| micahg | hggdh: the question is that if an upstream project which uses LP already has a bug for something that's in ubuntu and there is a bug in ubuntu, what do we do? | 00:00 |
|---|---|---|
| === pitoow is now known as A_pitoow | ||
| hggdh | ah | 00:10 |
| hggdh | micahg: then yes, I think dupping against the upstream would make sense. But this would have to be always done -- it does not make sense to have different options | 00:11 |
| micahg | hggdh: well, 1. should upstream see ubuntu bug spam? | 00:12 |
| micahg | 2. can LP have an upstream task to itself? | 00:12 |
| hggdh | two good questions, no easy answer. But (1) they already see bug spam (or, at least, can see if they so want); (2) it can, but does not make much sense, similar to the chicken and egg issue | 00:14 |
| micahg | ok, so dupe and open an ubuntu task | 00:15 |
| micahg | hggdh: right? | 00:16 |
| hggdh | this is how I would go, yes. Of course, mileage can vary | 00:16 |
| micahg | should I add a wiki page for it? | 00:16 |
| hggdh | I asked the same thing during UDS, and did not get a firm position | 00:16 |
| micahg | should I add it to the meeting agenda again for next month or ML it? | 00:17 |
| hggdh | Micah... not sure it warrants a wiki by itself, but at least a comment, yes | 00:17 |
| hggdh | ML it, if you do not mind | 00:17 |
| micahg | that's the main question I had for last months meeting :) | 00:17 |
| hggdh | this is a question I have had for a while ;-) | 00:18 |
| micahg | hggdh: well, the wiki shows info about the upstream bug trackers | 00:18 |
| micahg | and LP qualifies | 00:18 |
| micahg | with it's own process | 00:18 |
| micahg | *its | 00:18 |
| hggdh | yes. The problem is we are under LP | 00:18 |
| micahg | hggdh: -control or squad ML? | 00:18 |
| hggdh | we can go ortodox, but does it really make sense? | 00:18 |
| hggdh | -control | 00:18 |
| hggdh | methinks | 00:19 |
| micahg | hggdh: done | 00:25 |
| hggdh | I will add my .02 cents to it | 00:27 |
| hggdh | and yes, it is a .02 cents. Inflation adjustments | 00:28 |
| SoftwareExplorer | I'm getting ready to report a Lucid bug. Every time an application starts to play a sound or stops playing a sound there is a pop/click. Does anyone know if I should report this as a bug for the pulseaudio package or the linux package? | 00:55 |
| BUGabundo | SoftwareExplorer: ask crisum in #ubuntu+1 | 00:57 |
| BUGabundo | PA / alsa is his domain | 00:57 |
| SoftwareExplorer | BUGabundo: OK, I will thanks | 00:59 |
| BUGabundo | np | 01:00 |
| === nhandler_ is now known as nhandler | ||
| rCX | I contacted the upstream developers for bug #472818 and they confirmed that this behavior is intentional and not a bug. | 02:01 |
| ubot4 | Launchpad bug 472818 in playonlinux "guiv3.py crashed with IndexError in AutoReload() (dup-of: 392513)" [Undecided,New] https://launchpad.net/bugs/472818 | 02:01 |
| ubot4 | Launchpad bug 392513 in playonlinux "guiv3.py crashed with IndexError in AutoReload()" [Undecided,Confirmed] https://launchpad.net/bugs/392513 | 02:01 |
| rCX | oops I meant bug #472918 | 02:02 |
| ubot4 | Launchpad bug 472918 in quick-lounge-applet "Applet does not respect a size of 1 with 2 items" [Undecided,Confirmed] https://launchpad.net/bugs/472918 | 02:02 |
| rCX | Should it be marked as a wishlist item? | 02:02 |
| rCX | Should I mark it as invalid? thanks! | 02:03 |
| micahg | rCX: did they say they are willing to modify the behaviour? | 02:04 |
| rCX | micahg: no | 02:04 |
| micahg | so they won't change it, right? | 02:05 |
| rCX | micahg: yup, not going to change it | 02:05 |
| micahg | rCX: please comment with what they told you and I will mark Won't Fix | 02:05 |
| rCX | micahg: ok thanks | 02:06 |
| rCX | micahg: I commented but it won't let me select "won't fix" :( | 02:08 |
| micahg | rCX: that's why I said I'll mark it won't fix :) | 02:08 |
| micahg | only bug-control members can set it | 02:08 |
| rCX | micahg: Ok thanks :D | 02:09 |
| micahg | rCX: done | 02:09 |
| dnm | is any one using ubuntu on dell xps | 05:53 |
| dnm | wanted some help | 05:53 |
| dnm | can some one help... i am new to ubuntu and i cant use my wifi | 05:55 |
| JFo | dnm, the #ubuntu channel is your best bet. | 05:57 |
| dnm | thnaks | 05:57 |
| JFo | my pleasure | 05:57 |
| roark_deviated | I'm new to ubuntu bugs, learning how to triage. Do you know where I can find bugs that have been properly triaged so I can see how its done. I already read the how-to but I need an example to better guide me. | 06:01 |
| thekorn | bug 500000 | 10:59 |
| ubot4 | Launchpad bug 500000 in usplash "wrong aspect ratio boot splash on widescreen screen" [Undecided,New] https://launchpad.net/bugs/500000 | 10:59 |
| thekorn | nice! | 10:59 |
| === zoglesby_ is now known as zoglesby | ||
| === j is now known as Guest35297 | ||
| Guest35297 | hi there, I'm fairly sure I've found a bug in kubutntu but I have no diea what package it might be with. | 12:17 |
| Guest35297 | it's very similar to this https://bugs.launchpad.net/ubuntu/+source/libgphoto2/+bug/355998 but with an mp3 player not a camera | 12:17 |
| ubot4 | Launchpad bug 355998 in libmtp "usb-storage: waiting for device to settle before scanning" [Undecided,Fix released] | 12:17 |
| Guest35297 | also, it's kubuntu 9.10. any suggestions? | 12:17 |
| === Guest35297 is now known as vmos | ||
| vmos | anyone? | 12:25 |
| vmos | dust? | 12:25 |
| vmos | I've tried ubuntu-bug storage but it insists on a package name, any advice on narrowing it down? | 12:28 |
| qense | vmos: try executing just 'ubuntu-bug', that should provide you with a list of one item where you can choose the storage problems from. | 12:45 |
| vmos | thanks qense but when I try that it says "You need to specify a package or a PID. See --help for more information", I saw help and it didn't help | 13:15 |
| tgrundle | i am experiencing a bug in Luicid that has been reported in debian, should i create a bug report in launchpad and reference the debian? tickethttp://bugs.debian.org/cgi-bin/bugreport.cgi?bug=559289 | 14:37 |
| ubot4 | Debian bug 559289 in hplip "/lib/udev/rules.d/56-hpmud_support.rules: udev warnings: SYSFS{}= will be removed in a future udev version..." [Normal,Open] | 14:37 |
| qense | tgrundle: I would only report the issue here as well if it requires action from Ubuntu developers as well, otherwise it's just extra work to keep track of it. | 16:02 |
| charlie-tca | hggdh: you around? | 16:04 |
| hggdh | charlie-tca: I am | 16:09 |
| hggdh | charlie-tca: hope you are OK | 16:09 |
| charlie-tca | Never mind. I answered my question already | 16:09 |
| hggdh | heh | 16:09 |
| charlie-tca | Hey, I must be doing better. | 16:10 |
| hggdh | :-) | 16:10 |
| charlie-tca | I signed off that xchat bug report from karmic alpha3 | 16:10 |
| hggdh | oh | 16:11 |
| hggdh | and I moved away from xchat, into weechat (git head) | 16:11 |
| charlie-tca | oh | 16:11 |
| hggdh | current dev version looks really nice | 16:12 |
| charlie-tca | I got too many things I am trying to get my head around again now | 16:13 |
| === mac_v is now known as \vish | ||
| micahg | hggdh: if I send a frozen app a SIGSEGV, will the apport crash show why it froze? | 16:54 |
| === zoglesby_ is now known as zoglesby | ||
| hggdh | micahg: probably | 17:33 |
| hggdh | you will have to look at the threads, but if there is a deadlock somewhere you will see the wait | 17:33 |
| hggdh | (of course, discard the segv frames) | 17:34 |
| micahg | hggdh: ok, I'll try it | 17:40 |
| ali1234 | micahg: my firefox bug? | 17:42 |
| micahg | ali1234: yep | 17:43 |
| micahg | ali1234: I just added instructions to the bug | 17:44 |
| ali1234 | don't i have to start some service to get apport dumps if i'm not running a beta? | 17:46 |
| ali1234 | oh yeah you wrote it, sorry | 17:46 |
| micahg | ali1234: it's in the bug | 17:46 |
| ali1234 | ok, got it | 17:48 |
| ali1234 | The problem cannot be reported because I have obsolete software | 17:49 |
| micahg | ali1234: ok, so that ususally means you have to upgrade a few packages | 17:50 |
| ali1234 | libgtk2.0-common - but I need this version installed because there are no dbgsyms for 2.1, and i need those for another bug | 17:50 |
| ali1234 | ah whatever, i'll try to upgrade it | 17:50 |
| micahg | ali1234: no, the upgrade is not to 2.1 | 17:50 |
| ali1234 | yeah, you're right, the upgrade is ubuntu1 to ubuntu2 | 17:51 |
| ali1234 | however, there's no debug symbols package for -ubuntu2 | 17:51 |
| micahg | 2.18.3-1ubuntu2.1 | 17:51 |
| micahg | ali1234: libgtk2.0-0-dbg | 17:52 |
| ali1234 | yes, when i attempted to install that package the other day i got a message like "depends on =...-ubuntu1 but ...-ubuntu2 is going to be installed" | 17:53 |
| ali1234 | so i had to force version to get it installed | 17:53 |
| micahg | ali1234: apport retraces the crash server side so you don't need the dbg syms for it, if you need it for yourself, that's another story | 17:53 |
| ali1234 | i needed the debug symbols because somebody asked me to install them in another bug | 17:54 |
| micahg | ali1234: if you're done with that, you don't need them anymore | 17:54 |
| ali1234 | i'm not | 17:55 |
| micahg | ok | 17:55 |
| ali1234 | until the bug is closed :) | 17:55 |
| micahg | ali1234: do you know how to do a manual gdb backtrace? | 17:56 |
| ali1234 | yes | 17:56 |
| ali1234 | but i can already tell you what it's gonna say :) | 17:56 |
| ali1234 | java plugin is causing the freeze | 17:57 |
| micahg | ali1234: ok, that works too, can you reproduce the freeze and generate a full backtrace? | 17:57 |
| ali1234 | ok retrying apport | 17:57 |
| micahg | on a clean profile | 17:57 |
| ali1234 | the complete report is 172mb | 17:58 |
| ali1234 | i guess i should send the reduced one? | 17:58 |
| micahg | wow...either way should be fine | 17:58 |
| ali1234 | well i'll do reduced for now | 17:59 |
| ali1234 | otherwise it wont upload before xmas | 17:59 |
| ali1234 | wow there's loads of bug reports about this already, mine is probably a dupe of one of those... | 18:00 |
| micahg | :) | 18:02 |
| ali1234 | ok you want me to do a new bug still? i see you;ve already replied to a bunch of these | 18:02 |
| micahg | ali1234: are any of them triaged? | 18:02 |
| ali1234 | they're all in various states of incomplete, new, or invalid | 18:03 |
| micahg | ali1234: that generally means that the problem was on the user side in those cases or they're not responding | 18:03 |
| micahg | you can file a new one | 18:03 |
| micahg | and we'll try to get to the bottom of it :) | 18:03 |
| micahg | of apport finds the stacktrace is a dupe, it will mark the bug as such | 18:04 |
| ali1234 | i'll mark my original report a dupe | 18:05 |
| ali1234 | bug 500184 | 18:05 |
| ubot4 | Launchpad bug 500184 in firefox-3.5 "firefox freezes on pages with java" [Undecided,New] https://launchpad.net/bugs/500184 | 18:05 |
| micahg | ugh, that should be private | 18:06 |
| micahg | nm | 18:06 |
| ali1234 | why? | 18:06 |
| micahg | well, usually there's a coredump, but I guess since it's a reduced report, there isn't one | 18:07 |
| ali1234 | yeah, that part was 170MB | 18:07 |
| hggdh | too many missing symbols | 18:07 |
| micahg | yeah, I was about to say that :) | 18:07 |
| ali1234 | what's that apport-retrace command again? | 18:08 |
| hggdh | sudo apport-retrace -o a.file.name /var/crash/crash-whatever | 18:08 |
| hggdh | it will download & install all dbgsyms needed | 18:09 |
| hggdh | (as long as you have the ddebs in the sources.list) | 18:09 |
| ali1234 | a.file.name = what? | 18:09 |
| micahg | hggdh: that's actually really good to know :) | 18:10 |
| ali1234 | nvm got it | 18:10 |
| hggdh | ali1234: apport-retrace requires an output file | 18:11 |
| ali1234 | yeah, i figured it out. how do i resubmit the resulting regenerated report? | 18:12 |
| hggdh | ali1234: just upload the GDB output to the bug, micahg will follow up | 18:12 |
| hggdh | (i.e., run apport-retrace -g) | 18:13 |
| hggdh | and, of course, tell us. Probably good to run 'thread apply all bt full' and 'bt full' (so that we know the active thread) | 18:15 |
| ali1234 | hmm, i don't think this is working right | 18:15 |
| ali1234 | http://pastebin.com/m1d99a65b | 18:16 |
| ali1234 | (it's still running, but warns about missing packages - i have ddebs with the key imported) | 18:16 |
| ali1234 | oh and now it crashed | 18:16 |
| ali1234 | Exception ImportError: ImportError('No module named shutil',) in <bound method __AptDpkgPackageInfo.__del__ of <apport.packaging_impl.__AptDpkgPackageInfo instance at 0x933c44c>> ignored | 18:17 |
| hggdh | ali1234: there are many dbgsyms on wrong versions | 18:17 |
| === yofel_ is now known as yofel | ||
| ali1234 | fail | 18:20 |
| ali1234 | the backtrace is still missing almost all symbols | 18:20 |
| ali1234 | great, now i have 12 broken packages on my system too :) | 18:21 |
| ali1234 | firefox-3.5-gnome-support-dbgsym: Depends: firefox-3.5-gnome-support (= 3.5.3+build1+nobinonly-0ubuntu6) but 3.5.6+nobinonly-0ubuntu0.9.10.1 is installed | 18:22 |
| ali1234 | and loads of crap like that | 18:23 |
| hggdh | ali1234: weird, this is a bug on itself | 18:23 |
| ali1234 | yep | 18:23 |
| ali1234 | same bug i ran into the other day with libgtk | 18:23 |
| hggdh | this happened of old on apport (it would force-install wrong versins), but it was corrected | 18:24 |
| hggdh | so it is a regression :-( | 18:24 |
| ali1234 | this happened the other day when i tried to install the libgtk symbols in synaptic | 18:24 |
| ali1234 | basically afaict the ddebs repository is lagging behind the main repo by a few versions | 18:25 |
| ali1234 | so the only way to make it work is to install older versions of everything | 18:25 |
| hggdh | well, there are two different issues here | 18:25 |
| hggdh | (1) ddebs lagging behind official versions: expected, this happens | 18:25 |
| hggdh | (2) force-install of *wrong* dbgsym versions: this should *NOT* happen | 18:26 |
| ali1234 | the thing about (1) is, apport won't submit a report unless you have the newest version of everything, but without the ddebs, it can't submit a useful report | 18:26 |
| hggdh | I agree, almost a catch-22. But this is contention on finite resources -- the build machines --, and not much we can do on it | 18:27 |
| hggdh | except wait... | 18:28 |
| * ali1234 waits | 18:30 | |
| === \vish is now known as mac_v | ||
| * mac_v kicks the 5-a-day bot! | 18:35 | |
| hggdh | heh | 18:35 |
| mac_v | the stats havent been updated for nearly a week... almost 1-3 days ,as soon as i joined :( | 18:36 |
| hggdh | mac_v: you might want to ping bdmurray about it | 18:36 |
| hggdh | oh, just did it ;-) | 18:37 |
| mac_v | hehe :D | 18:37 |
| === BUGabundo is now known as SantaBUG | ||
| SantaBUG | Merry xtmas every one | 20:20 |
| matti | Thanks ;] | 20:21 |
| SantaBUG | hey matti | 20:24 |
| micahg | ali1234: I see you found your problem :) | 20:25 |
| ali1234 | yeah | 20:25 |
| ali1234 | i bet there's a few more dupes of this too | 20:26 |
| micahg | ali1234: you need to be careful with those | 20:27 |
| micahg | just because someone's having an issue with java, doesn't make it a dupe | 20:27 |
| micahg | like bug 478263 | 20:27 |
| ubot4 | Launchpad bug 478263 in firefox-3.5 "[KARMIC]Firefox 3.5.4 can't run java applets (dup-of: 484148)" [Undecided,New] https://launchpad.net/bugs/478263 | 20:27 |
| ubot4 | Launchpad bug 484148 in firefox-3.5 "apparmor-profiles freezes Firefox when using Java applets (Sun JRE)" [Undecided,Incomplete] https://launchpad.net/bugs/484148 | 20:27 |
| SantaBUG | micahg: speaking of bugs | 20:27 |
| SantaBUG | if Sun Java is being left behind | 20:27 |
| SantaBUG | shouldn't ubuntu-restricted stop pulling it ? | 20:28 |
| ali1234 | micahg: sure, but hopefully they'll read the comment i just wrote on the main bug, and then try the workaround, and then we'll know | 20:28 |
| micahg | SantaBUG: it hasn't been made official yet | 20:28 |
| === mac_v is now known as \vish | ||
| micahg | ali1234: the kern.log shows no apparmor issue with java | 20:28 |
| ali1234 | micahg: but stopping apparmor makes the issue not happen so... | 20:28 |
| micahg | ali1234: apparmor should NEVER be stopped entirely | 20:29 |
| ali1234 | hah | 20:29 |
| ali1234 | well, fix the bug then, and i won't have to | 20:29 |
| micahg | ali1234: that's not how it works...you can disable the firefox profile | 20:31 |
| ali1234 | yes, i see "sudo touch /etc/apparmor.d/disable/path.to.bin" | 20:32 |
| ali1234 | i read this too "Since Ubuntu 9.10 (Karmic), AppArmor ships with a profile for Firefox which is disabled by default. " | 20:34 |
| ali1234 | i never enabled it so what gives? | 20:34 |
| ali1234 | maybe this is the wrong channel for this type of question | 20:34 |
| micahg | well, was is enabled? | 20:34 |
| micahg | there was a bug with earlier FF35 packages that might have enabled it | 20:34 |
| ali1234 | yes, i see messages about firefox and apparmor all the time | 20:35 |
| micahg | ah, ok, well then, you might indeed have a duplicate | 20:35 |
| ali1234 | bug 489278 | 20:39 |
| ubot4 | Launchpad bug 489278 in apparmor "Apparmor should notify the user when it blocks access" [Undecided,New] https://launchpad.net/bugs/489278 | 20:39 |
| ali1234 | this is the same reason why torrents don't open in transmission any more | 20:39 |
| ali1234 | that's been happening since the day i upgraded to karmic | 20:39 |
| micahg | there's already a bug for that | 20:39 |
| ali1234 | yep | 20:39 |
| ali1234 | all these bugs are caused because apparmor is turned on | 20:40 |
| ali1234 | that bug ^ mentions torrents, and it just clicked | 20:40 |
| micahg | bug 476299 | 20:40 |
| ubot4 | Launchpad bug 476299 in firefox-3.5 "Firefox apparmor profile does not allow access to transmission" [Low,Fix released] https://launchpad.net/bugs/476299 | 20:40 |
| ali1234 | so, looks like all i need to do is disable the profile (which i never enabled) and my firefox will stop sucking | 20:42 |
| micahg | ali1234: yep | 20:42 |
| ali1234 | so, got a bug for that? | 20:43 |
| micahg | which one? | 20:43 |
| micahg | transmission? | 20:43 |
| ali1234 | "there was a bug with earlier FF35 packages that might have enabled it" | 20:43 |
| ali1234 | that one | 20:43 |
| micahg | idr | 20:43 |
| ali1234 | there's also bug 455792 | 20:44 |
| ubot4 | Launchpad bug 455792 in firefox-3.5 "deluge doesn't open .torrent files from firefox while apparmor is on" [Low,In progress] https://launchpad.net/bugs/455792 | 20:44 |
| ali1234 | and a load of others for various plugins and external files | 20:45 |
| ali1234 | but i don't see one about the profile getting enabled | 20:45 |
| micahg | ali1234: it was only for daily users | 20:45 |
| micahg | iirc | 20:45 |
| ali1234 | i upgraded this machine to karmic about 4 hours before the final ISO went public | 20:45 |
| ali1234 | and pretty much the first thing i noticed was torrents didn't open automatically any more | 20:46 |
| micahg | ali1234: daily ff users | 20:46 |
| ali1234 | nope, never used that afaik | 20:46 |
| micahg | hmmm | 20:46 |
| hggdh | so, ali1234 -- was your ffox apparmour profile enabled or disabled> | 20:46 |
| hggdh | ? | 20:46 |
| ali1234 | i'm not 100% sure how i tell | 20:47 |
| vmos | hi there, I've got a problem with karmic not recognising my mp3 player. I'm pretty sure it's a bug but I have no idea what package is at fault. any suggestions? | 20:47 |
| ali1234 | but i think it is enabled | 20:47 |
| ali1234 | given that i'm suffering from every single bug involved firefox, and having apparmor enabled for it | 20:47 |
| micahg | ali1234: bug 436221 | 20:48 |
| ubot4 | Launchpad bug 436221 in firefox-3.5 "apparmor profile is not disabled on upgrade from jaunty firefox-3.5" [High,Fix released] https://launchpad.net/bugs/436221 | 20:48 |
| hggdh | ali1234: ps auxZ | 20:48 |
| * micahg also can't get java to work with apparmor enabled... | 20:49 | |
| ali1234 | yeah it's enabled then | 20:49 |
| ali1234 | hmm ok, that bug doesn't tell me how to disable it now it's been erroneously enabled | 20:50 |
| ali1234 | if just updating the package is supposed to do that, then obviously there's been a regression | 20:51 |
| micahg | ali1234: I commented in the last bug you posted in about it | 20:51 |
| micahg | ali1234: http://www.ubuntu.com/getubuntu/releasenotes/910overview#New profiles | 20:51 |
| ali1234 | ok, rebooting to get my apparmor back how it's supposed to be | 20:59 |
| micahg | ali1234: ?? | 20:59 |
| ali1234 | i stopped it, remember? | 21:00 |
| micahg | ali1234: sudo /etc/init.d/apparmor start | 21:00 |
| ali1234 | that doesn't seem to work, now *all* my running processes are "unconfined" except the ones that i started after i restarted apparmor | 21:00 |
| micahg | oh, yeah, I guess that's a problem | 21:01 |
| ali1234 | right, apparmor profile disabled, ff works properly now | 21:09 |
| micahg | ali1234: :) | 21:12 |
| * micahg forgot how much trouble that profile can cause | 21:14 | |
| micahg | on the next ff3.6 update, I'll try to run ff with it enabled so that I can try to find the bugs in advance | 21:15 |
| micahg | ali1234: BTW, I filed bug #500215 about apparmor not warning people when disabling it entirely | 21:21 |
| ubot4 | Launchpad bug 500215 in apparmor "apparmor should warn users before allowing itself to be disabled/stopped" [Wishlist,Triaged] https://launchpad.net/bugs/500215 | 21:21 |
| ali1234 | micahg: yeah i saw it when logging in again :) | 21:24 |
| ali1234 | also, is the bug widely known about? i have a feeling a lot of people might be affected by it and no idea what's going on... | 21:25 |
| ali1234 | well, anyone who upgraded on day 1 basically | 21:25 |
| micahg | ali1234: it shouldn't be affecting too many people | 21:25 |
| * micahg needs to get an ubuntu blog... | 21:26 | |
| micahg | ali1234: it was fixed a month before release... | 21:26 |
| ali1234 | hmm..... well, must have regressed at some point then | 21:27 |
| ali1234 | although i don't know how | 21:27 |
| ali1234 | my other machine that i was using to test alphas (starting with a fresh install) was not affected, only the one i upgraded from jaunty | 21:28 |
| micahg | ali1234: do you still have your upgrade logs? | 21:31 |
| ali1234 | i didn't even know they existed | 21:31 |
| ali1234 | where would i find them? | 21:31 |
| micahg | /var/log/dist-upgrade/ | 21:32 |
| ali1234 | 20091029-0819 - so yeah | 21:32 |
| micahg | ok, so I might ask you for them later, I need to talk with someone about the issue | 21:33 |
| ali1234 | no problem | 21:34 |
| === rackIT is now known as rackIT_AFK | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!