=== asac_ is now known as asac [13:40] heya jdong [13:40] jdong, why the hell it's still not built? [13:40] bluekuja: I'm wondering the same [13:40] jdong, queued 16 hours ago [13:41] bluekuja: I'm discussing with fujitsu in -motu right now what to do with Azureus in feisty [13:42] jdong, following it a bit [13:42] jdong, let me look at the i386 build machines [13:43] jdong, seems busy [13:45] bluekuja: eep how much more in queue? [13:47] jdong, checking [13:48] jdong, the queue is huge! [13:48] jdong, https://launchpad.net/ubuntu/hardy/+builds?build_text=&build_state=pending&start=75&batch=75 [13:48] out build score is too low [13:48] dont know why [13:49] also because main packages come before universe [14:03] bluekuja: reupload with urgency=critical [14:03] * jdong ducks [14:03] lol [14:03] jdong, we should have set it as medium or high [14:04] :) [14:04] bluekuja: lol, yeah [14:04] I find it odd that arch-indep packages are forced to build on i386 [14:04] I'm sure the amd64 builders aren't as busy? [14:04] true [14:04] (gcc tends to be faster in compiling 64-bit binaries than 32-bit ones) [14:05] jdong, it will build it on hppa [14:05] too [14:05] I guess [14:05] lol [14:05] :D [14:05] so we will end up having it built [14:05] in a week [14:05] -.- [14:05] that long? :( [14:05] no, I guess [14:05] 1-2 days [14:05] more [14:07] meh not terrible [14:07] annoying if we're gonna have to wait that often for each build [14:07] yeah... [14:07] * jdong contemplates asking pitti to copy to -backports now to parallelize it [14:07] I hope not [14:08] jdong, yeah, it would be a nice idea. Does backports use a special build machine? [14:08] hah, nope. [14:08] ah darn [14:08] ^^ [14:08] wonder if there's a way to send backports into a binary-unapproved queue? [14:08] I don't want hte backport to come out then it turns out to be broken [14:09] though I'd have no idea how that happens :D [14:09] don't know...But as far as we need to wait the result for that buildd thing [14:09] yeah [14:10] if that works out [14:10] it's ok [14:10] if not, we have a big problem [14:10] but I *guess* it will work. [14:10] bluekuja: consdiering that it's this weekend when it comes out,l I'll set up a vmware hardy and test out-of-the-box install of azureus [14:10] jdong, true, it will be built around saturday [14:10] or sunday [14:10] I guess [14:11] jdong, hardy atm is quite the same of gutsy [14:11] so having it working for gutsy is a 90% to have it done for hardy as well [14:12] jdong, right? [17:38] bluekuja: hardy is virtually identical [17:38] bluekuja: we KNOW it works for gutsy as long as the buildd builds :) [17:40] jdong, lol [17:40] true :) [18:35] bluekuja: we've got trouble, btw. [18:36] bluekuja: icedtea running az on amd64 segfaults, but not on i386, and sun-java* works fine on amd64 [18:36] bluekuja: puts us in a tight spot on amd64..... [18:36] darn [18:37] why does it segfaults? [18:37] bluekuja: http://rafb.net/p/usAG6344.html [18:37] that's the traceback... [18:37] bluekuja: I asked the person to file a bug report [18:37] bluekuja: I think it's a bug in icedtea [18:38] bluekuja: I'll set up an amd64 ubuntu box over the weekend and investigate this... [18:38] jdong, sounds great [18:38] keep me updated when you're up on that box [18:38] I hope to be around this weekend [18:38] bluekuja: initial report claims it segfaults on the updater [18:39] bluekuja: I might just nuke the whole damn updater [18:39] can't say I didn't try ;-) [18:39] lol [18:39] what's the reason for using idedtea over sun-java anyway? [18:40] camrdale: to avoid multiverse demotion [18:40] I force az to use sun-java with no problems [18:40] camrdale: you can do that with the new package [18:40] camrdale, did you see the debtorrent sync bug? [18:40] because sun-java is non-free? [18:40] camrdale: correct [18:40] bluekuja: yes, responding now [18:40] camrdale: I think that'd be doko's last straw [18:40] camrdale, thanks [18:40] if I were to require sun-java* on an architecture, he'd go ballistic [18:40] is sun-java6 still non-free? [18:41] camrdale: correct [18:41] damn, I thought they opened it up [18:41] camrdale: they opened up java 7, aka icedtea [18:41] ahhh, now I see I need to do more reading, thanks [18:43] bluekuja: ok, for now, I'm going to change the logic of the launcher [18:43] bluekuja: it will perform a java -version on /usr/bin/java first.... and if that's either Sun or icedtea, it'll use it [18:43] bluekuja: that way amd64 users can update-alternatives to a sun java, and azureus will comply [18:44] jdong, we will require a new upload for it then [18:44] just let me know when it's tested and ready [18:44] bluekuja: yeah, it's a new upload.. and a suboptimal workaround [18:44] bluekuja: it's like "unofficially" multiverse. dodging the archive admins basically [18:44] sounds great [18:45] not ideal, but works until icedtea behave [18:45] bluekuja: apparent'y void^ has seen this error on openoffice & icedtea too [18:45] bluekuja: so it'll probably be fixed in a later icedtea [18:46] jdong, I hope we will have a 1ubuntu1 working package [18:46] and buildd wont complain [18:46] so we can move to fix the amd64 thing [18:46] you just told me [18:46] * bluekuja is going for a showa [18:46] bluekuja: right; that's my goal... [18:46] bluekuja: 1ubuntu1 working, then we just change the order of the launcher's checks [18:47] and call it done for gutsy-updates... [18:47] that'll be a simple change [18:47] jdong, if it builds fine, we can backport it [18:47] we'll deal with icedtea stack fixes as -backports [18:48] I asked void^ to open a bug report on it [18:48] so we can track it better [18:48] but meh, I'm not too concerned about the amd64 thing for now [18:48] jdong, news from doko? [18:48] bluekuja: nope [18:48] strange [18:48] maybe he's angry [18:48] ^^ [18:49] bluekuja: but as I said, demoting azureus to multiverse would definitely push him over the edge [18:49] bluekuja: I think he bought my last explanation and backed down [18:49] that java.nio is broken in GCJ [18:49] lol [18:49] now one can defeat mr. azureus [18:49] ;) [18:49] *no [18:50] :P [18:50] jdong, I go for a shower [18:50] I'll bll [18:50] *bbl [18:50] damn keyboard [18:51] ok [19:00] hello [19:00] hi [19:01] i would like to report Azureus crashing on Ubuntu 7.10. Is this only me or there have been others complaining about the same issue? [19:01] BaKKaR: there are around 30 duplicate bug reports of this [19:01] BaKKaR: I am working on a fix and am very close to it [19:01] bug 57875 [19:01] Launchpad bug 57875 in azureus "Azureus hangs or crashes showing splash screen at start" [High,Confirmed] https://launchpad.net/bugs/57875 [19:02] thank you jdong. Keep the spirit high and good luck. [19:02] any log from my files i can help you with? [19:02] thanks :) [19:02] or any fix you want me to try? [19:02] no, at this point I have all the information I need [19:02] there is a fix attached near the end of that bug report that you can play with [19:02] I know that fix is not correct on amd64, and am working towards that [19:03] jdong: hope i will know how :) other wise i will have to wait for the update-manager :-P [19:04] :) [19:04] it's simply a .deb file [19:04] BaKKaR: the update should hit Gutsy by end of next week [19:04] jdong: Super Cool. Thank you for you work :) [19:05] sure thing [19:05] wil try the dep now [19:05] will report in few seconds [19:05] :) [19:17] back [19:17] hi all [19:18] w/b [19:18] heya BaKKaR [19:18] hi [19:19] BaKKaR, azureus user [19:19] right? [19:19] yup [19:19] and YES it does crash with me [19:19] cool, we are waiting the package to be built [19:19] i am trying the fix now [19:20] great, report back to john [19:20] BRIGITTE LAHAIE [19:20] oops [19:20] one sec [19:20] he's following this [19:20] yesh this is why i am here waiting for the deb to install and will report to him [19:20] btw what is java icedtea? FOSS java? [19:22] BaKKaR: yes - icedtea.classpath.org [19:22] BaKKaR, yup based on OpenJDK [19:23] hi hansel [19:23] camrdale, answered? [19:23] yes [19:23] WoW, you guys are amazing [19:25] camrdale, great [19:25] BaKKaR, why? [19:25] :) [19:26] * jdong wonders if he can make Azureus self-aware [19:26] ROFL [19:26] lol [19:26] i thought Java is only made by Sun it was ported only because sun has it's own linux. But you guys have went all the way to make an all knew java? [19:26] or some less cool term for "azureus checks if it crashed last time thanks to the JVM" [19:27] ;-) [19:27] BaKKaR: IcedTea is a sun-sponsored project, we merely package it and make it available from the repositories [19:27] :) [19:27] BaKKaR: we are the only distro I know of with icedtea in a stable release... [19:27] Fedora Core 8 will sport icedtea [19:28] are you telling me in couple years Linux will be ( java wise) based on Java IcedTea? [19:28] for it's Java applications? [19:29] I better hit the FAQ, no? [19:29] * BaKKaR blushes [19:29] BaKKaR: that's the way I see it going [19:29] but I am not a Java expert [19:29] icedtea is also know as "Java 7" [19:30] it will be the next major release of Sun's Java [19:30] once we put this java 6, 5 behind us, even Sun's Java will be open source [19:30] jdong: com'n don't say that. You are fixing one heck of java app :) [19:30] BaKKaR: my "fixing" is minor code tweaks, and mostly changing the java compiler and runtime :) [19:30] I don't fully undrstand what's going on in there [19:30] Java's a big complex environment [19:31] btw don't mind my english, am not a nateive speaker :) [19:31] and it has to be -- for it's not a native compiled language and can sometimes turn 2x C performance [19:32] jdong: 2x C performance?!?! now this is very much the opposite of all what i know so far. [19:33] BaKKaR: under certain circumstances (particularly tight mathematical loops, or cases where input data type wildly differs, Sun's JVM can run Java code faster than the equivalent C program compiled. [19:33] BaKKaR: Java is not inherently slow. it's in fact quite the opposite. [19:34] camrdale, forgot to tell you, if you need any help maintaining some of your packages, just tell me. That's my field. [19:34] this is very new to me. [19:34] BaKKaR: Sun's HotSpot JVM is so miles above the rest, that Microsoft even purchased Sun Hotspot technology to power its .NET framework [19:34] camrdale, e.g p2p apps ;) [19:34] BaKKaR: roughly speaking, it is able to monitor interpreted code running inside of it, and whenever it identifies a bottleneck, it compiles that part down to machine code [19:34] it constantly watches over execution and tweaks parts that can be optimized [19:35] i was on the loop to try and learn how to code in JAVA but when i heared the 'rumors' about it's speed and the performance of it's binary against C and C++ i didn't feel too much like completeing the task [19:35] most p2p apps' bottlenecks are in the network, rather than the CPU, anyway [19:35] BaKKaR: for almost everything you can write, Java will likely execute just as fast as native [19:35] bluekuja: thanks for the offer :) [19:35] charles_: except when using GCJ, where it hash checks some 100x slower than icedtea ;-) [19:36] camrdale, np ;) [19:36] charles_: you know something's wrong when hash checking is totally bound by CPU, not iowait :) [19:36] jdong: i must spread the word, LOL [19:36] jdong: ouch [19:36] BaKKaR: don't forget that Java and similar rntimes make your life easier by managing low-level stuff, like RAM allocation, for you [19:36] all the ppl i know are into the slow-motion-java path [19:36] BaKKaR: humans cannot be upgraded with more memory and faster CPU's. [19:36] but computers always can. [19:37] a language that spares human time is often worth the tradeoff [19:37] lol @ trade off [19:37] well, fast and light will never go out of style [19:38] charles_: C and C++ soon will :) [19:38] but Java much better than its old bad reputation would make you think [19:38] jdong: what's going to replace them? :) [19:38] charles_: higher-level languages, as the performance gap is quickly shrinking [19:41] jdong: the deb pack is working [19:41] I doubt that C and C++ are in any danger in their niches [19:41] now i am trying to download the updates [19:42] I suspect perl is in more danger (from python) than C/C++ are [19:42] update done [19:42] but I'm getting off-subject. you're right about java [19:42] jdong: now i am trying to download tonight's movie for BRIGITTE LAHAIE [19:43] that was a bit private info [19:43] :) [19:43] thank you [19:43] BaKKaR: sure thing [19:43] that was automated, no? [19:43] I'm not that magical :) [19:45] haha [19:48] jdong: kind of interesting: http://radar.oreilly.com/archives/2006/08/programming_language_trends_1.html [19:50] charles_: very interesting data indeed [19:50] charles_: nice to see C# increasing in popularity though [19:50] * jdong is a mono fanboy on a 10-step addiction program ;-) [19:50] charles_: I remember seeing that before - there are others that should be current - I think ohloh has one [19:52] that oreilly article is from before Sun changing the Java license to GPL, which was a late but positive change [19:57] leaving [19:57] night all [20:03] me oo [20:03] g.night [20:55] bluekuja: doko marked the amd64 crasher as a dupe of bug 152362 [20:55] Launchpad bug 152362 in icedtea-java7 "icedtea-java7-plugin always crashes firefox" [High,Confirmed] https://launchpad.net/bugs/152362 [20:56] that's great news for us -- the fix seems trivial [20:56] (rebuild icedtea) === nuu is now known as nu === nu is now known as nuu === nuu is now known as nu === nu is now known as nuu