Timestamps are in GMT/BST.
[0:00] <mYone_> "2/18/0/0/0" :(
[0:01] <toad_> Cooo: disk full?
[0:03] <toad_> mYone_: is that the dominant reason for rejection?
[0:03] <toad_> mYone_: if the pInstantReject is less than 50% or so, don't worry about it
[0:05] <mYone_> toad_: its the only reason for rejection... currently "112505 >threadLimit"
[0:06] <mYone_> toad_: with 1075 i at least had a few rejections because of ping and/or bandwith limit... but now its only threads
[0:08] <mYone_> toad_: and "backedOffPercent: 71.1%" seems to be higher too... but maybe i'm only imagining things (paranoia?) ;)
[0:09] <toad_> mYone_: hmmm that's worrying
[0:09] <toad_> mYone_: pInstantReject?
[0:11] <toad_> mYone_: do you have queued local requests?
[0:11] <mYone_> toad_: currently "pInstantReject: 31.2%" but its fluctuating a lot ...
[0:11] <toad_> mYone_: any chance of a thread dump?
[0:12] <toad_> also a copy of the statistics page
[0:12] <toad_> please?
[0:12] * oogrr (n=aa@) has joined #freenet
[0:13] <mYone_> toad_: yep, coming... no i dont have queued requests (if you are refering to downloads in the global queue... only frost is running)
[0:13] <toad_> it's not downloading anything though?
[0:14] <toad_> well Frost uses the global queue now doesn't it? so no...
[0:14] <mYone_> toad_: nope
[0:16] <toad_> mYone_: is this a fairly heavy duty node?
[0:17] <toad_> 120KB/sec upload limit!
[0:18] <mYone_> toad_: i've got the bandwith available, so why not let freenet use it? but most of the time it doesnt use it
[0:18] <nextgens> now it will :)
[0:18] <mYone_> toad_: currently: Output Rate: 36.6 KiB/sec (of 120 KiB)
[0:20] <mYone_> toad_: i guess thats connected to the fact that most of the time the thread limit is reached... when i increased the thread limit to 750 i got the feeling that a bit more bandwith was used
[0:20] <mYone_> but thats just me guesssing ^^
[0:20] * Shadow_Vixen (n=panther@) Quit ("Leaving")
[0:20] <nextgens> mYone_> we have several load-limiting metrics
[0:21] <nextgens> the amount of free threads is one
[0:21] <nextgens> the bandwidth is an other one
[0:21] <nextgens> if you give it too much bandwidth for that metric to be useful, then it's normal that other "causes" will trigger the load-limiting
[0:22] <toad_> yeah, that's what it looks like
[0:22] <toad_> if it's using 1/4 of the available bandwidth presumably output liability isn't an issue
[0:25] <mYone_> anyway, currently ive got a very nice "pInstantReject: 3.5%" , so it doesnt seem to be that bad ^^ but still >60% backedoffpercent
[0:28] <pheenode> what could be wrong if I'm trying to connect one node to another with an opennet connection, the node that I'm telling to connect is sending packets which are arriving at the other end
[0:29] <pheenode> but the other end isn't sending anything back
[0:29] <pheenode> where should I look for indications of what's wrong?
[0:31] <Tommy[D]> pheenode, did both refs contain an ip?
[0:31] <pheenode> both?
[0:31] <pheenode> do I need to add refs on both sides to form an opennet connection?
[0:32] <Tommy[D]> connecting does only work, if 1 adds the ref of 2 and 2 adds the ref of 1
[0:32] <Tommy[D]> its the same for both opennet and darknet
[0:32] <pheenode> oic
[0:32] <pheenode> then what's the difference between opennet and darknet?
[0:32] <pheenode> if I need to add refs on both sides?
[0:33] <nextgens> FreenetLogBot> c'mon!
[0:33] <pheenode> will an opennet node automatically learn about other opennet nodes on the network once it gets a connection up?
[0:33] <mYone_> pheenode: yep, exactly
[0:33] <pheenode> mYone_: I see
[0:34] <pheenode> so opennet functionality is vastly different to what it was like in 0.5
[0:34] <pheenode> where you could add new nodes to the network without any manual interaction on the other nodes
[0:34] <mYone_> pheenode: everything is vastly different than in 0.5 ;)
[0:35] <Tommy[D]> pheenode, that should also work some time in the future with 0.7, its just not implemented atm
[0:35] <mYone_> pheenode: but as far as i know an automatic bootstrapping for opennet is planned (so that you wont need to make the initial manually)
[0:35] <pheenode> Tommy[D]: I see, ok
[0:36] <pheenode> mYone_: yes I know it's pretty much a complete remake
[0:37] * hrabal (n=captain@) has joined #freenet
[0:40] <pheenode> ok so I need to add noderefs on both ends in opennet mode
[0:40] <pheenode> and I got one connection up
[0:41] <pheenode> but does it take time to learn about other opennet nodes?
[0:41] <pheenode> or is that just not implemented yet
[0:41] <Tommy[D]> it takes some time to learn about other opennet nodes
[0:42] <mYone_> pheenode: if you have made the first few conections by hand freenet will automatically add new connections/peers. but yes, it will take some time
[0:42] <mYone_> ;)
[0:42] <Tommy[D]> and network activity, no requests on freenet => no new opennet peers
[0:42] <pheenode> oh yeah
[0:42] <pheenode> its the same with 0.5
[0:42] * hrabal (n=captain@) Quit (Read error: 104 (Connection reset by peer))
[0:42] <pheenode> I remember you had to run frost or something
[0:43] <pheenode> to get your node to actually make some connections
[0:45] <pheenode> will it start learning about other nodes with only one connection?
[0:45] <pheenode> or will it need to have a few before it starts learning?
[0:46] <mYone_> pheenode: i THINK it should work with only one, but it could take a bit longer... (thats me guessing again ;)
[0:47] <Tommy[D]> pheenode, only 1 is needed, but can of course need some more time as there are less connections for learning
[0:49] * earthsound (n=another1@) has joined #freenet
[0:49] * hrabal (n=captain@) has joined #freenet
[0:54] * oogrr (n=aa@) Quit ()
[1:09] * FrinkC|off is now known as FrinkC
[1:09] * skollll (n=chatzill@) has joined #freenet
[1:15] * skollll (n=chatzill@) Quit ("ChatZilla 0.9.79 [Firefox 2.0.0.10/2007111504]")
[1:15] * Hayona (i=lol@) Quit (Read error: 104 (Connection reset by peer))
[1:20] * mYone_ is now known as mYone
[1:32] * who9 (n=non@) Quit ()
[1:34] * toad_ (n=toad@) Quit (Remote closed the connection)
[1:47] * edt (n=Ed@) has joined #freenet
[1:47] * ChanServ sets mode +o edt
[1:47] <edt> anyone else seeing 1076 using excessive cpu?
[1:55] <FrinkC> yes, now i see it too. i have the same problem
[1:59] * Insurgent (n=kkllkj@) has joined #freenet
[2:02] <FrinkC> edt, what os you use?
[2:11] * mYone (i=schlepto@) Quit ("leaving")
[2:22] * Insurgent (n=kkllkj@) Quit (Read error: 110 (Connection timed out))
[2:22] * saces (n=saces@) Quit (Read error: 110 (Connection timed out))
[2:24] <FrinkC> edt, update your node to 1078
[2:27] * mYone (i=schlepto@) has joined #freenet
[2:34] * Tarz12 (n=test@) has joined #freenet
[2:35] <Tarz12> can anyone tell me how frost works...
[2:35] * Tarz12 (n=test@) Quit (Client Quit)
[2:35] * Tarz12 (n=test@) has joined #freenet
[2:37] * Tarz12 (n=test@) Quit (Client Quit)
[2:54] * NEOatNHNG_ (n=NEOatNHN@) has joined #freenet
[3:08] * earthsound_ (n=another1@) has joined #freenet
[3:10] * earthsound (n=another1@) Quit (Read error: 104 (Connection reset by peer))
[3:10] * earthsound_ is now known as earthsound
[3:11] * NEOatNHNG (n=NEOatNHN@) Quit (Read error: 110 (Connection timed out))
[3:21] * earthsound (n=another1@) Quit ("ChatZilla 0.9.79 [Firefox 2.0.0.10/2007111504]")
[3:23] * gasi____ (n=chatzill@) has joined #freenet
[3:23] * gasi____ is now known as gasi
[3:25] * sleon (n=sleon@) Quit (Remote closed the connection)
[3:42] * gasi___ (n=chatzill@) Quit (Read error: 101 (Network is unreachable))
[3:53] * hrabal (n=captain@) Quit ()
[4:30] * Orakio (n=hah@) has joined #freenet
[4:30] * friday (n=hah@) Quit (Read error: 104 (Connection reset by peer))
[4:38] * Wolf9696 (i=Wolf9696@) has joined #freenet
[4:38] * magu42 (n=martin@) Quit ("Verlassend")
[4:38] <Wolf9696> Anybody alive tonight?
[4:41] * Wolf9696 (i=Wolf9696@) has left #freenet
[4:58] * FrinkC is now known as FrinkC|off
[5:15] * sleon (n=sleon@) has joined #freenet
[5:15] * ChanServ sets mode +o sleon
[5:23] * ed_ (n=Ed@) has joined #freenet
[5:23] * edt (n=Ed@) Quit (Read error: 110 (Connection timed out))
[6:38] * bunnywabbits (n=bunnywab@) has joined #freenet
[6:54] * bunnywabbits (n=bunnywab@) has left #freenet
[7:21] * FrinkC|off is now known as FrinkC
[7:59] * lepoulpe303 (n=LePoulpe@) has joined #freenet
[7:59] <lepoulpe303> hi all
[7:59] <lepoulpe303> after a unwanted computer shutdown , my node doesn't wan't to restart
[8:01] <lepoulpe303> ./run.sh start says that node is already started (but nothing visible through ps aux commande)
[8:01] <lepoulpe303> ./run.sh restart starts in infinite loop "waiting for freenet 0.7 to exit"
[8:05] * edt (n=Ed@) has joined #freenet
[8:05] * ChanServ sets mode +o edt
[8:05] <lepoulpe303> ok finally fixed ; sry for troubles
[8:05] * ed_ (n=Ed@) Quit (Read error: 110 (Connection timed out))
[8:08] * lepoulpe303 (n=LePoulpe@) has left #freenet
[8:22] * edt (n=Ed@) Quit (Read error: 110 (Connection timed out))
[8:22] * edt (n=Ed@) has joined #freenet
[8:22] * ChanServ sets mode +o edt
[8:42] * edt (n=Ed@) Quit (Read error: 110 (Connection timed out))
[8:47] * Gudihl (n=Gudihl@) has joined #freenet
[9:04] * edt (n=Ed@) has joined #freenet
[9:04] * ChanServ sets mode +o edt
[9:08] * akw21 (i=uoyf@) has joined #freenet
[9:10] * akw21 (i=uoyf@) Quit (Client Quit)
[9:21] * edt (n=Ed@) Quit (Read error: 110 (Connection timed out))
[9:27] * TheSeeker (i=Fridlekh@) has joined #freenet
[9:29] <TheSeeker> looks like there was another couple of version bumps after the last sync with googlecode :/
[9:52] <Bombe> Yes, we had quite some activity yesterday.
[10:07] * invalidcom (n=in_valid@) has joined #freenet
[10:12] * Hayona (i=lol@) has joined #freenet
[10:18] * saces (n=saces@) has joined #freenet
[10:24] * Hayona (i=lol@) Quit (Remote closed the connection)
[10:45] * invalidcom (n=in_valid@) Quit ()
[10:45] <mYone> I have no idea what you have done, or if you have done anything at all, but 1078 seems much smoother than 1076! "backedOffPercent: 31.8%" and only "16896 >threadLimit" after 8 hours (not like more than 200k after one hour with 1076 ;). Thx!
[10:47] <dbkr> mYone: ooi, are your bandwidth limits set quite high?
[10:52] <mYone> dbkr: yes
[10:52] <dbkr> then you were probably getting the zombie insert senders bug that I was getting
[10:53] <dbkr> except that for you it was causing it to hit the thread limit rather than the bandwidth limit
[10:55] <mYone> dbkr: hm , no idea ^^ but its much better now with 1078, so everything's cool ;)
[10:55] <dbkr> on another note, does mantis really not have a search deature, or am I just blind?
[10:55] <dbkr> *feature
[10:56] * freenet32 (n=50aa247a@) has joined #freenet
[10:56] * freenet32 (n=50aa247a@) Quit (Client Quit)
[10:56] * luelue (i=luelue@) has joined #freenet
[10:58] <dbkr> nm, it's just poorly located :/
[11:09] * toad_ (n=toad@) has joined #freenet
[11:09] * ChanServ sets mode +o toad_
[11:21] <mYone> toad_: hi toad ! :)
[11:22] <mYone> toad_: there seems to be a problem with persistent downloads in the global queue (see frost post in freenet board)
[11:31] <nextgens> hmm
[11:31] <nextgens> it sounds like I screwed up
[11:38] <toad_> nextgens: and clobbered everyone's global queues? :(
[11:40] <nextgens> toad_> I still don't know why
[11:40] <nextgens> toad_> and btw, as you might have noticed, reports are very helpful, no logs, nothing, just flames
[11:41] <toad_> nextgens: 16117?
[11:41] <nextgens> I don't think it's releated
[11:43] <toad_> well you have to look at what changes you actually did
[11:43] <toad_> it's likely to be the result of one of them :)
[11:43] <nextgens> I've just fixed it
[11:43] <nextgens> on 16118
[11:44] <nextgens> let me verify I didn't make the same mistake somewhere else before you release :)
[11:44] <nextgens> toad_> btw, it's a win* only issue
[11:44] <nextgens> hence we didn't notice it: "worksforme"
[11:44] <toad_> okay, are there any other critical bugs that can be easily fixed before we release 1079?
[11:45] <toad_> the bandwidth bug can't be fixed until somebody gets it and sends us *REAL* logs with logleveldetails=freenet.node:MINOR or ideally loglevel=MINOR
[11:45] <toad_> bandwidth bug: output rate well over the limit, payload way down (<20%)
[11:45] <toad_> if anyone gets it please contact us and send us logs etc
[11:46] * dbkr is on the lookout for it
[11:48] <toad_> okay well we can't delay 1079 for it
[11:50] * toad_ won't tell them it was your bug, your reputation is bad enough already :)
[11:51] <nextgens> toad_> edontcare :)
[11:52] <nextgens> toad_> as I said, it works for me and my code is clean and works on proper OSes
[11:52] <nextgens> it's not my fault if their crappy os can't move files around when file descriptors are open :)
[11:52] <toad_> :)
[11:52] * dionyziz (n=dionyziz@) has joined #freenet
[11:52] <mYone> nextgens: do you think this was only an windows issue? i had a restarting download in global queue with linux !
[11:52] <toad_> fair enough, I suspect somebody will post this conversation to frost :)
[11:53] <nextgens> doh, you did release before my fix
[11:53] <dionyziz> Hi guys, I just inserted a CHK -- where can I find the key?
[11:53] <nextgens> toad_> we need 1081 afaic
[11:53] <nextgens> I told you to wait :)
[11:53] <toad_> nextgens: for what?
[11:53] <nextgens> [11:44] <@ nextgens> | let me verify I didn't make the same mistake somewhere else before you release :)
[11:53] <toad_> nextgens: you fixed the downloads didn't you?
[11:53] <nextgens> have a look to r16120 :)
[11:54] <toad_> okay, so that's 1080
[11:54] <toad_> do you want me to wait?
[11:55] <nextgens> downloads not resuming is the visible face of the iceberg
[11:55] <toad_> is there any more that needs to be done before 1080?
[11:56] <dionyziz> sorry for repeating my question, I think I didn't phrase it properly. I went to the "Queue" page of my fproxy and inserted a file from my local computer as "CHK". I waited for it to upload, and now it's done. However, I don't see the file anywhere. Is there any way to get a link to the CHK file I just inserted? How?
[11:56] <toad_> i haven't inserted 1080...
[11:56] <toad_> dionyziz: it's on the page
[11:56] <toad_> dionyziz: there should be a link
[11:57] <dionyziz> toad_: I only see "Completed downloads to temp" and "Downloads in progress". It was in "Uploads in progress" before, but now it's gone.
[11:57] <nextgens> I need to go, bbiab
[11:57] <nextgens> but yes, we will need 1080
[11:57] <nextgens> not that I'm aware of
[11:57] <nextgens> well, testing it maybe
[11:57] <nextgens> but I really need to go
[11:57] <toad_> it couldn't be any worse, i'll release it
[11:59] <toad_> folks, we need some testing of 1080: does it clobber your global queue? :|
[12:00] <mYone> toad_: how can i test? just run updater?
[12:01] <dionyziz> toad_: I'm probably missing something obvious, but I only see completed downloads to temp and downloads in progress. The uploads in progress section that I was seeing before is now gone. The file is kinda big so it took some time -- a few hours. Any clues on what I should do?
[12:01] <toad_> dionyziz: no idea
[12:01] <toad_> dionyziz: did you delete it from the queue?
[12:01] <mYone> dionyziz: your problem COULD be related to the current bug?
[12:01] <toad_> how exactly did you upload it?
[12:02] <toad_> did you restart afterwards?
[12:02] <toad_> are you running Windows?
[12:03] <dionyziz> toad_: no, I just waited. I went to the "Queue" section, selected the "CHK" radio box, browsed for my file, clicked on "Insert File". It took some time to load the page. Then it showed my file in the "uploading" section. I restarted my browser and it was still there. I did not restart my machine. I'm running windows, yes. I did not delete it from the queue, I just refreshed the page and now it's gone.
[12:03] * nextgens (n=nextgens@) Quit (Remote closed the connection)
[12:03] <toad_> dionyziz: that's bizarre
[12:03] <toad_> dionyziz: double check that it hasn't restarted - check the node uptime on the stats page
[12:04] <toad_> if it had restarted then it's possible it could be caused by the bug that was just fixed
[12:05] <dionyziz> toad_: how do I view the uptime of my node? there's no uptime in the "Statistics" page, only Node Version Information, JVM Info, Current Activity, Peer statistics, and Bandwidth.
[12:05] <toad_> nodeUptime ?
[12:05] <dionyziz> it's not there, I even tried Ctrl + F in firefox
[12:05] <toad_> under Node Status overview ?
[12:05] <toad_> okay, turn on advanced mode on the config page
[12:05] <toad_> then try
[12:05] <dionyziz> OK
[12:06] <dionyziz> toad_: apparently it has restarted -- uptime is 2 hours only. I don't know why though, I did not restart it manually. Could it have restarted itself?
[12:06] <toad_> yes, it happens sometimes
[12:06] <dionyziz> will I have to reinsert then?
[12:07] <toad_> due to new builds, if you have auto-update-fire enabled
[12:07] <toad_> or due to the node crashing
[12:07] <toad_> yes it looks that way, sorry
[12:07] * phrosty (i=phrosty@) Quit ("He who asks a question is a fool for five minutes; he who does not ask a question remains a fool forever.")
[12:07] <dionyziz> I have auto update enabled
[12:07] <toad_> well that would be why then
[12:07] * luelue (i=luelue@) Quit (Read error: 110 (Connection timed out))
[12:07] <dionyziz> thanks a lot for your help! it's appreciated :) I'll try reinserting now
[12:07] <toad_> result of restarting and the filebug in 1076-1079
[12:08] <toad_> gone in 1080
[12:08] <toad_> you might want to wait until 1080 has been deployed successfully before trying again
[12:08] <toad_> 1079 should be downloadable now, it's actually 1080
[12:08] * nextgens (n=nextgens@) has joined #freenet
[12:08] * ChanServ sets mode +o nextgens
[12:08] <toad_> so your node should auto-update to 1080 in the next few minutes
[12:08] <toad_> nextgens: rehi
[12:09] * luelue (i=luelue@) has joined #freenet
[12:12] * Shadow_Vixen (n=panther@) has joined #freenet
[12:12] * NEOatNHNG_ (n=NEOatNHN@) Quit (Read error: 104 (Connection reset by peer))
[12:12] <mYone> i've upgraded to #1080... after restarting it still seems to restart the download in the global queue (?)
[12:14] <mYone> INFO | jvm 4 | 2007/11/30 13:03:15 | Deleted 13312 temp files (0 non-temp files in temp dir)
[12:15] * Hayona (i=lol@) has joined #freenet
[12:15] <toad_> ok 1080 is inserted
[12:15] <toad_> mYone: that may happen, 1076-1079 broke it, 1080 can't recover from it because the data isn't there
[12:16] <toad_> mYone: sorry ... but 1080 shouldn't have the same problem if you change something and restart it
[12:16] <mYone> toad_: ok, trying again (restarting)
[12:16] <toad_> also w.r.t. downloads, you should have almost all the blocks in cache
[12:16] <toad_> so it should be able to get somewhere with it reasonably quickly
[12:18] <toad_> nextgens: emu is down
[12:18] <mYone> toad_: dont worry, i am not complaining ;) When i heard of the bug in frost i even restarted deliberatly to see if it is true
[12:18] <toad_> nextgens: hmmm it's back up again now, wierd
[12:22] * Nogaso (n=Nogaso@) has joined #freenet
[12:23] * NEOatNHNG (n=NEOatNHN@) has joined #freenet
[12:31] <saces> toad_: (r16121) uploads in progress seems ok, done uploads are lost after restart.
[12:33] <toad_> saces: hmmm?
[12:33] <toad_> saces: even AFTER upgrading?
[12:33] <toad_> i mean, upgrade, add an upload, it completes, restart, it's not there?
[12:35] <toad_> saces: hmmm?
[12:36] <saces> the uploads (3) was added on 1079, restart/upgrade to r16121 the clompeted one was lost, the 2 others was still running (restored seems ok), after all done restart again all was lost
[12:37] <toad_> when you say all, you mean completed uploads, or everything?
[12:37] <toad_> 16121 = 1080
[12:38] <saces> all uploads was completed = all are lost.
[12:39] <toad_> gah
[12:39] <toad_> but it's only completed uploads that are lost? what about in-progress uploads?
[12:39] <saces> I have only the test-uploads running.
[12:39] <saces> the in progress seems ok.
[12:40] <toad_> ok
[12:40] <toad_> add an upload. wait for it to finish. wait a couple of minutes and see whether it's written a new downloads.dat.gz. open it up (un-gzip it), and see if there's any mention of the completed request.
[12:40] <nextgens> well, only inserts *started* with an up to date node can be resumed
[12:40] <toad_> this is bizarre because that code hasn't been touched afaics
[12:41] <nextgens> check the size of the file
[12:41] <nextgens> if it grows, it's good
[12:41] <nextgens> toad_> I would have made it mandatory earlier on
[12:42] <toad_> afaics, he upgraded, the two running ones were preserved, the completed ones dropped... the two running ones completed, then he restarted, then they were gone
[12:42] <toad_> saces: that's right?
[12:42] <saces> yes.
[12:42] <toad_> okay, so 1080 is still dropping them. i have no explanation for this since that code wasn't touched
[12:42] <nextgens> anything in logs ?
[12:42] <toad_> but see if they are in the downloads.dat.gz
[12:43] <nextgens> I might close the descriptors not in order
[12:43] <toad_> nextgens: why would that selectively lose stuff that has been completed already?
[12:43] <toad_> well can you handle this? i need to bbiab
[12:44] <toad_> release 1081 when you're sure it's fixed
[12:45] <saces> dowloads.dat and queue page says : 1 done, 1 in progress.
[12:47] * edt (n=Ed@) has joined #freenet
[12:47] * ChanServ sets mode +o edt
[12:47] <edt> I noticed something on the new connect page
[12:48] <edt> I have no IPV6 here yet opennet is trying to connect to ipv6 sites
[12:48] <edt> these always fail
[12:48] <saces> restarted: the done is lost, the incomplete falls back from 21% to 15% (should be one or two chunks)
[12:50] <nextgens> edt> that's a feature, not a bug
[12:50] <nextgens> saces> is any exception raised in the logfiles ?
[12:51] * atomicplayboy (n=kesakuu@) Quit (calvino.freenode.net irc.freenode.net)
[12:51] * MrDomino (i=steve@) Quit (calvino.freenode.net irc.freenode.net)
[12:51] * HarryR (n=harryr@) Quit (calvino.freenode.net irc.freenode.net)
[12:51] * darth (n=klghd@) Quit (calvino.freenode.net irc.freenode.net)
[12:51] * FuriousRage (n=XPPRO@) Quit (calvino.freenode.net irc.freenode.net)
[12:51] * geeku (n=geeku@) Quit (calvino.freenode.net irc.freenode.net)
[12:51] * hramrach (i=hramrach@) Quit (calvino.freenode.net irc.freenode.net)
[12:51] * Tommy[D] (i=tsbnc@) Quit (calvino.freenode.net irc.freenode.net)
[12:51] * Shadow_Vixen (n=panther@) Quit (calvino.freenode.net irc.freenode.net)
[12:51] * luelue (i=luelue@) Quit (calvino.freenode.net irc.freenode.net)
[12:51] * Jflesch (n=jflesch@) Quit (calvino.freenode.net irc.freenode.net)
[12:51] * geeo (n=geeop@) Quit (calvino.freenode.net irc.freenode.net)
[12:51] * FrinkC (n=FrinkC@) Quit (calvino.freenode.net irc.freenode.net)
[12:51] * |R (i=bob@) Quit (calvino.freenode.net irc.freenode.net)
[12:51] <nextgens> grep for FCPServer
[12:52] * _dd_ (i=dima@) has joined #freenet
[12:52] * _dd (i=dima@) Quit (Read error: 104 (Connection reset by peer))
[12:52] * Shadow_Vixen (n=panther@) has joined #freenet
[12:52] * luelue (i=luelue@) has joined #freenet
[12:52] * Jflesch (n=jflesch@) has joined #freenet
[12:52] * FrinkC (n=FrinkC@) has joined #freenet
[12:52] * |R (i=bob@) has joined #freenet
[12:52] * geeo (n=geeop@) has joined #freenet
[12:52] * MrDomino (i=steve@) has joined #freenet
[12:52] * atomicplayboy (n=kesakuu@) has joined #freenet
[12:52] * Tommy[D] (i=tsbnc@) has joined #freenet
[12:52] * HarryR (n=harryr@) has joined #freenet
[12:52] * FuriousRage (n=XPPRO@) has joined #freenet
[12:52] * darth (n=klghd@) has joined #freenet
[12:52] * hramrach (i=hramrach@) has joined #freenet
[12:52] * geeku (n=geeku@) has joined #freenet
[12:53] * ChanServ sets mode -o Tommy[D]
[12:54] <saces> http://dark-code.bulix.org/denh4o-63107
[12:54] <nextgens> okay
[12:54] <edt> nextgens how is sending to an address that can never work a feature?
[12:55] <edt> or were you joking?
[12:55] * Nogaso (n=Nogaso@) Quit (Remote closed the connection)
[12:56] <edt> it should be easy enough to notice that all sends to ipv6 sites fail and so to have the node stop trying to use ipv6 addresses...
[12:57] <Bombe> Why?
[12:57] <Bombe> Just let it try, no harm done.
[12:57] <Bombe> Mostly this is done because there's no way to detect whether we have IPv6 connectivity or not.
[12:57] <edt> why keep and attempt to use addresses that cannot work - here there are some ipv6 address with 1000s of attempts
[12:58] <nextgens> saces> update
[12:58] <edt> Bombe if all sends to ipv6 fail its a good indication that there is not ipv6
[12:58] <Bombe> edt, maybe we could put in a config option to skip IPv6.
[12:58] <edt> or that
[12:58] <Bombe> edt, not at all. It could also means that your ISP currently sucks.
[12:58] <edt> Bombe mainly by dsl router
[12:59] <Bombe> If all IPv4 connects within a certain time frame fail, do we assume we don't have IPv4? :)
[12:59] <edt> s/by/my/
[12:59] <nextgens> saces> please update and keep me posted
[13:00] <edt> Bombe that would be a good conclusion provided that some ipv6 was working. It could be just that a connection is down.
[13:00] <saces> nextgens: the next testupload should be done in a minute...
[13:00] <Bombe> edt, internet connectivity is a bitch. You never know whether the problem is on the local or on the remote end.
[13:01] <edt> yes
[13:01] <edt> but we have a really good indication that ipv6 is a problem if ipv4 is working correctly and ALL ipv6 is failing
[13:02] <edt> we just cannot make a choice if ipv4 is failing and ipv6 is also failing
[13:03] * Nogaso (n=Nogaso@) has joined #freenet
[13:03] <Bombe> edt, define "ALL". After what time do you want to device that IPv6 is non-functional? And what if you really only get bad IPv6 peers and the next connect would succeed but you'd never try?
[13:03] <Bombe> I don't really see the problem.
[13:03] <Bombe> Let the node try.
[13:04] <Bombe> An IPv6 connect does barely resume any resources.
[13:04] <Bombe> s,resume,consume,
[13:04] <edt> Bombe on my connection page not one ipv6 packet has been recieved yet 1000s have been sent
[13:04] <edt> that ALL
[13:05] <Bombe> So what? This is a non-issue the way I see it.
[13:05] <edt> s/that/thats/
[13:05] <saces> nextgens: the done upload is present after restart.
[13:05] <Bombe> edt, if you really want to, file a bug report for the skip-IPv6 config option.
[13:06] <edt> Bombe I guess that 30 years of computers make me really want tight code
[13:07] <edt> OK
[13:09] * Zothar_Work (n=zothar@) has joined #freenet
[13:09] * ChanServ sets mode +o Zothar_Work
[13:10] <nextgens> saces> meaning that it works ?
[13:10] <nextgens> saces> may you test for a download before I release please ?:)
[13:12] <nextgens> edt> ipv6 has several advantages for us
[13:12] <nextgens> we want to support it and to use it where it's available
[13:12] <nextgens> so we try it everywhere
[13:12] <edt> nextgens its great to use it IF it is working
[13:12] <edt> It does not work here
[13:13] * Hayona (i=lol@) Quit (Remote closed the connection)
[13:13] <NEOatNHNG> also got the bandwith over limit problem, after i changed the loglevel to minor it grew to 245MB
[13:13] <edt> and we can detect it looking at the connections page - here not one ipv6 packet has been received...
[13:13] <NEOatNHNG> node version: 1080
[13:14] <NEOatNHNG> nextgens, do you still want to have it?
[13:15] <saces> nextgens: I repeat my last download, takes a moment...
[13:17] <nextgens> NEOatNHNG> toad will want the logs
[13:17] <nextgens> edt> what if you're firewalled and you need udp-holepunching ?
[13:17] <nextgens> udp-holepunching isn't only about ipv4 and NATs
[13:17] <nextgens> it's also about firewalls
[13:17] <nextgens> and any ipv6 host could be firewalled as well
[13:17] <nextgens> is there any other bug I should be aware of before I release ?
[13:17] * MrDomino (i=steve@) Quit (calvino.freenode.net irc.freenode.net)
[13:18] * MrDomino (i=steve@) has joined #freenet
[13:19] * Shadow_Vixen (n=panther@) Quit (calvino.freenode.net irc.freenode.net)
[13:19] * geeo (n=geeop@) Quit (calvino.freenode.net irc.freenode.net)
[13:19] * luelue (i=luelue@) Quit (calvino.freenode.net irc.freenode.net)
[13:19] * FrinkC (n=FrinkC@) Quit (calvino.freenode.net irc.freenode.net)
[13:19] * |R (i=bob@) Quit (calvino.freenode.net irc.freenode.net)
[13:19] * Jflesch (n=jflesch@) Quit (calvino.freenode.net irc.freenode.net)
[13:19] * Shadow_Vixen (n=panther@) has joined #freenet
[13:19] * luelue (i=luelue@) has joined #freenet
[13:19] * Jflesch (n=jflesch@) has joined #freenet
[13:19] * FrinkC (n=FrinkC@) has joined #freenet
[13:19] * |R (i=bob@) has joined #freenet
[13:19] * geeo (n=geeop@) has joined #freenet
[13:19] <NEOatNHNG> ok restarted the node kept it running for 30s now it's just 35MB
[13:19] <nextgens> NEOatNHNG> I hope you kept the logs for toad
[13:20] * dra9on (n=dra9on@) has joined #freenet
[13:22] * nextgens updates to trunk
[13:23] * Zothar_Work updates
[13:25] * mYone (i=schlepto@) Quit (Remote closed the connection)
[13:26] * FreedomSupporter (n=Gabber_F@) has joined #freenet
[13:27] * greycat (i=rfc1413@) has joined #freenet
[13:27] <FreedomSupporter> Good day all :)
[13:28] * mYone (i=schlepto@) has joined #freenet
[13:30] * wth (n=opera@) has joined #freenet
[13:30] <wth> oh great
[13:31] <wth> I misclicked and launched setup.bat instead of start.bat
[13:31] <wth> Now my config got resetted
[13:31] <FreedomSupporter> Just a brief question, I'm trying to install 0.5 but freenet.jar won't download (404ing)
[13:32] <nextgens> FreedomSupporter> which url are we talking about ?
[13:34] <FreedomSupporter> During the main install process the file freenet.jar doesn't work, I've tried http://downloads.freenetproject.org/freenet.jar and there's no sign of it; I don't know if that's the right url or not
[13:34] <nextgens> wth> usually we delete that file
[13:34] <wth> just a brief question too: Is my Datastore useless with a new config?
[13:34] <wth> i never mentioned it :(
[13:34] <pheenode> FreedomSupporter: that's because the project authors wish for the 0.5 network to disappear off the face of the earth for some reason
[13:35] <nextgens> wth> you need to change the paths in freenet.ini when the node is down for it to be used
[13:35] <toad_> rehi
[13:35] <nextgens> FreedomSupporter> where did you get that link from ? .5 has never used that filename
[13:35] <nextgens> FreedomSupporter> it's freenet-stable-latest.jar
[13:35] <nextgens> no ?
[13:36] <nextgens> maybe I broke it yesterday too
[13:36] <FreedomSupporter> Well it says in the installer 'downloading freenet.jar' and it isn't found
[13:36] <nextgens> the installer ?
[13:36] <FreedomSupporter> I'm trying to remember how I found the url
[13:36] <nextgens> which installer ?
[13:36] <FreedomSupporter> Windows
[13:36] <nextgens> hrrrmmm
[13:36] <nextgens> ok, will fix it
[13:36] <toad_> saces: yay
[13:37] <toad_> <nextgens> edt> ipv6 has several advantages for us
[13:37] <toad_> it does? people will still use stateful firewalls even if they don't have to NAT
[13:37] <toad_> <NEOatNHNG> also got the bandwith over limit problem, after i changed the loglevel to minor it grew to 245MB
[13:37] <toad_> NEOatNHNG: big deal :)
[13:37] <toad_> NEOatNHNG: expect several gigabytes per hour
[13:38] <nextgens> FreedomSupporter> retry in 5 mins please
[13:38] <FreedomSupporter> Earlier in the week the install went fine, but due to experimentation and XP problems this is something like the 3rd try within the past 5 days I've installed it
[13:38] <nextgens> toad_> the main advantage is that they have a static ip address!
[13:39] <nextgens> FreedomSupporter> as I said, give it 5mins for the mirrors to update
[13:39] <FreedomSupporter> Ok, thanks :)
[13:39] <wth> phew thank god my datasotrage is still okay
[13:39] <wth> it works now... with 2 refs >.>
[13:39] <nextgens> wth> when did you install freenet ?
[13:40] <nextgens> I need to know whether the current installer deletes those scripts or not :)
[13:40] <wth> i installed it 2 weeks ago
[13:40] <wth> that was when you released that big plugin fix update
[13:40] * FreedomSupporter heads off to sort out his system
[13:40] * FreedomSupporter (n=Gabber_F@) Quit ("Leaving.")
[13:41] <toad_> nextgens: *maybe* they have a static IP ... why do you think so many ISPs regularly change IPs on 24x7 connections?
[13:41] <toad_> NEOatNHNG: do you have logs for me?
[13:41] <nextgens> toad_> well, on ipv6 they *will* have a static ip address
[13:41] <toad_> nextgens: of course they won't
[13:41] <nextgens> why ?
[13:42] <toad_> nextgens: why do ISPs rotate IPs on permanent connections even if they stay up? to prevent servers!
[13:42] <nextgens> no
[13:42] <toad_> no?
[13:42] <toad_> most german ISPs rotate every 24 hours for example
[13:42] <toad_> I don't see any other plausible motive
[13:42] <nextgens> well, at the beginning it was to reset their hardware on a regular basis because it was unreliable
[13:42] <wth> there isnt
[13:43] <toad_> does anyone have logs of the bandwidth problem?
[13:43] <nextgens> NEOatNHNG told me that he is uploading some
[13:43] <toad_> symptoms: output bandwidth is way up (well over the bwlimit, maybe several times), payload is way down (20% or less)
[13:43] <toad_> NEOatNHNG: uploading to a website?
[13:44] <nextgens> toad_> http://dl.free.fr
[13:44] <toad_> NEOatNHNG: how long will it take?
[13:44] <nextgens> it's gonna be fast for me to download them afterwards :p
[13:44] <toad_> Total Output: 1.27 GiB (28.6 KiB/sec)
[13:44] <toad_> curious
[13:44] <toad_> the average, over quite a long period, is over the limit
[13:45] <toad_> but not much over - i could do 100kB/sec on this uplink, but it's not much higher than the limit of 25k
[13:45] <toad_> also i dunno if this is before or after the limiting fix
[13:46] <toad_> any theories on the Bandwidth Bug?
[13:46] <toad_> it has to be something low level, but if it was the PacketSequenceException thing that would have shown up on the normal level logs I saw
[13:47] <toad_> NEOatNHNG: if you can hear me ... go to the friends and strangers pages, click on "more detailed", there's a "total traffic" column, does any one node have a really huge amount of data transfer?
[13:47] <toad_> if it's one node that would be interesting
[13:51] <toad_> GRRRRRRRR
[13:51] <toad_> everyone gets this fscking bug except me!
[13:51] <toad_> now i've got an N2NTM saying JustMe is shutting his node down because of it!
[13:53] * TheBishop_ (n=bishop@) has joined #freenet
[13:54] <wth> argh i was wrong
[13:54] * FreedomSupporter (n=Gabber_F@) has joined #freenet
[13:54] <wth> my datastore is useless now
[13:54] <FreedomSupporter> Just a note, it works fine now
[13:54] <FreedomSupporter> Now to pick a name that isn't as long
[13:55] <toad_> wth: you can get it back, just change the listenPort back to what it was
[13:55] <wth> listenport? isnt it alwas 8888?
[13:55] <toad_> no
[13:55] <toad_> that's the fproxy port
[13:56] <toad_> i mean the listenPort, the UDP port
[13:56] <wth> ah okay
[13:56] <wth> I think i still have it
[13:56] <wth> ill give it a try
[13:56] <toad_> well the datastore filenames include it
[13:56] <wth> do I have to reboot the node now?
[13:56] <wth> ah thanks for the hint :D
[13:58] <luelue> the "freenet.googlecode" trunk is at 16106 when will it be in sync with emu?
[13:58] <nextgens> wth> edit freenet.ini only when the node is down
[13:58] <nextgens> luelue> tonight
[13:58] <luelue> ok
[13:59] * FreedomSupporter (n=Gabber_F@) has left #freenet
[13:59] <wth> pleeeeeeeeeeeeeeease work now >.<
[14:00] <wth> my old refs work again
[14:00] <wth> holy crap :D
[14:00] <wth> the magic port changed everything :D
[14:00] <wth> ty
[14:10] <wth> update successful!
[14:12] <wth> yeah the flow is back!
[14:12] <wth> it works perfect again, thanks for support
[14:16] <nextgens> 1081 is out
[14:16] <saces> nextgens: completed testdownload is present after restart, the incomplete is restarted. (seems work like expected)
[14:16] <nextgens> ok
[14:17] * NEOatNHNG (n=NEOatNHN@) Quit ("Ex-Chat")
[14:27] <Cooo> wooah. a lot of activity since I was here yeasterday, better upgrade.
[14:33] <saces> r16128 have still MAX_INT as completiontime on clientput: PutSuccessful {CompletionTime=9223372036854775807, URI=USK@MYLAnId-ZEyXhDGGbYOa1gOtkZZrFNTXjFl1dibLj9E,Xpu27DoAKKc8b0718E-ZteFrGqCYROe7XBBJI57pB4M,AQACAAE/KeyExplorer/2, StartupTime=1196432205136, Identifier=test1196432205097} EndMessage
[14:35] <nextgens> saces> when did you start the request ?
[14:35] <nextgens> saces> using a fixed build or not ?
[14:36] <saces> started with r16128.
[14:36] <nextgens> ok, hold on
[14:40] * Gudihl (n=Gudihl@) Quit ("Leaving")
[14:40] * friday (n=hah@) has joined #freenet
[14:40] * Orakio (n=hah@) Quit (Read error: 104 (Connection reset by peer))
[14:44] <nextgens> saces> please get r16132
[14:45] <nextgens> saces> ensure that insert resuming/ persistent downloads/ completionTime work
[14:49] * vi390 (n=fc@) has joined #freenet
[14:51] <saces> on persistance=connction comletiontime is now 0 :(
[14:51] <saces> PutSuccessful {CompletionTime=0, URI=USK@MYLAnId-ZEyXhDGGbYOa1gOtkZZrFNTXjFl1dibLj9E,Xpu27DoAKKc8b0718E-ZteFrGqCYROe7XBBJI57pB4M,AQACAAE/KeyExplorer/2, StartupTime=1196433650227, Identifier=test1196433650199} EndMessage
[14:52] <nextgens> well, that's better
[15:00] * Cooo (n=liquido@) Quit ()
[15:01] * Cooo (n=liquido@) has joined #freenet
[15:04] <toad_> saces: update, do an insert, restart, check the completiontime
[15:11] <mYone> is there a progress indicator for the automatic update somewhere?
[15:14] * wth (n=opera@) has left #freenet
[15:16] <toad_> mYone: do you have 1081?
[15:18] <mYone> toad_: nope not yet... the node says its currently downloading the update... i'd just like to know how much it already has downloaded ^^
[15:22] * vi390 (n=fc@) Quit (Connection timed out)
[15:23] <nextgens> saces> may you try it with 1082 please ?
[15:23] * ddoc (n=ddoc@) Quit ("null")
[15:36] <saces> competion time still 0: PutSuccessful {CompletionTime=0, URI=CHK@jrlPUMhTu7mkCw0EBAJlYjKrF8ROOeoxYGBspJId9C8,yhz94q13w2b8NjSLpVwvga2GMU~nK12l5Mlq5W9OrKQ,AAIC--8/KeyExplorer.jar, StartupTime=1196436346259, Identifier=test1196436346192} EndMessage
[15:37] <nextgens> ok
[15:37] <nextgens> so it's still broken
[15:44] <saces> hm. on request with persistance the completion time seems ok, on persistance=connection it is 0.
[15:48] <toad_> 1082 has been inserted
[15:48] <toad_> let me know if you successfully update / fail to update
[15:48] <nextgens> saces> grab what I've just commited please
[15:48] <nextgens> saces> r16137
[15:49] <nextgens> there is still one border case where we send completionTime=0 ...
[15:49] <toad_> bbiab
[15:49] <nextgens> when a request is removed
[15:49] <nextgens> is that bad ?
[15:50] * sanity (n=ian@) Quit ()
[15:51] <Cooo> toad_: my node just gave me a message that 1081 is avaible even though I am running 1081 r16132. I am gonna let it update to see what happens
[15:54] * Ratchet (n=xxxx@) has joined #freenet
[16:01] * grawity (n=grawity@) has joined #freenet
[16:03] <nextgens> saces> I'm confident that r16137 fixes it
[16:11] * sanity (n=ian@) has joined #freenet
[16:11] * ChanServ sets mode +o sanity
[16:15] <saces> nextgens: its maybe fixed now ;)
[16:15] <saces> PutSuccessful {CompletionTime=1196438546041, URI=CHK@jrlPUMhTu7mkCw0EBAJlYjKrF8ROOeoxYGBspJId9C8,yhz94q13w2b8NjSLpVwvga2GMU~nK12l5Mlq5W9OrKQ,AAIC--8/KeyExplorer.jar, StartupTime=1196438370134, Identifier=test1196438370039} EndMessage
[16:19] <nextgens> yay :)
[16:19] <nextgens> ok
[16:20] <nextgens> well, as I said there is only one race condition remaining
[16:20] <nextgens> what shall I return when a request is cancelled ?
[16:20] <nextgens> current time ?
[16:20] <nextgens> 0 ?
[16:21] <Bombe> 0. It never completed.
[16:21] <nextgens> fair enough
[16:22] <nextgens> so the current trunk behaves as it should
[16:22] * vi390 (n=fc@) has joined #freenet
[16:22] <saces> CanceledTime=timestamp. "how long was it running?"
[16:23] * CoolDude (n=no@) has joined #freenet
[16:25] * vi1390 (n=fc@) has joined #freenet
[16:27] * PhilKC (i=PhilKC@) has joined #Freenet
[16:30] <grawity> hi, I just got an internal error while using Fproxy.
[16:30] <grawity> the error message is: http://dark-code.bulix.org/es3z34-63133?raw
[16:30] <grawity> I hope this was the correct place :)
[16:32] <nextgens> grawity> which version of freenet are you running ?
[16:33] <grawity> Freenet 0.7 Build #1082 r16135M, Freenet-ext Build #17 r14866
[16:35] * vi391 (n=fc@) has joined #freenet
[16:35] <nextgens> :s
[16:35] <nextgens> ok
[16:35] <nextgens> well, I've fixed it in trunk
[16:36] <nextgens> next version should fix it
[16:36] <nextgens> in the meantime I'm not sure you will be able to see any jpeg :/
[16:39] * CoolDude (n=no@) Quit ()
[16:41] * vi390 (n=fc@) Quit (Read error: 113 (No route to host))
[16:41] * CoolDude11 (n=no@) has joined #freenet
[16:49] * vi1390 (n=fc@) Quit (No route to host)
[16:50] <TheSeeker> Can a cron job be made to update googlecode anytime there's a bump in version, in addition to the once a day thing?
[16:51] <Bombe> That wouldn't be a cronjob but a post-commit hook.
[16:52] <TheSeeker> hmm, well, in any case, googlecode is still at 16106 :P
[16:52] * nextgens doesn't regard that as a problem
[16:54] <TheSeeker> ok, dumb question #2 ... why not just use googlecode as the main repository, and mirror that one to emu for a backup in case googlecode goes down?
[16:56] <nextgens> hmm
[16:56] <nextgens> I've broken the bookmark code somehow :<<
[16:57] <nextgens> TheSeeker> because it's silly ?
[16:57] <nextgens> where would we put the auto-builder ?
[17:02] <TheSeeker> leave it on emu, and auto-build when you update from google... if it's not important that regular users have access to the latest source the minute it's out, then why would it be important that they have access to the latest compiled version the minute it's out?
[17:03] <nextgens> TheSeeker> because we need testers to try out our latest mileage of bugs
[17:03] <TheSeeker> by using google as the main repo, devs would still have the latest source, and regular users would be able to see the commits before their node updates automagically to the next manditory version.
[17:03] <nextgens> and sadly, we can't expect our testers to build from source themselves... so we do it for them
[17:04] <nextgens> TheSeeker> commits go through @cvs
[17:04] <nextgens> suscribe to it if you need to be informed of what's going on
[17:04] <nextgens> it's updated in real-time
[17:05] <nextgens> TheSeeker> back when we were using CVS and sourceforge, from time to time there was a 48h lag in between the developer's CVS and the public one
[17:05] <TheSeeker> *shrug* ah well. anyway... I'm late for work again x_x stupid pointless conversations *grumble* :P
[17:05] <nextgens> btw, I'm not sure that googlecode's svn is updated in real time for public access either
[17:06] <nextgens> TheSeeker> stupid pointless conversations you have initiated :)
[17:06] <Cooo> nextgens: Thanx for r16139, that button would come in handy :)
[17:06] <TheSeeker> nextgens: why do you think I'm so irritated? :)
[17:08] <nextgens> Cooo> it seems that there is a url-encoding bug in the bookmark manager somewhere
[17:08] <nextgens> but I dunno where ^-^
[17:09] <TheSeeker> so... when can freenet start using java 1.5? when all of the jdk is opensourced?
[17:09] <Cooo> nextgens: hmm.
[17:16] * grawity (n=grawity@) Quit ("Leaving.")
[17:17] <TheSeeker> I re-added the default bookmark set, then tried to delete the tree, and it left the developer flogs in the list. (16139)
[17:18] <TheSeeker> Trying to delete the tree again says it does not exist. and the developer flogs 'folder' won't delete, but the bookmarks within it deleted fine.
[17:18] * Nogaso (n=Nogaso@) Quit ("Leaving")
[17:19] <TheSeeker> aaaand, I gotta go to work.
[17:21] <toad_> nextgens: how much bandwidth would it cost us to autosync from emu to google?
[17:24] <toad_> TheSeeker: yes, when either a workable openjdk ships or GCJ 4.3 comes out.
[17:29] <Cooo> toad_: My node has autoupdated to #1082
[17:30] <nextgens> toad_> dunno, it depends on how much/often we commit ;)
[17:30] <dbkr> surely post-commit hook?
[17:32] <nextgens> we can't do that on a post-commit hook
[17:32] <nextgens> we don't want the commit to hang forever when google is unreachable :p
[17:32] <nextgens> we could have a cron to do it on a regular basis, when autobuilder does something for instance
[17:33] <nextgens> but again, I don't see why that would be necessary
[17:33] <dbkr> oh, hum. I'd imagined that a post-commit hook was run, well, after the commit.
[17:33] <dbkr> actually, hang on - the post commit hook can trigger the sync in the background and then return, couldn't it?
[17:33] * Shadow_Vixen (n=panther@) Quit ("Уз послала так послала... Иди, говорит, и без ёлоськи не возврасяйся!")
[17:33] <nextgens> it's done after the commit but the client won't "return" until it's finished
[17:34] <nextgens> the post commit hook could fork and sync off thread, yes
[17:34] <nextgens> but we can't do it *on* the post-commit hook
[17:35] <nextgens> and once again, I won't do it unless I'm given a good reason to do it
[17:35] <nextgens> we didn't have that when were using SF's CVS
[17:35] <dbkr> what's the reason for not doing it?
[17:36] <nextgens> I'm not even sure that the public repository is updated in realtime on googlecode either
[17:36] <nextgens> dbkr> it's yetAnotherTodo, noone will use it, and it consumes bandwidth/cpu time on each commit
[17:36] <nextgens> all of them beeing independently good reasons not to do it
[17:37] * dbkr was using googlecode in places until he realised it was usually chronically out of date
[17:37] <nextgens> provided that toad_ grants rights on svn to whoever needs^wwants them, it's useless
[17:38] <dbkr> hang on, you're losing me here - are you saying we shouldn't have googlecode at all?
[17:39] <nextgens> that's not what I'm saying
[17:39] <nextgens> it's usefull to backup our tree reliably :p
[17:39] * ed_ (n=Ed@) has joined #freenet
[17:39] <dbkr> only if the backups are up to date :D
[17:39] <nextgens> anyway, I'm done argueing about that
[17:40] <nextgens> unless someone comes up with a real good reason to do it, I won't do it.
[17:41] * edt (n=Ed@) Quit (Read error: 110 (Connection timed out))
[17:41] <dbkr> we really ought to have a decent, up to date anyonymous svn miror :/
[17:41] <toad_> the reason to do it is that we are distributing binaries
[17:41] <dbkr> we are an open source project, after all
[17:41] <toad_> therefore we MUST distribute source
[17:41] <nextgens> they are hundreds of things I should do on emu and I don't...
[17:41] <toad_> this is both the letter of the law, the spirit of the law, and the spirit of the movement
[17:41] <nextgens> toad_> we do distribute it when we release
[17:41] <toad_> it's just good practice
[17:42] <nextgens> non-released builds aren't meant to be used
[17:42] <nextgens> that's also the spirit
[17:42] <toad_> they're meant to be tested
[17:42] <nextgens> toad_> again, if testers were able to compile builds themselves, my position would be different
[17:43] <toad_> well, if it takes less time to implement near-commit-sync than to argue about it then please do it
[17:43] <nextgens> I wouldn't auto-build anything but would provide access to an up to date repository
[17:43] <toad_> i don't see why it shouldn't be easy
[17:43] <toad_> if for some reason it will take an hour, then don't do it
[17:43] <toad_> but if it will take 5 minutes it's probably worthwhile
[17:44] * CoolDude11 (n=no@) Quit ()
[17:44] <nextgens> that's typically the kind of stuffs which is gonna take 5 ... until it breaks and then I'm gonna spend *hours* fixing it
[17:45] <nextgens> I've already spent hours on the post-commit stuffs (the auto-builder and so on...)
[17:45] <nextgens> and I know what I'm talking about
[17:45] <toad_> well IMHO it's extremely unfriendly to would-be devs to ask them to get an account BEFORE they can have access to the source
[17:45] <toad_> I don't see why you can't\=
[17:45] <toad_> I don't see why you can't add it to the end of the auto-builder script
[17:46] <nextgens> because I get a mail each time the auto-builder fails and usually debug any problem I find there
[17:46] <nextgens> I don't want to be spammed when google is unreachable
[17:46] <toad_> how often is that gonna happen?
[17:46] <toad_> i mean we're talking about google!
[17:46] <nextgens> (even though the problem is likely to come from our end rather than theirs)
[17:47] <toad_> yes and if it's at our end, then it's our problem, it probably affects other more important services like the web site
[17:47] <nextgens> yes, and ?
[17:47] <toad_> and they need to be fixed
[17:47] <dbkr> we should spam the committer, surely :)
[17:47] <nextgens> I still don't want to be spammed about it
[17:47] <nextgens> dbkr> send a patch
[17:47] <toad_> you don't want to be spammed when emu goes offline?!
[17:47] <nextgens> as I said, many things could be done/improved
[17:47] <toad_> surely emu going offline is exactly something that you would want to know about?
[17:48] <nextgens> that's really the less urgent item I've on my todo
[17:48] <nextgens> toad_> my mails are relayed through emu... I'll get spammed when it gets back online!
[17:48] <toad_> :)
[17:48] <dbkr> is the auto-builder source avaialble somewhere?
[17:48] <toad_> well what is your TODO now?
[17:48] <nextgens> anyway, as said previously I'm done argueing about that
[17:49] <nextgens> fixing that damn bookmark code
[17:49] <nextgens> then have a look to download-resuming, again
[17:49] <toad_> well I still think it's gonna seriously alienate casual contributors and generally fuck up any efforts to get new devs
[17:49] <toad_> how about we change the cronjob to run every 5 minutes?
[17:49] <nextgens> maybe give a shot to a bandwidth/load-limiting simulator (I didn't want to get involved in that area but as no significant progress is being made, I guess I will have to)
[17:50] <toad_> it will only upload anything if something has changed, right?
[17:50] <nextgens> toad_> the cronjob runs every 2mins
[17:50] <toad_> nextgens: i thought it ran every day?
[17:50] <nextgens> well, the auto-builder's one
[17:51] <toad_> nextgens: i mean the googlesync cronjob
[17:51] <nextgens> the svn-sync one runs every day from my account
[17:51] <nextgens> because it uses my credentials
[17:51] <toad_> nextgens: so it has to be done by hand?
[17:51] <toad_> nextgens: what if you're away for a while?
[17:51] <nextgens> what do you mean by hand ?
[17:52] <nextgens> no, it's done in a cronjob
[17:52] <toad_> "because it uses my credentials"
[17:52] <toad_> okay so why not increase the frequency of the cronjob to every 5 minutes
[17:52] <nextgens> well, they are stored in my profile in my ~
[17:52] <toad_> or even every 2 minutes
[17:52] <nextgens> because of bandwidth usage ?
[17:52] <toad_> it won't do anything if nothing has changed locally
[17:52] <toad_> right?
[17:52] <toad_> is it SVK?
[17:52] <nextgens> false
[17:52] <nextgens> no, it's svnsync
[17:53] <toad_> well it just has to check the revision number, surely?
[17:53] * PhilKC (i=PhilKC@) Quit (Client Quit)
[17:53] <toad_> so that's a few hundred bytes including crypto setup, no?
[17:53] <nextgens> that means establishing a tcp session, negociating using SSL, talking WEBDAV...
[17:54] <toad_> right but finding the overall revision number is cheap
[17:54] <nextgens> and sending me a mail if it fails
[17:54] <nextgens> what we need is to figure out locally if we need to sync or not
[17:54] <toad_> yeah
[17:54] <nextgens> ie if something has been commited before we last synced
[17:55] <nextgens> and that's cheap
[17:55] <toad_> svnversion
[17:55] <nextgens> but tricky as it's done by a different account
[17:55] <toad_> svnversion to a file, compare it with the previous copy...
[17:55] <nextgens> (I don't have access to the repository from my uid and don't need to)
[17:55] <nextgens> and anyway; it involves scripting => some work
[17:56] <toad_> some work which is helpful if we want to get more devs
[17:56] <nextgens> I don't think so.
[17:56] <toad_> some work which will take less time than we've spent arguing over it so far
[17:56] <nextgens> agreed
[17:56] <toad_> however, if it's too much, just change it to hourly
[17:56] <nextgens> hence I repeat for the 3rd time that I'm done debating it
[17:56] * ed_ (n=Ed@) Quit (Read error: 110 (Connection timed out))
[17:56] <toad_> hourly is better than daily and is affordable bandwidth-wise
[17:57] <toad_> and it's only gonna be a few megs a month at most
[17:57] <nextgens> do it yourself if you think that's important
[17:57] <toad_> well, then i'd have to get you to explain where the cronjob is
[17:57] <nextgens> don't mess up my profile, don't copy my credentials anywhere else
[17:57] * dra9on (n=dra9on@) Quit ("Leaving")
[17:57] * ed_ (n=Ed@) has joined #freenet
[17:57] <nextgens> figure it out, you've got the r00t
[17:58] <nextgens> and I still think that it's a waste of time
[17:58] <toad_> i can use @hourly in a crontab, right?
[17:58] * Ratchet (n=xxxx@) Quit (Read error: 110 (Connection timed out))
[17:58] <toad_> ok, done, we can stop arguing about it
[18:09] <toad_> nextgens: my network connectivity to emu is going down periodically
[18:09] <toad_> nextgens: i can ping google, but not emu, and i can't ssh into emu either
[18:09] * toad_ supposes being a good paranoiac you've disabled ping ... but i can't get a connection
[18:09] <toad_> i had something similar yesterday, it went away after a while
[18:09] <toad_> so i'm asking if there is a problem with emu's connectivity?
[18:09] <toad_> From 43-246.rbsov.bogons.net (85.158.43.246) icmp_seq=139 Time to live exceeded
[18:09] * nextgens_ (n=nextgens@) has joined #freenet
[18:09] * nextgens (n=nextgens@) Quit (Read error: 104 (Connection reset by peer))
[18:11] * nextgens_ is now known as nextgens
[18:11] * ChanServ sets mode +o nextgens
[18:12] <nextgens> toad_> I'm setting up my maildroprc to bounce you related mails
[18:12] <nextgens> toad_> you should have received the first of a long list
[18:15] * ed_ (n=Ed@) Quit (Read error: 110 (Connection timed out))
[18:15] <toad_> nothing yet
[18:15] <nextgens> toad_> for some reason emu went down for a short while (I got disconnected from irc) ... and the svnsync cron was running => a mails has been generated
[18:21] <toad_> yeah i killed the sync so i could see what it was doing
[18:21] <toad_> it finished, complained about MERGE request failed on /svn, MERGE of /svn: 502 bad gateway
[18:21] <toad_> then emu went down again
[18:21] <toad_> and is still down
[18:24] * nextgens (n=nextgens@) Quit (Read error: 104 (Connection reset by peer))
[18:26] <Cooo> Got an NPE.
[18:26] * vi391 (n=fc@) Quit (Read error: 113 (No route to host))
[18:28] <Cooo> added the default bookmarks with the new button. When trying to delete the "Freenet devel's flogs" I got this http://dark-code.bulix.org/l4da2a-63147?raw
[18:30] * nextgens (n=nextgens@) has joined #freenet
[18:30] * ChanServ sets mode +o nextgens
[18:30] <nextgens> :s
[18:30] <nextgens> I wonder why emu's connectivity is sucking
[18:30] * |Sunlight| (n=nothing@) has joined #freenet
[18:32] <nextgens> toad_> there is a high packet-loos for some reason
[18:34] * nextgens has caught the bookmarks get mangled bug :)
[18:38] <Cooo> nextgens: did you see the NPE i posted?
[18:40] <nextgens> no
[18:40] <Cooo> [19:21] <Cooo> added the default bookmarks with the new button. When trying to delete the "Freenet devel's flogs" I got this http://dark-code.bulix.org/l4da2a-63147?raw
[18:42] * jel_ (n=jel@) has joined #freenet
[18:42] <nextgens> Cooo> it might have been fixed by r16143
[18:42] <nextgens> Cooo> if you manage to reproduce it with it, I'll have a look
[18:42] <jel_> hi all. Can someone tell me how to make freenet work on ubuntu amd64?
[18:42] <jel_> (x86)
[18:43] <Cooo> nextgens: r16143 is starting up here now. Gonna check it out in a minute or two
[18:43] <jel_> uhh... forget I added that last part :)
[18:43] <nextgens> toad_> may you close the mantis tickets you've openned and fixed please ?
[18:46] <Cooo> jel_: what seems to be the problem with runiing freenet?
[18:46] <toad_> sure, when i come across them
[18:47] <Cooo> damn. My comp is slow sometimes :(
[18:47] <jel_> Cooo: actually, it's not a java issue like I thought (I automatically assume java is going to screw up lately :). It's just having permissions problems, since I installed as root into /usr/local, but want to run as a user. It's currently complaining that it can't create a log dir?
[18:49] <nextgens> you shouldn't use the installer as root
[18:49] <nextgens> toad_> that's also something I need to work on, yes.
[18:49] <jel_> OK. Can I uninstall?
[18:49] <nextgens> sure
[18:49] <nextgens> but you need to do it as root :)
[18:50] <jel_> java -jar Uninstaller/uninstaller.jar?
[18:50] <nextgens> yep
[18:51] * |Sunlight| (n=nothing@) Quit ()
[18:51] <jel_> OK, now installing as user. Thanks :)
[18:56] <Cooo> nextgens: NPE seems to be gone in r16143 :)
[18:58] * vi390 (n=fc@) has joined #freenet
[19:03] <Cooo> NPE again. http://dark-code.bulix.org/s45x08-63152?raw almost the same as the one before
[19:09] <nextgens> hrmm
[19:09] <nextgens> doing what ?
[19:10] <Cooo> I tried to delete one of my own bookmarks.
[19:13] <nextgens> was it within a category or at toplevel ?
[19:13] <nextgens> did it have any weird character in it ?
[19:13] <nextgens> can you share your bookmarks.dat file ?
[19:13] * nextgens will try to fix it without
[19:15] <Cooo> it was within a category, first category to be exact. That category has been with me for some while now so it made me suspect something else fishy was going on. Deleted bookmarks.dat to start from scratch and try to replicate it. No wierd characters in the bookmark or in the category.
[19:15] <nextgens> might have found the problem
[19:16] <esmhammer> i get process execution failed error when i try to install freenet on a freebsd amd64, any thoughts.
[19:17] <nextgens> Cooo> update and try to reproduce please
[19:17] <Cooo> nextgens: :)
[19:17] <Cooo> nextgens: will do.
[19:17] <nextgens> Cooo> didn't your bookmark or category have a trailing space somewhere ?
[19:17] <nextgens> esmhammer> pastebin your error please :)
[19:18] <Cooo> nextgens: no spaces in bookmark or category.
[19:18] <nextgens> hmm
[19:18] <nextgens> maybe I didn't fix it then
[19:18] <esmhammer> noob here how to do that
[19:18] <nextgens> Cooo> you need r16147
[19:19] <Cooo> nextgens: updating and trying my best to reproduce :P
[19:19] <nextgens> esmhammer> copy paste your error on http://code.bulix.org/ and give me the url it gives back please :)
[19:19] <nextgens> Cooo> if you have kept the bookmark file it should be trivial
[19:21] <Cooo> nextgens: Yep..
[19:23] <esmhammer> nextgens: in lzpack it gives me the error and a chance to continue anyway, i tried to continue and after ward it won't work. should i continue and run it after to get the error.
[19:26] <nextgens> esmhammer> and it doesn't tell you anything except that there was an error ?
[19:31] <esmhammer> nextgens: ya it starts complaining after enabling XMLLibrarian plugin and the next line about detecting tcp-ports and can't bind fproxy to 8888.
[19:32] <esmhammer> binding fproxy to 9999.
[19:35] <esmhammer> nextgens: by the way i've got my browser showing successful installation but first time wizard gives me unable to connect to 127.0.0.1:9999
[19:38] * merethan (n=merethan@) has joined #freenet
[19:38] <merethan> Hi al
[19:38] <merethan> all
[19:39] <nextgens> esmhammer> that's weird
[19:40] <nextgens> esmhammer> pastebin the wrapper.log file from your freenet folder please
[19:40] <toad_> any opinions on performance of 1082 relative to previous builds? more bandwidth usage? looks that way to me
[19:40] <merethan> I have a node installed on my headless server, now I want to acces it using my desktop (trough my LAN). in freenet.ini I added the IP of my desktop to the default values at fproxy.allowdHosts and the full acces variant.
[19:40] <merethan> but, no go
[19:41] <Cooo> nextgens: Lot of things going on now :P I havent got any NPE yet, other strange things are happening though. I updated to r16147 without any bookmarks at all. When the node was up and running I added the default bookmarks with the add button. After that I tried to add a category named Etc to root ( / ) but instead the category seems to be located under "Default bookmarks - 1196450532930" after that I tried to add
[19:41] <Cooo> anoter category named Etc2 to root ( / ) but that one got located in the first Etc I added.
[19:42] <Cooo> toad_: Payload seems to be up to normal now :) Thats good.
[19:43] <toad_> Cooo: "up to normal" ?
[19:43] <nextgens> merethan> you shouldn't hand-edit the freenet.ini file; the wizard should have asked you about LAN hosts to allow
[19:43] <toad_> Cooo: so there's no improvement relative to previous builds?
[19:43] <merethan> nextgens, I did mention the word "headless" right?
[19:44] <Cooo> 82% payload. I have been down to 70 or lower in the last days.
[19:44] <nextgens> ah, sorry
[19:44] <toad_> Cooo: woah
[19:44] <toad_> 82% is pretty nice
[19:44] <nextgens> merethan> still, you should use the wizard : http://127.0.0.1:8888/wizard/
[19:44] <toad_> i never get that high
[19:44] <nextgens> merethan> using a text browser if you don't have anything else :p
[19:45] <merethan> nextgens, tied that already
[19:45] <merethan> links won't do.
[19:45] <merethan> What's wrong with edditing a conf file?
[19:45] <Cooo> toad_: seems to bee good, yes. I have been having that kind of numbers for long now, except for the latest days that is.
[19:45] <toad_> Cooo: what about overall transfer?
[19:45] <merethan> Once I can use my desktop's Firefox I will do everything as it is supposed to be done again
[19:47] <nextgens> merethan> well, read the wiki then, it explains how to set that up using freenet.ini
[19:48] <esmhammer> nextgens: it is also complaining about missing wrapper-freebsd-x86-64 and exception in thread "main" java.lang.noclassdeffounderror: freenet/node/nodestarter.
[19:48] <sanity> hey guys, how is it going?
[19:48] <toad_> sanity: hey
[19:48] <toad_> sanity: did you get time to look at my mail on load management yet?
[19:49] <sanity> toad_: i don't think i saw that
[19:49] <sanity> toad_: did you send it directly to me?
[19:49] <toad_> sanity: opennet is progressing...
[19:49] <toad_> sanity: I CC'ed you
[19:49] * jel_ is now known as jel
[19:49] <toad_> [freenet-dev] Why Freenet is so SLOW! Proposal for new load management
[19:49] <nextgens> esmhammer> you are using a free-jvm I guess, and that doesn't work with the installer
[19:49] <Cooo> toad_: overall transfer seems descent but not better than it has been the last months.
[19:49] <sanity> ah, will read it now
[19:50] <toad_> basically I want it to get out of the way, propagate load, constrain misrouting to some degree, and be simple, possibly at a short term cost in latency
[19:50] <sanity> it got filtered into my freenet-dev folder, and i haven't been following that too closely
[19:50] <esmhammer> nextgens: any way around it, can i install or compile it manually?
[19:50] <nextgens> toad_> I suggest you explain it on -dev to avoid confusing people here
[19:50] <nextgens> toad_> I suggest you explain it on -dev to avoid confusing people here
[19:51] <nextgens> sorry
[19:51]