:26] * christl_s (i=christin@) Quit (Remote closed the connection)
[7:29] <Ralith> betatester, nope, development is still very active
[7:29] <Ralith> you can tell if you hang out here much
[7:29] <Ralith> lots of code being submitted
[7:33] * betatester (n=WeisBj02@) has left #freenet
[7:36] * Q-collective (n=Q-collec@) Quit (Read error: 104 (Connection reset by peer))
[7:49] * railk (n=railk@) Quit ("Cya, wouldn't want ta be ya!")
[7:51] * mazzanet (n=mazzanet@) Quit (Read error: 104 (Connection reset by peer))
[8:21] * whiterabbit (n=Whiterab@) has joined #freenet
[8:26] * ZwiSter (n=arne@) has joined #freenet
[8:26] * ZwiSter is now known as ZwiSter_away
[8:27] * timppax3m (n=timppa@) Quit (Read error: 110 (Connection timed out))
[8:46] * _ph00 (n=z@) has joined #freenet
[9:00] * grillide (n=hjubal@) has joined #freenet
[9:05] * pipex (n=pipes@) has joined #freenet
[9:05] * pipex (n=pipes@) has left #freenet
[9:19] * mazzanet (n=mazzanet@) has joined #freenet
[9:23] * ShipHead (i=ShipHead@) Quit (Connection reset by peer)
[9:28] * NullAcht15 (n=NullAcht@) has joined #freenet
[9:52] * christl_s (i=christin@) has joined #freenet
[10:16] * railk (n=railk@) has joined #freenet
[10:19] * HellViper (n=hellvipe@) has joined #freenet
[10:20] <HellViper> Hello, my node is updated to v999. My upload bandwith limit is set to 50K. My upload rate is actually 160K... Last night i uploaded 10.8G to one peer! Is the rate limiter broken?
[10:22] * Iceman_B (n=Ice@) Quit (Read error: 104 (Connection reset by peer))
[10:23] * Iceman_B (n=Ice@) has joined #freenet
[10:23] * lat (n=gerhard@) has joined #freenet
[10:35] * railk (n=railk@) Quit ("Cya, wouldn't want ta be ya!")
[10:44] * ShipHead (i=ShipHead@) has joined #freenet
[10:50] * Iceman_B (n=Ice@) Quit (Read error: 104 (Connection reset by peer))
[10:50] * Iceman_B^ (n=Ice@) has joined #freenet
[10:50] * Iceman_B^ (n=Ice@) Quit (Remote closed the connection)
[11:23] * mazzanet (n=mazzanet@) Quit ("Leaving")
[11:24] * mazzanet (n=mazzanet@) has joined #freenet
[11:32] * guruz (i=guruz@) Quit (Read error: 104 (Connection reset by peer))
[11:35] <fridim> HellViper, it seems there is a kind of loop making this bandwidth usage
[11:40] <HellViper> ftidim: i ask some of my peers. Most of these have the same problem. I hope it will be solved very soon... Thanks for your answer
[11:40] <fridim> Running threads: 1 503 wow!
[11:44] * mizery_guts (n=miz@) has joined #freenet
[11:44] <mizery_guts> anybody help in getting librarian plugin to run?
[11:45] <mizery_guts> I type in *@http://downloads.freenetproject.org/alpha/plugins/Librarian.jar.url into the plugin box and get nothing
[11:46] <mizery_guts> no errors in the Java console, just nothing
[11:47] * colione (n=colione@) Quit ("Leaving")
[11:51] * colione (n=colione@) has joined #freenet
[11:56] * toad_ (n=toad@) has joined #freenet
[11:56] * ChanServ sets mode +o toad_
[12:16] <nextgens> mizery_guts> you have to download it onto your hard drive and to start it from there
[12:16] <nextgens> mizery_guts> using *@file:///
[12:16] <nextgens> hey toad_
[12:18] <nextgens> toad_> it seems that we have got a problem with 999 : see the b/w graph fridim has drawn : http://fridim.org/~gcore/mrtg/127.0.0.1_1-day.png
[12:18] <nextgens> toad_> the b/w usage seems to increase with linearly uptime
[12:18] <nextgens> toad_> the b/w usage seems to increase with uptime linearly
[12:19] <nextgens> toad_> and we can notice an increase of the size of the unmatchedFIFO as well
[12:19] <nextgens> toad_> it might be related to Zothar's changes on the "matcher" loop
[12:19] <nextgens> I bet we aren't matching everything ... and we ask for ACKs
[12:20] <toad_> very possibly
[12:20] <toad_> is it staying within the limit?
[12:20] * JustMe (i=JustMe_@) has joined #freenet
[12:20] <nextgens> no
[12:21] <toad_> we also have another problem which prevents freesite inserts
[12:21] <toad_> related to USK insertion
[12:21] <nextgens> I'm not sure about that one
[12:21] <toad_> hmm?
[12:21] <nextgens> someone had the problem, I told him to switch his insert scheduler to the soft policy and it worked
[12:21] <toad_> that's bizarre
[12:21] <nextgens> meaning that he had higher priority jobs queued
[12:22] <nextgens> but it's true that manifest are often taking ages to insert
[12:22] <nextgens> ...
[12:22] <toad_> well we're getting reports of it from MANY places
[12:22] <toad_> and it's a critical bug if it's true
[12:23] <nextgens> I've been experiencing that since the beginning as far as I remember
[12:23] <nextgens> (the slowness of manifest insertion)
[12:23] <toad_> nice of you to tell me
[12:23] <toad_> it's not slow, it's broken
[12:23] <nextgens> :D
[12:23] * nextgens thinks it's slow
[12:24] <nextgens> maybe I should try to update my freesite to see
[12:25] <mizery_guts> thank you nextgens :)
[12:25] <mizery_guts> still cannot load the plugin
[12:25] <mizery_guts> but everything else works flawlessly so it is probably me putting in the wrong UNC name
[12:25] <nextgens> mizery_guts> where is the plugin located on you hard drive ?
[12:26] <mizery_guts> c:\Program Files\Freenet\plugins\Librarian.jar.url
[12:27] <mizery_guts> and I tried @file://localhost/c/Progra~1/Freenet/plugins/Librarian.jar.url
[12:28] <nextgens> that can't work : it would be :
[12:29] <nextgens> *@file:///c:/Progra~1/Freenet/plugins/Librarian.jar.url
[12:29] <nextgens> mizery_guts> try that one ^
[12:32] <mizery_guts> sorry that did not work
[12:32] * HellViper (n=hellvipe@) has left #freenet
[12:33] * HellViper (n=hellvipe@) has joined #freenet
[12:33] <nextgens> hmm
[12:33] <mizery_guts> I will redownload the jar file and try again
[12:33] * HellViper (n=hellvipe@) has left #freenet
[12:33] <nextgens> mizery_guts> try with "progam files" maybe too
[12:33] <nextgens> mizery_guts> or move the jar file to c:/Librarian.jar.url for testing purposes
[12:33] * HellViper (n=hellvipe@) has joined #freenet
[12:35] <JustMe> nextgens: If you have time look at USK@5QOjqouIpkngIq8RnzQeStvB3F-QKO9hUC2pnIoGZwY,zI0ngWE4OyOPbjF5FxAFxMrBaL-uw-LdqQxe3kP54Jk,AQABAAE/mytwocents/4/ for some information on the site insertion problem.
[12:37] <mizery_guts> *@file:///c:/Local/Librarian.jar syntax works :)
[12:37] <mizery_guts> Librarian plugin is loaded
[12:38] <nextgens> ok, so it's "program files" which matters
[12:39] <nextgens> because of the space I bet
[12:42] * edt (n=Ed@) has joined #freenet
[12:42] * ChanServ sets mode +o edt
[12:42] <mizery_guts> no biggie, I will load plugiins from a directory with no spaces :)
[12:43] <nextgens> there must be some way to escape it ... but I don't know how
[12:43] <nextgens> :$
[12:43] <nextgens> JustMe> well, frost slowness is something else ... it's by design :p
[12:43] <nextgens> JustMe> I would suggest him to use freesitemgr insteed
[12:44] <nextgens> JustMe> and to forward his feature requests to Bombe (regarding jSite) or to fill in a ticket on the BTS
[12:44] <nextgens> Bombe> they are some feature requests for jSite on USK@5QOjqouIpkngIq8RnzQeStvB3F-QKO9hUC2pnIoGZwY,zI0ngWE4OyOPbjF5FxAFxMrBaL-uw-LdqQxe3kP54Jk,AQABAAE/mytwocents/4/
[12:45] <JustMe> Where is freesitemgr?
[12:46] <nextgens> in the SVN repository
[12:46] <nextgens> it's a python script
[12:46] <nextgens> to automate "batch" inserts of given freesites
[12:46] <nextgens> https://emu.freenetproject.org/svn/trunk/apps/pyFreenet/
[12:47] * nextgens wonders if we have a tarball for it somewhere
[12:47] <nextgens> we do :)
[12:47] <nextgens> JustMe> http://downloads.freenetproject.org/alpha/pyFreenet/pyFreenet-9498.tbz
[12:48] <JustMe> Ok thanks for the info. I guess it (python) can be made to run on Windows?
[12:48] <nextgens> it ought to
[12:48] <nextgens> I dunno
[12:49] <JustMe> I'll look into it. Thanks!
[12:49] <mizery_guts> there is a windows port of python
[12:51] <JustMe> It's just that jSite used to work flawlessly. I don't know why some of us are having problems with it now.
[12:52] <nextgens> I'm going to fix the auto-builder on emu first
[12:53] <nextgens> then I might have a look at the b/w bug if toad_ hasn't fixed it in the meantime
[13:01] <nextgens> FreenetLogBot> c'mon
[13:03] <nextgens> toad_> we need to smatch that b/w bug soon, otherwise people are going to shutdown their nodes ...
[13:03] <nextgens> especially as it doesn't obey the limit
[13:13] * sbc (n=sbc@) has joined #freenet
[13:25] * ZwiSter_away (n=arne@) Quit (Remote closed the connection)
[13:31] <toad_> nextgens: yes
[13:32] * MikeW (n=em@) has joined #freenet
[13:51] <Zothar> JustMe, nextgens: that tarball is quite old 'cause most of the mirrors are out of date for pyFreenet; try this from one of the specific mirrors that's got it: ftp://ftpmirror.sectoor.de/freenet/alpha/pyFreenet/pyFreenet-11027.tbz.sha1
[14:03] <lat> I've problems with binding the port because of a router ... is "node.listenPort=14496" the correct entry in freenet.ini ... couldn't anything useful in the mailingslist except the mention of this entry
[14:04] <lat> the port is opened on the router and the ip forwarded
[14:04] * Iceman_B^Ltop (n=ice@) has joined #freenet
[14:05] <Zothar> that's your FNP port, yes; if you're running 999, it's mentioned directly at the bottom of the /darknet/ page
[14:05] * Iceman_B^Ltop (n=ice@) Quit (Client Quit)
[14:11] <lat> ok, didn't really care about it ... so the settings are ok ... will take a second look on it
[14:11] <Zothar> nextgens, toad_: I don't see how it the bandwidth thing could be related to my changes on UdpSocketManager.checkFilters() as all we're doing there is dropping packets in more circumstances; _unclaimed has always grown like that (and more actually because of the pings), it just wasn't easily visible; removing ping from BlockTransmitter _may_ have done it, but I don't see how unless other...
[14:11] <Zothar> ...parts of the code assume we have that much extra traffic since I know DMT.ping directly was only logged and nothing else
[14:12] * Iceman_B^Ltop (n=ice@) has joined #freenet
[14:13] * sopues (n=sandos@) has joined #freenet
[14:13] <nextgens> toad_> any chance you get around debugging that bug?
[14:13] <sopues> http://sandos.se/munin/sandserv/sandserv-ip_3478.html
[14:14] * nextgens has a lot of non-freenet related things to do today
[14:14] <sopues> do you know why the ups/downs happen yet?
[14:14] <Zothar> BTW, my bandwidth graphs don't demonstrate that behavior; I've only seen the instantaneous bandwidth numbers above the limit for seconds at a time in a oscilate high/low scenariou
[14:14] <sopues> ( I started logging after ~1d of uptime, when bw usage had just dropped dramatically)
[14:16] <sopues> oh, sorry.. its been up 1d now!
[14:16] <sopues> thats weird.. I think I saw the exact same uptime this morning!
[14:16] <sopues> unclaimedFIFOSize: 10332
[14:16] <sopues> whats that? =)
[14:18] <Zothar> what may be happening for the # Output Rate: 2.81 KiBps # Input Rate: 6.41 KiBps values is that the number of seconds of data they use is shorter than the life of the data the limiter averages to limit; it looks like others may be getting total session lifetimes over the limit (I don't know their limits), but I'm not seeing it
[14:18] * mizery_guts (n=miz@) Quit ("Peace and Protection 4.22.1")
[14:19] <Zothar> unclaimedFIFOSize is the size of an internal FIFO of messages that are unclaimed; it shouldn't get that big, except there's a bug; the stat was added to the stats page because the size of the FIFO can have a very big impact on performance (and so maybe it'll get more attention and the real problem get fixed :)
[14:20] <sopues> aha ok
[14:20] <sopues> mybe it goes up with the bw-usage?
[14:20] <sopues> my node is soon using 1Mbit.. higher than ever before, afaik :)
[14:21] <Zothar> actually, because of the performance impact, if it get's really high (it maxes at 50k), I'd expect the bandwidth to plateau or drop in the higher range; I don't think it's directly related to the bandwidth problem otherwise
[14:22] <sopues> it is claiming 51/102Kbit.. while munin/iptables is claiming 462/920 KBps... hmm
[14:22] * colione (n=colione@) Quit ("This computer has gone to sleep")
[14:22] <sopues> thats what you were talking about maybe?
[14:24] <sopues> 1,3GB in, 1,9GB out used during ~10 hours time
[14:24] <Zothar> so you're saying that the numbers the node is using don't match the real numbers? Are you looking at Output Rate/Input Rate? Note that those numbers are very simplisticly calculated, they're something like the last three seconds or something like that; I'm not familiar with munin; MRTG usually uses a 5 minute average
[14:24] <sopues> pretty good :)
[14:25] <sopues> yes, munin is also 5 minute average
[14:25] <sopues> but the node consistently claims way too low it seems
[14:25] <sopues> if this is all overhead it means the efficiency is ridiculously low!
[14:25] <Zothar> munin/iptables are for the whole computer or just the ports/traffic of Freenet on FNP?
[14:25] <Zothar> -s
[14:26] <Zothar> Freenet should know it's actual bandwidth usage on the FNP port
[14:29] <sopues> just the proper port
[14:31] <sopues> its also the only thing using bw, so I can clearly see it on my regular graphs for the entire interface...
[14:31] <sopues> Ive stopped running tor..
[14:31] <Zothar> note that the bandwidth limiter isn't as simplistic as the Output Rate/Input Rate on the stats page; have you installed the SNMP plugin and pointed an MRTG-like thing at it for graphing what the node thinks more concretely?
[14:31] <sopues> no, I ahve not
[14:32] <Zothar> does Total Output seem to mismatch what iptables tells you?
[14:35] <NullAcht15> In case anyone cares, here's another graph demonstrating erratic bw-usage in 999: http://freenet.thisisatest.de/freenettraffic.html
[14:35] <sopues> I dont think so, actually
[14:35] <sopues> it says 6GB for 24+19 hours
[14:35] <sopues> seems atleast about right
[14:38] <NullAcht15> (green: outbound, blue: inbound, red:total)
[14:45] * sbc (n=sbc@) Quit ("Ex-Chat")
[14:54] * railk (n=railk@) has joined #freenet
[15:09] * HellViper (n=hellvipe@) has left #freenet
[15:12] * Iceman_B^Ltop (n=ice@) Quit ("Swan is hotter than Jasmine ^_^ -=SysReset 2.53=-")
[15:15] * timmy2chk (n=violent@) has joined #freenet
[15:21] <CIA-14> zothar * r11040 /trunk/apps/pyFreenet/refbot.py: refbot: bot won't try to add its node's own ref
[15:22] * MarcelXitan (n=sfilippi@) has joined #freenet
[15:24] * MarcelXitan (n=sfilippi@) has left #freenet
[15:28] * another_user (n=Mathias@) has joined #freenet
[15:30] <another_user> Hi there! I had a freenode running before the darknets came, and I'm somewhat lost now. As I understand it, I have to impress someone, so they'll invite me, right?
[15:33] <another_user> In case anybody cares, my ID is: identity=tUjg5FhqKLONyZMOcI4CM29E5cSej6cHDasUQQrbfXc
[15:36] <Zothar> if you're running an 0.5 node, it should hopefully just get connections; if you're running an 0.7 node, just run over to #freenet-refs and find some people willing to trade refs with you
[15:37] <toad_> nextgens: any progress?
[15:37] <another_user> Ah, thanks for the info. I didn't know 0.5 should still be functional. I'll try that first.
[15:37] * timmy2chk (n=violent@) Quit ("so what?")
[15:37] * another_user (n=Mathias@) Quit ("Download Gaim: http://gaim.sourceforge.net/")
[15:38] <toad_> Zothar: any idea about the bandwidth bug? didn't you do some work on MessageFilter.match() ?
[15:39] <Zothar> hmm, lemme look at that one, since I saw match() and thought checkFilter()
[15:42] <Zothar> toad_: I check for a source match earlier and swapped the left and right side of a !=, so I don't think it should matter unless there's some reason I don't know why the source check was after the field iteration; here's what I did: http://emu.freenetproject.org/cgi-bin/viewcvs.cgi/trunk/freenet/src/freenet/io/comm/MessageFilter.java?rev=11017&r1=10490&r2=11017
[15:43] <Zothar> I think you'll find that reverting that commit won't affect unclaimedFIFOsize; it's affect on the bandwidth stuff would rather unexpected to me
[15:45] <toad_> and checkFilters()?
[15:47] <Zothar> checkFilters() is merely dropping stuff from _unclaimed in more circumstances and a logging tweak to match
[15:47] <Zothar> (the new circumstance is that the item has been in _unclaimed for an hour)
[15:48] <toad_> hmmm
[15:48] <toad_> and you're sure about the calculation of the time?
[15:50] <Zothar> rather sure, yes, since we don't get log messages of dropping after an hour before the node's been up an hour
[15:50] <nextgens> toad_> not yet :/
[15:50] <nextgens> toad_> and I need to go shopping before shops close :/
[15:50] <Zothar> 60*60*1000; // 1 hour
[15:50] <nextgens> the fridge is empty
[15:50] * timmy2chk (n=violent@) has joined #freenet
[15:51] <toad_> nextgens: ok
[15:51] <toad_> nextgens: i will investigate
[15:51] <Zothar> now - MAX_UNCLAIMED_FIFO_ITEM_LIFETIME
[15:51] <Zothar> m.localInstantiationTime < messageDropTime
[15:54] * railk (n=railk@) Quit ("Cya, wouldn't want ta be ya!")
[15:54] <toad_> it's probably related to the high overhead bug ...
[15:57] * Iceman_B^Ltop (n=ice@) has joined #freenet
[15:58] * OctobersDark (n=October@) has joined #freenet
[16:09] <toad_> ok, i doubt very much that the ping removal causes the current problems...
[16:12] <Zothar> good to hear
[16:17] <toad_> hmmm
[16:17] <toad_> nothing obvious in recent commits
[16:21] * MikeW (n=em@) Quit (Remote closed the connection)
[16:22] <Zothar> side effect of slightly better performance and we've had the problem for a long while?
[16:22] <toad_> Zothar: your MessageFilter changes look good
[16:22] <Zothar> ok, that's good
[16:22] <toad_> did you make any changes to USM.match?
[16:23] <toad_> yes, checkFilters
[16:23] * toad_ checks
[16:23] * Zothar reminds self
[16:23] * mozillaman (n=chatzill@) has joined #freenet
[16:23] <Zothar> right
[16:24] <rah> does anybody else have a problem with their node doing nothing and sucking 100% CPU?
[16:24] <rah> 981784fa71a93079cfcf98a9b91d3106d8b020
[16:24] <rah> Nov 24, 2006 16:18:32:976 (freenet.node.PeerManager, WrapperListener_start_runner, NORMAL): Added freenet.node.PeerNode@8cd72fb9@66.30.80.156:15245@e9584ccd620614b4091073e541fabca33833f2fea34e8eb93ac51720769e4cad
[16:24] <rah> Nov 24, 2006 16:18:32:977 (freenet.node.PeerManager, WrapperListener_start_runner, NORMAL): Read 29 peers from ./peers-43654
[16:24] <rah> that's the last thing mine said
[16:25] <Zothar> rah: what's your unclaimedFIFOsize on the /stats/ page? uptime?
[16:25] <rah> I can't get to the stats page
[16:25] <rah> the daemon isn't listening
[16:25] <rah> it hasn't gotten that far
[16:25] <toad_> stack dump?
[16:25] <Zothar> your node's apparently not starting up correctly for some reason then; what's the last few lines of your wrapper.log?
[16:26] <rah> INFO | jvm 59 | 2006/11/24 16:18:32 | Creating PeerManager
[16:26] <rah> INFO | jvm 59 | 2006/11/24 16:18:33 | Read 29 peers from ./peers-43654
[16:26] <rah> Connecting to proxy[172.18.111.16]:3128... connected.
[16:26] <rah> Proxy request sent, awaiting response... 503 Service Unavailable
[16:26] <rah> oops
[16:26] <rah> Connecting to localhost[127.0.0.1]:8888... failed: Connection refused.
[16:27] <Zothar> yeah, because it didn't get that far in the startup; can you produce a stack dump? On unix: run.sh dump
[16:28] <rah> STATUS | wrapper | 2006/11/24 16:28:30 | Dumping JVM state.
[16:28] <rah> heh
[16:28] <rah> that's it
[16:28] <toad_> Zothar: any ideas then?
[16:28] <rah> the node is quite dead
[16:28] <Zothar> rah: I've got the same problem when I try a stack dump; I haven't been able to find out why it won't dump
[16:29] <toad_> right now my node is respecting the bwlimit, and has 66% payload (which is good)
[16:29] <toad_> and 1144 unclaimed FIFO size
[16:29] <rah> I'd note that this didn't happen before updating to 999
[16:29] <toad_> rah: if it doesn't produce a stack dump when you ask it to, it's a JVM bug
[16:29] <toad_> rah: restart it
[16:29] * Zothar checks on what's after reading peers in the startup
[16:30] <rah> I have done
[16:30] <rah> it was in the same state this morning
[16:30] <toad_> get a new JVM then :)
[16:30] <rah> I restarted it, it came up with 100% CPU, and has obviously been there all day
[16:30] <toad_> seriously, stack dumping does not involve us at all
[16:30] <toad_> it's a JVM function, and if it's broken, it's definitely a JVM problem
[16:30] <toad_> post your entire wrapper.log somewhere
[16:31] <rah> freenet@teasel:~$ ls -lh wrapper.log
[16:31] <rah> -rw-r--r-- 1 freenet nogroup 1.7M Nov 24 16:31 wrapper.log
[16:31] <rah> you want that?
[16:31] <toad_> hmmm
[16:31] <toad_> email me it :)
[16:31] <toad_> gzip it first
[16:31] <rah> k
[16:32] <toad_> Running threads:?563
[16:32] <toad_> hmmm
[16:32] <toad_> that's a lot of threads
[16:32] <toad_> especially considering memory usage is pretty low
[16:33] <Zothar> "Initializing CHK Datastore" should be next output, so it could be several things in there
[16:33] <toad_> 366 of the 563 threads are FCP related
[16:33] <toad_> so logically it's Frost's fault
[16:33] <rah> 2006-11-24 16:34:04 1Gndzy-0007mY-24 => toad@amphibian.dyndns.org R=smarthost T=remote_smtp_smarthost H=smtp.tiscali.co.uk [212.74.114.61]
[16:33] <rah> 2006-11-24 16:34:04 1Gndzy-0007mY-24 Completed
[16:34] <toad_> okay, nothing to do right now
[16:34] <toad_> on that
[16:34] <Zothar> toad_: right on the JVM being the whole shebang on stack dumps; I think my problem started when I switched to Sun's 1.5 packaged by Debian sarge's java package builder
[16:35] <toad_> rah: wrapper.log doesn't include an attempt to get a stack dump
[16:36] <toad_> rah: actually the wrapper.log you sent me is way out of date, it only goes up to 2006/10/27 20:47:14
[16:36] <toad_> oh
[16:36] <toad_> i'm looking at the wrong file
[16:36] <toad_> doh
[16:37] <rah> heh
[16:37] <toad_> rah: looks broken
[16:37] <toad_> rah: what JVM?
[16:37] <rah> same as Zothar
[16:38] <rah> 1.5 made with make-jpkg
[16:38] <toad_> "Dumping JVM state." is what you get when you ./run.sh dump ?
[16:38] <rah> yes
[16:38] <Zothar> yes :)
[16:40] <toad_> hmmm
[16:40] <toad_> well it still looks like a problem on sun's part
[16:40] <toad_> incidentally i'm using 1.5 with make-jpkg
[16:41] * toad_ suggests you add logging to Node etc until you can isolate exactly where it hangs
[16:41] <toad_> in the meantime, i have more important (and more tractable!) bugs to track, which might actually be in our code
[16:42] <toad_> the main one being the bandwidth problem
[16:42] <toad_> which naturally doesn't show itself for me :|
[16:43] * timmy2chk (n=violent@) Quit (Read error: 104 (Connection reset by peer))
[16:43] * Iceman_B^Ltop (n=ice@) Quit ("Swan is hotter than Jasmine ^_^ -=SysReset 2.53=-")
[16:45] <mozillaman> toad_: Heh
[16:49] * Zothar updates from 1.5.0_07 to 1.5.0_09
[16:49] <toad_> good idea
[16:50] * toad_ is running 1.5.0_08
[16:50] <toad_> maybe it's a specific problem with _09 ?
[16:50] <Zothar> I'm running 07; what are you running rah? java -version
[16:50] <rah> 09
[16:51] <mozillaman> question
[16:51] <Zothar> bleah; well, I should try 08 then maybe
[16:51] <mozillaman> If I had a freesite
[16:51] <mozillaman> managed by Jsite
[16:51] <mozillaman> then copies over the inster nd requests keys for a backup
[16:51] <mozillaman> then lost jsite and reinstalled from scratch
[16:51] <mozillaman> and was at revision 21 for a ssk site
[16:51] <mozillaman> How do I get jsite to begin instering at 22?
[16:52] <mozillaman> Or, is my old freesite now toast?
[16:53] <Zothar> rah: is your wrapper.conf unmodified?
[16:53] <toad_> if you have the privkey, you should be able to insert
[16:53] <toad_> Zothar: 1.5.0_09 works?
[16:53] <rah> Zothar: erm yes
[16:53] <rah> (why would it be modified?)
[16:53] * timmy2chk (n=violent@) has joined #freenet
[16:54] <Zothar> toad_: dunno; I'm 07 and not working, you're 08 and working and rah is 09 and not working; based on the Java changelogs, I don't think it's the JVM version specifically
[16:55] <rah> I'm not sure why you think it's a JVM problem
[16:55] <rah> this never happened before 999
[16:55] <toad_> Zothar: did you install _09?
[16:55] <toad_> Zothar: does it fix it?
[16:56] <toad_> rah: it is a JVM problem that the JVM does not generate a stack dump
[16:56] <Zothar> not yet
[16:56] <rah> the problem isn't getting a stack trace
[16:56] <toad_> rah: without one, debugging it - whatever it is - is far more tedious
[16:56] <rah> indeed
[16:56] <toad_> although of course it may be possible
[16:56] <rah> the JVM may yet produce a stack dump
[16:56] <toad_> and since the JVM is broken, we have to assume that, well, the JVM is broken
[16:56] <toad_> rah: how long have you given it?
[16:56] <rah> if the freenet code stops taking 100% of the CPU
[16:57] <rah> toad_: not long enough
[16:57] <toad_> no, the JVM takes precedence
[16:57] <rah> it probably would not produce a backtrace within our lifetimes
[16:57] <toad_> it's quite capable of producing a stack dump while fred uses all the cpu
[16:57] <rah> because there's a bug in freenet
[16:57] <toad_> and most likely the cpu usage is from java anyway not from fred
[16:57] <rah> ok..
[16:57] <toad_> a bug in freenet that just happens to coincide with a bug in the jvm, yeah, right
[16:58] <toad_> anyway, i've explained how you can narrow down exactly where it breaks with lots of logging
[16:58] <toad_> for me to do that remotely would be much slower than for you to do it
[16:58] <rah> s/bug in/problem with/
[16:58] <toad_> if i could replicate it here then i could debug it here
[16:59] <toad_> i can give you jars with more and more logging in to find out how far it's got
[16:59] <toad_> but that will take ages
[16:59] <toad_> maybe setting log level to MINOR would help
[17:04] <Zothar> rah: I don't think your not getting a stack dump is related to your high CPU usage
[17:04] <Zothar> I'm currently downloading 09 to try it
[17:07] * timmy2chk (n=violent@) Quit ("so what?")
[17:10] <mozillaman> toad_: So, jsite will detect the correct revisition to insert into?
[17:10] <mozillaman> err, revision
[17:11] * hopeatikari (n=hopeatik@) has joined #freenet
[17:12] <hopeatikari> hi all
[17:16] <Zothar> darn, no change on the stack dump front
[17:16] <Zothar> toad_: paste your wrapper.conf somewhere so I can compare?
[17:16] <toad_> mozillaman: it should
[17:17] <toad_> mozillaman: if it doesn't, just keep inserting until it does
[17:18] <mozillaman> toad_: k :)
[17:18] * mozillaman wants to bring FreeNetWatch back now that he has more time
[17:19] <mozillaman> Have I mentioned jsite is slow yet today?
[17:19] <mozillaman> If not, let me correct that error :)
[17:20] * mozillaman needs to remove some of the archives and link to them as previous revisions to make inserting quicker
[17:20] * JacknJill (n=vengeanc@) has joined #freenet
[17:22] <JacknJill> can anyone tell me how to get "on" the freenet?
[17:22] * mozillaman shaved .12kb of the freenet bunny logo :)
[17:22] <mozillaman> JacknJill: Sure
[17:22] <mozillaman> JacknJill: Go to http://freenet.sf.net
[17:22] <mozillaman> download 0.7 from the download page
[17:22] <mozillaman> install it
[17:23] <mozillaman> then
[17:23] <mozillaman> go to #freenet-refs and get some references
[17:23] <mozillaman> After you have three you're "on" freenet :)
[17:23] <JacknJill> okay. thanks
[17:24] * JacknJill (n=vengeanc@) has left #freenet
[17:26] * hopeatikari (n=hopeatik@) Quit ("Advanced IRC 2.0.4 - don't ask, don't tell")
[17:29] <mozillaman> networkSizeEstimateSession: 85 nodes
[17:32] * Q-collective (n=Q-collec@) has joined #freenet
[17:32] * sich (n=sich@) has joined #freenet
[17:34] * tsukuba1971_Berl (n=chatzill@) has joined #freenet
[17:35] <tsukuba1971_Berl> hi
[17:42] * rah_ (n=rah@) has joined #freenet
[17:46] <mozillaman> toad_: jSite has been running on a 20kb site for 1hour exaclty with 6 connected peers.
[17:47] <mozillaman> toad_: At this point, I think something is up :P
[17:49] * timmy2chk (n=violent@) has joined #freenet
[17:57] * mozillaman thinks he's lost his site ;(
[17:57] <Zothar> silly human
[17:59] * rah (n=rah@) Quit (Read error: 110 (Connection timed out))
[17:59] <mozillaman> Zothar: You will be assimilated....resistance is futile.
[17:59] <Zothar> eh
[18:02] <mozillaman> http://en.wikipedia.org/borg
[18:02] <Zothar> right
[18:02] <Zothar> that was more of a "don't care" expression... :)
[18:03] <mozillaman> ah
[18:03] <toad_> mozillaman: known bug
[18:03] <mozillaman> toad_: Is there a work-around?
[18:04] <toad_> insert as SSK?
[18:04] * mozillaman notes FreeNetWatch needs more news:
[18:04] <mozillaman> http://127.0.0.1:8888/freenet:USK@jotJldLVFPDEnvRqfhBWsnXPQpOS~QrawxFjgsLZcFQ,xnNqE4Z~zMHmIUmqrA0oziUFSXNOAC7OhOOH4yhcBq4,AQABAAE/freenetwatch/22/
[18:04] <toad_> i dunno i should fix it
[18:04] <mozillaman> toad_: Please do! :P
[18:04] * mozillaman tries as SSK
[18:05] <toad_> it's just that i should fix the bandwidth bug too
[18:05] <mozillaman> toad_: Gah
[18:05] <toad_> and i have no idea how to
[18:05] <mozillaman> toad_: How am I going to keep FreeNetWatch updated :'(
[18:05] * toad_ will have a look in a bit
[18:05] <toad_> mozillaman: insert it as an SSK
[18:05] <toad_> mozillaman: SSK@.../freenetwatch-22
[18:05] <toad_> mozillaman: tell me if that owrks
[18:06] <toad_> mozillaman: use the console if you need to
[18:06] <mozillaman> k, but it half-inserted
[18:06] <mozillaman> http://127.0.0.1:8888/freenet:USK@jotJldLVFPDEnvRqfhBWsnXPQpOS~QrawxFjgsLZcFQ,xnNqE4Z~zMHmIUmqrA0oziUFSXNOAC7OhOOH4yhcBq4,AQABAAE/freenetwatch/22/
[18:06] <mozillaman> half didn't.
[18:06] * mozillaman tries ssk insert....
[18:07] * JustMe says at least it's not JustMe ;)
[18:09] * PraiseChaos (n=kcecil@) has joined #freenet
[18:10] <Bombe> Wow. After like six weeks of uptime my node ended itself with a deadlock.
[18:10] <Bombe> Time to update, finally. :)
[18:10] <sandos> my node has gone crazy, I think: http://sandos.se/munin/sandserv/sandserv-ip_3478.html
[18:11] <sandos> http://sandos.se/munin/sandserv/sandserv-cpu.html
[18:11] <sandos> Running threads: 1,415
[18:13] <sandos> btw, I am not running a _09 jvm, fyi
[18:13] <sandos> I have to check what my bandwidth-limits ae
[18:13] <toad_> sandos: interesting
[18:13] <toad_> sandos: stack dump would be interesting, if you can get one
[18:13] <sandos> sure, I can try
[18:14] <sandos> my bandwidth limit is 200000, its probably not reached that quite yet.. :)
[18:16] <sandos> where is my stackdump supposed to end up now?
[18:16] <sandos> aha, on console it seems
[18:17] <sandos> http://sandos.se/~sandos/thread_dump_01
[18:18] <sandos> all my stats: http://pastebin.ca/raw/256826
[18:20] <toad_> ok thanks
[18:29] <sandos> oh, I guess my node is hittig my bw-limit: Output Rate: 214 KiBps
[18:29] <sandos> that is a first for sure
[18:29] <mozillaman> PUTSSKDIR:*********************************SSKKEY******************************************#freenetwatch#index.html
[18:29] <mozillaman> What
[18:29] <mozillaman> did I do wrong?
[18:29] * mozillaman isn't used to the telnet
[18:29] <mozillaman> PUTSSKDIR:AK5***********************Bq4#freenetwatch#index.html
[18:32] <sandos> toad_, I also have a small snippet of "DEBUG" log
[18:32] <sandos> "small" meaning 19MiB
[18:33] <sandos> wooha
[18:33] <sandos> are packets buffered very much or what? I see VERY long sequences of "PS Sending: ..." lines in the log =)
[18:33] <sandos> like, a few hundred packets sent at once
[18:34] <sandos> toad_, give me a privmsg if you want a peek at that log
[18:52] * _ph00 (n=z@) Quit ("Leaving")
[18:53] * _ph00 (n=z@) has joined #freenet
[18:54] * grillide (n=hjubal@) Quit ("..3 2 1 ???check inignition and may God's love be with you???..")
[19:02] * MikeW (i=Mike@) has joined #freenet
[19:08] <nextgens> hi
[19:09] <nextgens> Bombe> can you provide details about the deadlock please ? do you have a wrapper.log describing it ?
[19:09] <nextgens> toad_> any progress regarding the b/w usage going crazy bug ?
[19:10] <nextgens> unclaimedFIFOSize: 11560
[19:10] <nextgens> my input rate is going crazy
[19:10] <nextgens> I receive more than 30 times what I send
[19:10] * timmy2chk (n=violent@) Quit ("so what?")
[19:12] <sandos> I think I am punding my neighbours pretty good :)
[19:12] <sandos> pounding*
[19:12] <sandos> peers, even
[19:12] <Zothar> I limited one of my nodes to 10K, which is below what it's been using and the node seemed to honor it, though it was limiting it to about 7K, so not sure what was up there unless a payload overhead of 70% had something to do with it
[19:17] * mozillaman (n=chatzill@) Quit ("Chatzilla 0.9.76 [Firefox 2.0/2006101023]")
[19:20] <sandos> yeah, mine seem to obey my 200K limit
[19:22] * mozillaman (n=chatzill@) has joined #freenet
[19:23] <toad_> nextgens: i haven't found any obvious code changes, nor have i been able to replicate it locally
[19:23] <toad_> nextgens: so no
[19:23] <toad_> :(
[19:23] <mozillaman> Completed 0% 0/2 (failed 0, fatally 1, total 2)
[19:23] <mozillaman> Is that bad?
[19:23] <toad_> mozillaman: fatally 1 is bad yes
[19:23] <nextgens> toad_> I've experienced it once, I restarted and it has disapeared
[19:23] <mozillaman> Will it continue, or, should I start again?
[19:23] * Qt-collective (n=Q-collec@) has joined #freenet
[19:23] <nextgens> toad_> it might be related to MTU changes too
[19:24] <nextgens> toad_> if you get around to it, I suggest you have a look to the unclaimedFIFO size
[19:24] <toad_> mozillaman: you should probably restart, but it would be interesting to see the resulting error message
[19:24] <nextgens> there is definitly a problem with it
[19:24] <toad_> nextgens: 7600 here
[19:24] <nextgens> unclaimedFIFOSize: 10727
[19:24] <nextgens> nodeUptime: 6h21m
[19:24] <toad_> it looks like there may be several issues here
[19:24] <nextgens> # Output Rate: 3.50 KiBps
[19:24] <nextgens> # Input Rate: 46.5 KiBps
[19:24] <toad_> sandos for example has 1400 threads
[19:24] <nextgens> my outgoing payload is fine
[19:25] <mozillaman> Finished insert but: Insert collided with different, pre-existing data at the same keyURI would have been: freenet:SSK@jotJldLVFPDEnvRqfhBWsnXPQpOS~QrawxFjgsLZcFQ,xnNqE4Z~zMHmIUmqrA0oziUFSXNOAC7OhOOH4yhcBq4,AQABAAE/freenetwatch/
[19:25] <nextgens> I'm below 500 here
[19:25] <mozillaman> toad_: ^
[19:25] <toad_> mozillaman: it should be SSK@blah/freenetwatch-<edition number>
[19:25] <mozillaman> Oh :P
[19:25] * mozillaman fixes
[19:26] <mozillaman> one dash or two?
[19:26] <mozillaman> /freenetwatch-23
[19:26] <mozillaman> or
[19:26] <mozillaman> /freenetwatch--23
[19:26] <nextgens> one
[19:27] * mozillaman retries
[19:27] <mozillaman> thanks
[19:27] <toad_> mozillaman: one dash
[19:28] <mozillaman> thx
[19:28] <toad_> tell me if it works
[19:28] <mozillaman> freenet:SSK@jotJldLVFPDEnvRqfhBWsnXPQpOS~QrawxFjgsLZcFQ,xnNqE4Z~zMHmIUmqrA0oziUFSXNOAC7OhOOH4yhcBq4,AQABAAE/freenetwatch-23/
[19:28] * mozillaman hugs toad_ !
[19:28] <mozillaman> Work perfectly :)
[19:29] <toad_> ok
[19:29] <toad_> now try inserting it as a USK
[19:29] <mozillaman> FreeNetWatch is saved! :P
[19:29] <mozillaman> From commandline
[19:29] <mozillaman> or jsite?
[19:29] <toad_> try inserting it as a USK
[19:29] <toad_> same way you inserted the SSK
[19:29] <mozillaman> k ;)
[19:30] <mozillaman> freenetwatch/24
[19:30] <mozillaman> or just -1?
[19:31] <toad_> USK@.../freenetwatch/23/
[19:31] <mozillaman> k
[19:35] * Q-collective (n=Q-collec@) Quit (Read error: 110 (Connection timed out))
[19:35] * rah_ is now known as rah
[19:35] <toad_> hi rah
[19:36] <rah> hey
[19:36] * rah looks over there
[19:36] <rah> :)
[19:37] <CIA-14> zothar * r11041 /trunk/apps/pyFreenet/ (botplugin.py.dist refbot.py): refbot.py can now use a pre-add and post-add plugin. Also, it closes the FCP connection more cleanly.
[19:38] <Zothar> darn, calling getAllStackTraces() didn't give me a dump either; wish I knew why
[19:38] <Zothar> ...so it's not the wrapper, unless there's a setting that turns it off in general
[19:39] <toad_> Zothar: so is it a _09 issue or not a _09 issue?
[19:39] <Zothar> toad_: I don't think it has anything to do with 09 or 07; one thing I haven't tried is 08, but the changelog from 08 to 09 was very short
[19:42] <toad_> Zothar: so regardless of JVM in use, you still can't get a stack dump?
[19:42] <toad_> Zothar: that's bizarre
[19:42] <toad_> Zothar: did you give it a minute or so to generate it?
[19:42] <Zothar> yes, that's it; bizarre
[19:42] <Zothar> oh, waiting isn't the issue, I'm pretty sure of that :)
[19:42] <Zothar> I guess I'll try 08, just because
[19:43] <toad_> well if it doesn't produce a stack dump then i very much doubt it's a problem with our code
[19:43] <mozillaman> toad_: Took aot longer, but it worked :)
[19:43] <toad_> on the other hand if it persists over many JVM versions it's probably not a JVM problem
[19:43] <toad_> so it's wierd
[19:43] <toad_> mozillaman: hmmm
[19:43] <Zothar> I would agree that it's very much not likely to be a problem with our code
[19:43] <toad_> mozillaman: that's very strange
[19:44] <mozillaman> That it worked?
[19:44] <toad_> mozillaman: because many many people are reporting that USK site inserts aren't working
[19:44] <Zothar> I suppose I could try it without starting from the wrapper to see how that differs; will try 08 first
[19:44] <toad_> unless it's actually site inserts, not USKs, that have the bug
[19:45] <mozillaman> toad_: I could use some new for freenetwatch BTW
[19:45] * blaaaah (n=blah@) has joined #freenet
[19:45] <blaaaah> hey, how come pircbot is banned?
[19:45] <sandos> toad_, do you want to have a peek at my debug-level log snippet?
[19:45] <toad_> sandos: good idea
[19:46] <toad_> sandos: what is it supposed to show?
[19:46] <sandos> well, it does show packet-sender sending huge amounts of packets all at once
[19:46] <nextgens> blaaaah> where ?
[19:46] <sandos> not sure if thats expected with this behavious were seeing anyway
[19:46] <blaaaah> freenet-refs
[19:46] <nextgens> because we don't want it to be logged ?
[19:47] <blaaaah> riiiight, but there's a million ways around that
[19:47] * timmy2chk (n=violent@) has joined #freenet
[19:47] <nextgens> and ?
[19:47] <blaaaah> and there's bots in there trading references
[19:47] <blaaaah> so they are capable of logging anyway
[19:47] <nextgens> still, what would be the purpose of allowing it ?
[19:48] * nextgens is even half inclined to ban *!*@*.demon.co.uk]
[19:48] <blaaaah> i see your point, i just don't see how banning the ident pircbot is going to work :/
[19:48] <blaaaah> fine ban me
[19:48] <nextgens> nah, I'm just jocking
[19:48] <blaaaah> ~_~
[19:49] <nextgens> but the question stands : why would you want to log it ?
[19:50] <blaaaah> i dont
[19:50] <blaaaah> im saying that people can log it
[19:50] <nextgens> "known issue"
[19:50] <blaaaah> so banning one small logging program seems a bit ineffective
[19:50] * NullAcht15 (n=NullAcht@) Quit (Nick collision from services.)
[19:50] * NullAcht15_ (n=NullAcht@) has joined #freenet
[19:51] <nextgens> I saw one pirbot there, and have banned it, that's all
[19:51] <blaaaah> ah ok
[19:53] <nextgens> mozillaman> I got edition 21 from here
[19:53] <mozillaman> nextgens: It should be 23
[19:53] <mozillaman> Which, has yesterday's post
[19:53] * nextgens tries to grab it
[19:54] <nextgens> hmm
[19:54] <nextgens> the CSS is missing
[19:54] <mozillaman> Which #?
[19:54] <sandos> my node is (trying) to send 10581 packets during one second according to my debug log =)
[19:54] <nextgens> #22
[19:54] <nextgens> #23 is fine
[19:54] <nextgens> I got it
[19:55] <mozillaman> :)
[19:55] <mozillaman> 22 was done with jsite
[19:55] <mozillaman> which gave up half way through :P
[19:56] * egon2003 (n=fargod@) has joined #freenet
[20:03] * colione (n=colione@) has joined #freenet
[20:06] * oierw (n=oierw@) has joined #freenet
[20:07] * rah_ (n=rah@) has joined #freenet
[20:07] * Qt-kollective (n=Q-collec@) has joined #freenet
[20:09] * [MineHaunter] (n=van85@) has joined #freenet
[20:09] * MineHaunter (n=van85@) Quit (Nick collision from services.)
[20:09] * [MineHaunter] is now known as MineHaunter
[20:11] <mozillaman> nextgens: Any news item ideas?
[20:15] * mozillaman needs news!
[20:17] <nextgens> what about reading the changelog ?
[20:18] <nextgens> toad_> it might be related to the massive update of freenet-ext.jar too
[20:18] <nextgens> I don't see how though
[20:18] <nextgens> but the timeframe would match
[20:18] <nextgens> Zothar> do you have up to date wrapper libraries ? does the node complain at startup?
[20:19] * Qt-collective (n=Q-collec@) Quit (Read error: 110 (Connection timed out))
[20:19] <egon2003> # Output Rate: 28.1 KiBps
[20:19] <egon2003> # Input Rate: 46.8 KiBps
[20:19] <egon2003> over double the speed with the last build :)
[20:19] <Zothar> nextgens: yes, they match the ones in my freenet-ext.jar after self-building them earlier today
[20:19] <egon2003> i run 999
[20:20] <Zothar> (no compaints at startup after updating the bin/wrapper lib/wrapper.jar and lib/libwrapper.so
[20:20] <Zothar> )
[20:20] * rah (n=rah@) Quit (Read error: 110 (Connection timed out))
[20:20] <Zothar> mozillaman: compared to a month or two ago or whenever you had your last news item, there should be _tons_ of news by now :)
[20:21] <mozillaman> I know
[20:21] * mozillaman just needs to find it :P
[20:22] <Zothar> release messages on the mailing lists might be a good start; check the archives
[20:24] * MikeW (i=Mike@) Quit (Read error: 113 (No route to host))
[20:26] <sandos> egon2003, yes, something is wrong ;) (probably)
[20:30] <egon2003> perhaps :)
[20:33] * Zorix_ (n=Brandon@) has joined #freenet
[20:34] * Zorix_ is now known as Zorix
[20:41] * kcecil_ (n=kcecil@) has joined #freenet
[20:41] * PraiseChaos (n=kcecil@) Quit (Read error: 104 (Connection reset by peer))
[20:41] * kcecil_ is now known as PraiseChaos
[20:52] * ferdsio (i=ferdsio@) has joined #freenet
[20:54] * whiterabbit (n=Whiterab@) Quit ("KVIrc 3.2.5 Anomalies http://www.kvirc.net/")
[21:02] * blaaaah (n=blah@) Quit ()
[21:06] * ferdsio (i=ferdsio@) Quit ()
[21:24] * rah_ is now known as rah
[21:32] * TheShado (i=no@) Quit ("I am what I am and I do what I can.")
[21:40] * Mikus (i=Mikus@) has joined #freenet
[21:40] * Mikus (i=Mikus@) Quit (Client Quit)
[21:44] * oierw (n=oierw@) Quit (Read error: 104 (Connection reset by peer))
[21:44] * oierw` (n=oierw@) has joined #freenet
[21:45] * oierw` is now known as oierw
[21:54] <toad_> hmmm
[21:54] <toad_> Frost has 318 sockets open to the node on FCP
[21:55] * burby (n=chicken@) has joined #freenet
[21:55] <toad_> which the node is servicing using 408 threads
[21:59] * makomk (n=aidan@) Quit (Remote closed the connection)
[22:01] * oierw (n=oierw@) Quit (Read error: 104 (Connection reset by peer))
[22:01] * oierw (n=oierw@) has joined #freenet
[22:05] <burby> i have tried to connect to my new 0.5 install, on a new machine, and it does not work. I have not used freenet in a long time, so I may be missing something. Is there a 0.5 link that should work so that I can test my install?
[22:05] <toad_> 167 _senderThread's and 211 Sender threads
[22:05] <toad_> the individual transfers however don't seem to be blocked by the bwlimiter ...
[22:06] <toad_> the packets are allowed immediatelyt
[22:06] <toad_> as the transfer rate isn't high enough for them to be a problem
[22:08] * burby (n=chicken@) Quit (Remote closed the connection)
[22:08] * tsukuba1971_Berl (n=chatzill@) Quit (Read error: 104 (Connection reset by peer))
[22:09] <toad_> HRRRRM
[22:09] <toad_> we got an InsertReply despite never receiving the full data (and therefore never sending it)
[22:09] * burby (n=chicken@) has joined #freenet
[22:10] <burby> anyone?
[22:16] * IMCensored1 (n=KMIntern@) Quit (Read error: 110 (Connection timed out))
[22:18] * IMCensored1 (n=KMIntern@) has joined #freenet
[22:18] <_ph00> burby; yes, there are ppl on line
[22:19] <_ph00> og, OK, I found the question
[22:19] <_ph00> ok*
[22:19] <_ph00> hm... .5 is not supported any more
[22:20] <_ph00> and I, persnoally, never used it, so I can't help, not even "unofficially"
[22:20] <_ph00> [tip: get .7]
[22:21] * egon2003 (n=fargod@) Quit ("using sirc version 2.211+KSIRC/1.3.12")
[22:28] * oierw` (n=oierw@) has joined #freenet
[22:31] * oierw`` (n=oierw@) has joined #freenet
[22:34] * Aladin (n=lars@) has joined #freenet
[22:38] * burby (n=chicken@) Quit (Remote closed the connection)
[22:40] * toad_ hmmmm
[22:40] * toad_ thinks there's a bug causing many of the transfer threads to stick around
[22:40] * toad_ plans to swat it
[22:40] <toad_> munch it
[22:40] <toad_> devour it
[22:41] * [JAFNFARPWNRDN] (n=GeneralB@) has joined #freenet
[22:41] <edt> payload percentage dropped from 67% to 36% here? Is this typical? What has happened?
[22:41] <toad_> woah, this transfer started at 16:45 and is still going at 22:40
[22:41] * burby (n=chicken@) has joined #freenet
[22:42] <toad_> edt: is output bandwidth usage really high?
[22:42] <edt> no
[22:42] <toad_> edt: i have heard of this bug, but i haven't been able to produce it here consistently
[22:42] <toad_> nor find any obvious cause in the code
[22:42] <burby> _ph00: Thanks, I will try it, I just did not want to go through all the hassle of setting up contacts.
[22:42] <edt> its never very high - its about 13k/s which is more than it has been for a while (limit is _much_ higher)
[22:43] <[JAFNFARPWNRDN]> interesting bookmarks any1 ?
[22:43] * oierw (n=oierw@) Quit (Read error: 110 (Connection timed out))
[22:43] * oierw`` is now known as oierw
[22:43] <_ph00> burby; if you wait for version .8 you won't need to manually add peers any more (chice between opennet and daknet)
[22:44] <_ph00> darket = manually adding peers; opennet= auto adding
[22:44] <burby> when is that suppossed to be testing?
[22:44] <_ph00> darknet is way safer, I think it's worth the hassle of manually adding peers
[22:44] <_ph00> I don't know nothing about .8 yet, I'm no developer
[22:45] <_ph00> I'm only telling you what I 'heard' here
[22:45] * Qt-collective (n=Q-collec@) has joined #freenet
[22:45] <burby> but, if i dont know anyone on darknet, then its whoever posts on the other channel
[22:45] * Webbi (n=chatzill@) has joined #freenet
[22:46] <_ph00> burby; yes I get your point
[22:46] <_ph00> I reason like this
[22:46] <_ph00> I go to #freenet-refs hang around and get to know ppl there
[22:46] <_ph00> and here
[22:47] * Webbi (n=chatzill@) has left #freenet
[22:47] <_ph00> I connect to those I "think" I can trust, and keep hanging around there and randomly chat
[22:47] <burby> i can see what happens i guess.
[22:47] <_ph00> in the eventuality of freenet getting banned, I *will* have a bunck of *known* perople all around the world that I can connect to
[22:48] <burby> can i connect with some people and remove them later
[22:48] <_ph00> OK, I still can't be 100% safe that they ar3e who they say they are
[22:48] * [JAFNFARPWNRDN] (n=GeneralB@) has left #freenet
[22:48] <_ph00> but that's anyways better than having my node "yelling" that there is a freenet node here, and any peers are welcome
[22:49] <_ph00> burby; yes, you can remove peers
[22:49] * oierw` (n=oierw@) Quit (Read error: 110 (Connection timed out))
[22:49] <_ph00> that's one of the grat things about darknet, you control who your node connects with
[22:49] <_ph00> great*
[22:57] * colione (n=colione@) Quit (Read error: 110 (Connection timed out))
[22:57] * Qt-kollective (n=Q-collec@) Quit (Connection timed out)
[22:59] * Zothar (n=Zothar@) Quit ("Chatzilla 0.9.74 [Firefox 1.5.0.8/2006102516]")
[22:59] <sandos> toad_, if you want you can connect to me and see what packets I send you.. or something ;)
[22:59] <sandos> heh, my node is not limited by my bw-limit but rather by my cpu it seems
[23:02] <_ph00> when a node is limited by the box' cpu, does a system monitor show 100% cpu usage all (or most of) the time?
[23:03] <_ph00> (my gnome system monitor applet shows a cpu usage around 30% on average)
[23:03] * rguerra (n=rguerra@) has joined #freenet
[23:03] <_ph00> no, low average. say around 20-22%
[23:04] * rguerra (n=rguerra@) Quit (Client Quit)
[23:04] <_ph00> lower than 30, I mean
[23:04] * rguerra (n=rguerra@) has joined #freenet
[23:04] * TheShado (i=no@) has joined #freenet
[23:05] <CIA-14> toad * r11042 /trunk/freenet/src/freenet/io/xfer/BlockTransmitter.java: Minor indent fix.
[23:06] * rguerra (n=rguerra@) Quit (Client Quit)
[23:08] * sandos (n=sandos@) Quit (Read error: 54 (Connection reset by peer))
[23:08] * sandos (n=sandos@) has joined #freenet
[23:08] * Qt-collective is now known as Q-collective
[23:10] * colione (n=colione@) has joined #freenet
[23:10] <CIA-14> toad * r11043 /trunk/freenet/src/freenet/io/comm/UdpSocketManager.java: Logging
[23:11] <CIA-14> toad * r11044 /trunk/freenet/src/freenet/io/xfer/BlockReceiver.java:
[23:11] <CIA-14> Cancel the PRB when we are disconnected in the middle of a transfer.
[23:11] <CIA-14> This bug was causing leaked sender threads.
[23:12] <CIA-14> toad * r11045 /trunk/freenet/src/freenet/node/NodeDispatcher.java: Logging
[23:12] <CIA-14> toad * r11046 /trunk/freenet/src/freenet/node/PeerNode.java: Throw a NotConnectedException so that the situation can be properly handled.
[23:13] * burthold10 (n=adfd@) has joined #freenet
[23:14] * burthold10 (n=adfd@) has left #freenet
[23:14] <sandos> I should probably try updating ;)
[23:15] <toad_> well, that wasn't the bug i was after
[23:15] <toad_> but it was a real, and long standing bug
[23:16] <sandos> leaking threads sounds a bit bad, atleast
[23:17] <sandos> uhm..
[23:17] <sandos> "freenet.config.InvalidConfigValueException: Cannot update because not running under wrapper"
[23:17] <sandos> but I am!
[23:20] <sandos> started at r11039
[23:21] <sandos> so whats changed in the config :)
[23:22] <sandos> blah, why arent you using viewSVN :(
[23:23] <mozillaman> New freenetwatch released, get it while it's hot!
[23:23] <mozillaman> http://127.0.0.1:8888/freenet:USK@jotJldLVFPDEnvRqfhBWsnXPQpOS~QrawxFjgsLZcFQ,xnNqE4Z~zMHmIUmqrA0oziUFSXNOAC7OhOOH4yhcBq4,AQABAAE/freenetwatch/26/
[23:23] <sandos> hmm, weird
[23:24] <mozillaman> sandos: It can't be that bad :(
[23:24] <sandos> no, I guess not
[23:25] <mozillaman> There is only one thing not in freenet's favor
[23:25] <mozillaman> It's Java...memeory, corporate, buggy, ugly, horrid, huge, hard to configure Java.
[23:35] * Rudolf_ (i=Rudolf_H@) has joined #freenet
[23:39] * paulux (n=chatzill@) has joined #freenet
[23:39] * sich (n=sich@) Quit ("Quitte")
[23:40] <sandos> blah, I cant get my wrapper to work
[23:41] <paulux> Do someone know how many connection do we need to have a reactive network ?
[23:45] <_ph00> depending on your connection speed, some 10 connected or more
[23:46] * Asc_ (n=asc@) has joined #freenet
[23:47] * NullAcht15_ (n=NullAcht@) Quit (Remote closed the connection)
[23:48] <Asc_> Is there anywhere 0.5 noderefs can be downloaded, these days?
[23:49] <paulux> newby question : what is a 0.5 noderef ?
[23:49] <sandos> :)
[23:50] <sandos> Oh well, restarted node. Lets see how it behaves during the night
[23:50] <sandos> seems fine now
[23:51] <Asc_> paulux: Version 0.5 needs a file containing information about nodes (seednodes.ref) in order to connect.
[23:52] <paulux> ok thanks ...
[23:55] * burby (n=chicken@) Quit (Remote closed the connection)
[23:58] <Asc_> Ah, found it. The download server's address has changed.
Irc logs of #freenet : 2008 2007 2006 2005
These logs were automatically created by FreenetLogBot on chat.freenode.net using the Java IRC LogBot.
