/srv/irclogs.ubuntu.com/2008/04/10/#launchpad-meeting.txt

=== mwhudson__ is now known as mwhudson
=== mrevell is now known as mrevell-lunch
=== mrevell-lunch is now known as mrevell
=== salgado_ is now known as salgado
=== salgado is now known as salgado-lunch
=== salgado-lunch is now known as salgado
* Rinchen looks around and sees MootBot and breathes a sign of relief.18:47
mpt[WC] <- sign of relief18:57
intellectronicathanks for letting us know!18:58
Rinchenheh18:58
mptah, intellectronica, you missed the context by ten minutes :-)18:58
* thumper has coffee18:58
intellectronicai'm not sure whether i should be happy or sad about that18:58
intellectronicai'll default to happy18:59
salgadome19:00
schwukyou19:00
sinzuihim19:00
bacus19:00
thumperI19:00
thumperaye19:01
* allenap is feeding his daughter; 7pm is not such a good time for the meeting :)19:01
carlosme19:01
bigjoolsallenap: seconded19:01
al-maisanme19:01
thumper6am ain't so hot either19:01
gmbme19:01
leonardrme19:01
thumperRinchen: c'mon19:01
Rinchenvroom19:01
allenapthumper: granted19:02
bigjoolsthumper: I thought you always got up at 6am? :)19:02
* schwuk waits for rollcall19:02
Rinchengeez, anxious bunch19:02
flacosteme19:02
Rinchen#startmeeting19:02
MootBotMeeting started at 20:02. The chair is Rinchen.19:02
MootBotCommands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE]19:02
thumperbigjools: almost, but not always coherent19:02
RinchenWelcome to this week's Launchpad development meeting. For the next 45 minutes or so, we'll be coordinating Launchpad development.19:02
Rinchen[TOPIC] Roll Call19:02
abentleyme19:02
MootBotNew Topic:  Roll Call19:02
mrevellme19:02
gmbme19:02
Rinchenme19:02
jt1me19:02
flacosteme19:02
salgadome19:02
allenapme19:02
bigjoolsme19:02
marsme19:02
barryme19:02
schwukme19:02
mptme19:02
herbme19:02
leonardrme19:02
intellectronicame19:02
matsubarame19:02
Rinchen(and my beeps aren't working today again)19:02
bacme19:02
=== jt1 is now known as jtv
adeuringme19:02
BjornTme19:02
sinzuime19:02
cprovme19:02
gmbthumper: Is that a definition of bigjools? Cos that's how it reads ;)19:02
thumperme19:02
jtvme19:02
SteveAme19:02
kiko-afkme19:03
statikme19:03
EdwinGrubbsme19:03
Rinchenhmm.... ok, let's keep going!19:04
Rinchen[TOPIC] Agenda19:04
MootBotNew Topic:  Agenda19:04
Rinchen * Next meeting19:04
Rinchen * Actions from last meeting19:04
Rinchen * Oops report (Matsubara)19:04
Rinchen * Critical Bugs (Rinchen)19:04
Rinchen * Bug tags19:04
Rinchen * Operations report (mthaddon/herb)19:04
Rinchen * DBA report (stub)19:04
Rinchen * Sysadmin requests (Rinchen)19:04
Rinchen * New packages required (salgado)19:04
Rinchen * A top user-affecting issue (mrevell)19:04
Rinchen * Doc Team report (mrevell)19:04
Rinchen * Blockers19:04
Rinchen[TOPIC] Next meeting19:04
MootBotNew Topic:  Next meeting19:04
Rinchensame time, same channel?19:04
SteveAI'd prefer a tad earlier19:05
SteveAbit it's no biggie19:05
thumper:(19:05
kiko-afkSteveA, what about thumper?19:05
=== kiko-afk is now known as kiko
kikowell, last week we said we'd experiment rotating team leads19:05
kikothe team leads call, sorry19:05
kikoturns out we didn't agree to do that for this round19:05
Rinchenour next call is on Wednesday of next week so we can bring the topic up again19:06
SteveAso, as the meeting is late for europeans, I ask that we keep it firmly on time19:06
kikoI really want a time that works for tim and jtv19:06
SteveAand end exactly 45 minutes after starting19:06
kikowe make a serious effort to end 45 minutes after19:06
SteveAwell19:06
SteveAwe just end then ;-)19:06
SteveAand take the consequences19:06
Rinchen[AGREED] Same time next week19:06
MootBotAGREED received:  Same time next week19:06
Rinchen[TOPIC] Actions from last meeting19:06
MootBotNew Topic:  Actions from last meeting19:06
Rinchennone19:06
Rinchen[TOPIC] Oops report (Matsubara)19:06
MootBotNew Topic:  Oops report (Matsubara)19:06
Rinchen(vrooom)19:06
matsubaraToday's oops report is about bugs 214843, 19836819:07
* Rinchen chuckles.19:07
ubotuLaunchpad bug 214843 in launchpad "Cycles in team data kill mailing lists" [Critical,Confirmed] https://launchpad.net/bugs/21484319:07
ubotuBug 198368 on http://launchpad.net/bugs/198368 is private19:07
matsubara#214843 is already assigned do barry.19:07
matsubarato barry19:07
matsubarabarry: any news about #198368?19:07
barryi think the data problem is already fixed, as is the resync issue.  we now have bug 215118 in review19:07
ubotuLaunchpad bug 215118 in launchpad "Odd mailing list resync behavior" [High,Confirmed] https://launchpad.net/bugs/21511819:07
barryonce that lands we should be good to go19:08
barrythat's it19:08
matsubarathanks barry19:08
matsubaraback to you Rinchen.19:08
Rinchenthanks!19:08
Rinchen[TOPIC] Critical Bugs (Rinchen)19:08
MootBotNew Topic:  Critical Bugs (Rinchen)19:08
Rinchen[LINK] https://bugs.edge.launchpad.net/launchpad/+bug/20665419:09
RinchenThis is the memory issue. flacoste, SteveA.  Any updated status?19:09
MootBotLINK received:  https://bugs.edge.launchpad.net/launchpad/+bug/20665419:09
Rinchen[LINK] https://bugs.edge.launchpad.net/launchpad-bazaar/+bug/20755819:09
Rinchen thumper, what's the next step with this?19:09
MootBotLINK received:  https://bugs.edge.launchpad.net/launchpad-bazaar/+bug/20755819:09
kikoI'm really unhappy about that one19:09
flacosteRinchen: mthaddonis copying the references leak log19:09
ubotuLaunchpad bug 207558 in bzr ""bzr branch http://bazaar.launchpad.net/...." fails with bzr.dev >= r3309" [Critical,Fix released]19:09
thumperI think that was just with bzr.dev19:09
SteveAkiko: what one?19:09
thumperand is fixed19:09
flacosteRinchen: i'll investigate after that, but we are not seeing any big leak on the monitoring server, probably due to the fact that it receives a lot less requests19:09
Rincheninteresting flacoste19:09
kikothe server issue, SteveA19:10
Rinchenthumper, can you please update the bug then since there is a code hosting task assigned?19:10
thumperah19:10
kikowe need a new strategy for dealing with it19:10
SteveAkiko: it's a tricky one, for sure19:10
thumperI recall now19:10
flacosteRinchen: i'm hoping to correlate a small leak in the monitoring log to the big one based on the difference in number of requests19:10
thumperlifeless filed an RT to get apache config updated19:10
SteveAkiko: what are you talking about?19:10
kikoSteveA, well..19:10
SteveAkiko: we have a strategy for dealing with it, which is in progress19:10
kikoSteveA, is that strategy going to work? it seems we've made the change and the change we've made has hidden the problem from that environment.19:11
SteveAkiko: I'd appreciate a bit more support and positivity considering the work people are putting into diagnosing this19:11
kikomeanwhile, clock ticking, etc19:11
thumperRinchen: confirmed with config that the RT has been executed19:11
SteveAwe don't know that the change has hidden anything19:11
* thumper updates task19:11
Rinchenthumper, thank you sir.19:11
flacostekiko: well, since the problem now disapperaed from edge, maybe it will go away by itself in the next roll-out ;-)19:12
SteveAwe are collecting data, and we need to analyze it and tweak how we're collecting it based on the analysis19:12
kikoSteveA, that's all just words to me. I haven't seen us progress to a solution after more than a week, and that's what's concerning me19:12
SteveAkiko: I have no idea what you're talking about19:12
kikoflacoste, completely? no edge instances are balooning?19:12
* kiko chuckles19:12
kikoheisenbug indeed19:12
flacostekiko: i wouldn't count on it though, because until last week, that was isolated to edge19:13
flacostekiko: and it's not like a cherry-pick could explain it19:13
Rinchenflacoste, perhaps it time for a quick update on status to be posted to the bug. That way kiko will see the work you and others are doing.  Would you be ok with that?19:13
kikoflacoste, f*cking crazy.19:13
flacosteRinchen: sure19:13
kikoI know what work's been done, though.19:13
kikothat's not really the issue :)19:14
kikobut go ahead, it'll be interesting19:14
flacostewhat kiko wants is results!19:14
SteveAwhat kiko wants is a vacation somewhere sunny19:14
kikoor really good excuses :)19:14
Rinchen[LINK] https://bugs.edge.launchpad.net/launchpad/+bug/21484319:14
Rinchenbarry, how is the refactoring going?19:14
MootBotLINK received:  https://bugs.edge.launchpad.net/launchpad/+bug/21484319:14
ubotuLaunchpad bug 214843 in launchpad "Cycles in team data kill mailing lists" [Critical,Confirmed]19:14
kikoit's always sunny where /I/ live19:14
flacosteSteveA: he just needs a vacation then I guess, i heard it was very sunny where he lives19:14
matsubaraRinchen: barry just gave a update about that one in the oops section :-)19:15
SteveAflacoste: the memory problems have only happened when kiko's not been on vacation.  so obviously that is the solution.19:15
kikocan I not be the subject of the conversation, and instead the problem we're trying to solve?19:15
barryRinchen: the refactoring has not been done and probably isn't critical19:15
barryRinchen: it can be done this cycle as normal19:15
kikoflacoste, can you go on describing where we are?19:15
Rinchenbarry, great...that means we can demote the bug to HIGH.  Would you mind doing the honours?19:15
flacostekiko: i'll analyse the current data we collect and see if there is anything interesting19:16
barryRinchen: done19:16
Rinchenbarry, thank you sir19:16
kikoflacoste, okay. is there a way we could monitor across multiple threads?19:16
flacostekiko: no19:16
kikobummer19:17
SteveAwell19:17
SteveAthere are ways.  it's a lot more complicated.19:17
kikoyeah, I had that feeling19:17
flacostekiko: we could bump up the number of requests it receives19:17
kikothat's a great idea! can we do that mthaddon?19:17
SteveAso we should only go there once we've established we can't get decent data the way we're doing it19:17
kikoright19:17
SteveAfirst, let's have flacoste analyze the data we have19:17
mthaddonkiko, what's that, bump up the number of requests?19:18
SteveAthen we'll look at increasing the number of requests it gets19:18
sinzuiI blame Scho√ędinger's cat19:18
mthaddonkiko, we tried that (from 1/16 -> 1/10) and had to revert since it was timing out19:18
kikomthaddon, make it more preferred in the server rotation I guess19:18
kikomthaddon, hah19:18
kikomthaddon, is there a smaller increment?19:18
* kiko high 5s jtv 19:18
jtvhuh?19:19
mthaddonkiko, any increment you like - 1/15 for instance... :)19:19
kikomthaddon, yeah, we could do a smaller step to see. okay, let flacoste make that call once he's looked at the data.19:19
mthaddonkiko, sounds good19:19
SteveAfrom looking at the data, we may see that we want to look more closely at *particular* kinds of request19:19
SteveAso, I'm not so interested in adding shavings to the number of requests it processes19:20
SteveAbut rather directing requests that matter to it19:20
SteveAwe'll know soon19:20
Rinchenok, thanks for the lively discussion....19:20
RinchenI'm going to move on19:20
Rinchen[TOPIC] Bug tags19:20
MootBotNew Topic:  Bug tags19:20
Rinchennone to be considered19:20
Rinchen[TOPIC] Operations report (mthaddon/herb)19:21
MootBotNew Topic:  Operations report (mthaddon/herb)19:21
herb2008-04-07 - Cherry pick 6013 to lpnet*19:21
herb2008-04-09 - Cherry picks 6006, 6026, 6031 to forster, lpnet*, xmlrpc-internal, mailman19:21
herbMemory utilization issues continue, which has forced us to restart multiple lpnet instances daily.19:21
herb... But we have a debug release in place on edge3 and lpnet6. As discussed above, it doesn't seem we're getting any real feedback because the debug instances are throttled due to running them with 1 thread. They don't seem to be getting enough traffic to create the memory leak.19:21
herbThat's it from Tom and I unless there are any questions.19:21
RinchenThis is a good time to talk about cprov's item19:21
cprovyup19:22
Rinchenupdate-cache19:22
mthaddonwhat's to talk about?19:22
mthaddonsounds like looptuner (sp?) is the way to go, right?19:22
cprovmthaddon: exactly, we have to find a way to reduce the commit-rate19:22
jtvmthaddon: spelled right19:23
Rinchencprov, so it sounds like the email threads have been updated and we can skip this topic.19:23
Rinchen..yes?19:23
mthaddonype19:23
mthaddonyep, even19:23
kikoI just replied to cprov and we should indeed use looptuner19:23
cprovRinchen: yes, all sorted19:23
kikoit's easy to do19:24
kikogmb, can you help him out?19:24
Rinchengreat thanks gents19:24
gmbkiko: Will do.19:24
Rinchen[TOPIC] DBA report (stub)19:24
MootBotNew Topic:  DBA report (stub)19:24
kikoi know you made that change for update-bugtargetnames19:24
kikothanks so much19:24
Rinchenno stub19:24
SteveAon vacation19:24
SteveAor leave19:24
SteveAor whatever19:24
Rinchenok, we'll skip the report this week.19:24
SteveAanyway, not at work today19:24
Rinchen[TOPIC] Sysadmin requests (Rinchen)19:24
MootBotNew Topic:  Sysadmin requests (Rinchen)19:24
RinchenHi! Is anyone blocked on an RT or have any that are becoming urgent?19:25
Rinchenthat I don't already know about19:25
RinchenIS is still congested due to Hardy19:25
Rinchenok, if you do have something, ping me later or email me please. Thanks19:26
Rinchen[TOPIC] New packages required (salgado)19:26
MootBotNew Topic:  New packages required (salgado)19:26
* bigjools waves at Rinchen19:26
salgadoif any of the branches you're working on right now depends on any library which is not part of the launchpad-dependencies package, come talk to me ASAP.19:26
* kiko looks around19:26
kikosalgado, do you know if we still require lxml?19:26
* Rinchen acks bigjools and the ppa issue19:26
salgadofyi, I recently made the package depend or pgsql 8.2 or 8.3, so you don't need to keep both versions around anymore19:26
salgadokiko, I don't19:26
kikoI'm asking because I know that abel isn't using it to validate hwtest schema anylonger19:27
kiko(is that right flacoste?)19:27
adeuringsalgado, kiko: I think I was the only one who used it19:27
jtvkiko: we still require xmlproc.19:27
Rinchenright and there's a bug open for this cycle to remove that dependency19:27
flacostekiko: i want to use it for other stuff19:27
Rinchenit's the only other hardy item on my list19:27
flacostekiko: like pagetests helpers and other general XML usage19:27
bigjoolswhat;'s going on with python-xml in hardy now?19:28
Rinchenhe newest python-xml package in hardy (uploaded April 1st) has made it impossible to run/test Launchpad because it moved its installed files to a place where we can't easily import from.19:28
kikoflacoste, yep, that's what I was trying to remember19:28
Rinchentranslations has a dependency on xmlproc which, when removed, should allow LP to run/test fine on dev machines19:29
Rinchenanything further for salgado?19:29
kikowill it be removed, though?19:29
Rinchentranslations has a bug open to remove the dep19:29
Rinchenthe package itself will remain in hardy19:30
jtvkiko: we definitely plan to19:30
Rinchenok, onward! (to make up for the last few meetings which overran)19:30
kikojtv, milestone eta?19:30
Rinchen[TOPIC] A top user-affecting issue (mrevell)19:30
MootBotNew Topic:  A top user-affecting issue (mrevell)19:30
Rinchenkiko, this cycle19:30
mrevellHi. I don't have one outstanding user-affecting issue this week. However, there are a couple of issues I've seen that I think are of interest.19:31
mrevell1. We've had a request from beeseek to import mbox files from their existing mailing lists into their Launchpad-hosted mailing lists. Barry, is this something we can/will do?19:31
kikocool cool19:31
barrymrevell: yes.  matsubara is this the same request we're working on?19:31
Rinchenkiko, https://bugs.edge.launchpad.net/rosetta/+bug/21340619:31
matsubarabarry: yes19:31
barrymrevell: ^^19:31
mrevellbarry: Ah, thanks. Would you be happy to do this generally? If so, I'll update ListHelp.19:32
* beuno has been holding back from using LP mailing lists due to not being able to migrate archive/subscribers from lists.ubuntu19:32
matsubaramrevell: I talked to barry about that already. we're going to do that after the meeting with mthaddon help :-)19:32
mrevellmatsubara: Ah, okay :) Thanks!19:32
mrevell2. Bug 213430 requests that "request feedback" in Blueprint should trigger an email notification.19:32
ubotuLaunchpad bug 213430 in blueprint ""Request feedback" should send an e-mail to the person who is asked" [Undecided,New] https://launchpad.net/bugs/21343019:32
barrybeuno: migrating subscribers isn't planned, but we can do mboxes19:32
mrevellJust thought I'd highlight that request.19:32
beunobarry, can you plan it?  :)19:33
mrevellRinchen: Do you want to talk about bug 213406?19:33
barrybeuno: let's chat after this meeting19:33
mthaddonI'd also like to mention https://answers.launchpad.net/launchpad/+question/29265 as well - registering PLD Linux as a distro19:33
ubotuBug 213406 on http://launchpad.net/bugs/213406 is private19:33
kikoyep19:33
beunobarry, sure, thanks  :)19:33
kikoI'll do that today, mthaddon19:33
Rinchenmrevell, no, that's ok. Thanks :-)19:33
mthaddonkiko, cool, thx19:33
intellectronicamrevell: i'm aware of this, but i doubt that we'll have time to look at it for some time19:33
mrevellRinchen: Thanks for pointing me at it.19:33
mrevellintellectronica: Thanks. It'd be cool when it arrives.19:34
intellectronicamrevell: just like many other blueprint feature requests19:34
RinchenI'm hoping to have more time applied to Blueprint love after our currently schedule work completes.  *fingers crossed*19:34
mrevellThat's it from me, thanks Rinchen.19:34
RinchenThank mrevell19:34
Rinchen[TOPIC] Doc Team report (mrevell)19:34
MootBotNew Topic:  Doc Team report (mrevell)19:34
mrevellAh, me again19:34
mrevellThis week mpt and I met with the designers who are working on our new Launchpad tour. We spent most of the day with them, explaining the concepts behind LP and the ten primary features we want to communicate in the tour.19:34
kikomrevell, what do you think of them?19:34
mrevellkiko: I was impressed with how quickly they "got" the concepts behind LP.19:35
kikothat's cool19:35
mrevellkiko: They had good input into how we could represent them visually, too.19:35
mrevellI'll post a report on the meeting to the list.19:35
mrevellPlease, if you're a team lead, could I have a fifteen minute call with you tomorrow or early next week to go over the parts of the tour relevant to your part of LP?19:35
mrevellPing me with some good times :)19:35
Rinchenmrevell, and grab someone from soyuz too please :-)19:36
mrevellI'll be posting a new draft of the tour content to the team list either tomorrow or Monday. The tour designers are going to give us some mock-ups next week, which I'll post to the list.19:36
mrevellRinchen: Will do :)19:36
* Rinchen loves mock-ups.19:36
mrevellThanks, back to you Rinchen.19:36
RinchenThanks mrevell19:36
Rinchenwow, we're really zooming today19:36
Rinchen[TOPIC] Blockers19:36
MootBotNew Topic:  Blockers19:36
thumperCode: not blocked19:36
flacosteFoundations: not blocked19:37
flacosteactually19:37
mptI'm not blocked, but bug 214690 will make me much slower fixing anything that involves CSS19:37
ubotuBug 214690 on http://launchpad.net/bugs/214690 is private19:37
kikowhat's that bug about?19:37
jtvTranslations: not blocked19:37
BjornTBugs: not blocked19:37
bigjoolsSoyuz: just the usual19:37
mptkiko, launchpad.dev using style-slimmer.css instead of style.css19:37
mptkiko, so I think it's about launchpad.dev's config19:37
RinchenReleases Team: Not blocked.19:37
flacostei see nothing in the config related to that19:38
flacostei thought that was an apache-side issue19:38
SteveAI think I kniw how that works19:38
kikompt, I think nobody else is aware of that problem19:38
statiklpcomm: no blocked19:38
SteveAI'll look into it shotyly19:38
Rinchenthanks SteveA19:38
mptkiko, I discovered it only yesterday19:38
mptthanks SteveA19:38
sinzuiDoes slimmer rely on devmode == False19:38
flacostei thnk sinzui got an hypothesis here19:39
schwukhwdb: not blocked19:39
sinzuiI have not messed with that until today, and it absolutely still works in the branch that removes launchpad ZConfig.19:40
BjornTmpt: can't you run scripts/make-static.css as a workaround? if it works, it's at least better than restaring the web app19:40
BjornTs/css/py19:40
mptI don't know -- I'll try it19:41
RinchenOk, I'm going to wrap up now19:41
mptthanks for the hint19:41
Rinchenfeel free to chat about this on -code :-)19:41
RinchenThank you all for attending this week's Launchpad Developer Meeting. See the channel topic for the location of the logs.19:41
jtvThank you!19:41
Rinchen#endmeeting19:42
MootBotMeeting finished at 20:42.19:42
thumper\o/19:42
mrevellthanks Rinchen19:42
Rinchenyo19:42
jtvEarly even.19:42
Rinchenvrooooom19:42
mrevellthanks everybody19:42
Rinchen;-)19:42
statikthanks rinchen19:42
cprovgreat !19:42
intellectronicacheers Rinchen19:42
gmbCheers Rinchen.19:42
abentleyThanks Rinchen19:42
* gmb -> evening things19:42
cprovthanks everybody19:42
=== cprov is now known as cprov-out
Rinchenthumper <-> rockstar_,  rockstar_ <-> thumper19:42
al-maisanthanks!19:42
thumperRinchen: we know19:42
Rinchen:-)19:42
rockstar_Rinchen, yea, we've met!  :)19:42
kikomoonies19:42
Rinchenwhere's my bandana ?19:43
thumperRinchen: bandana?19:43
* gmb misread that as the Pascal version of thumper != rockstar19:43
thumpergmb: and that'd be right19:43
bigjoolsRinchen's symbols look like an Imperial Tie Fighter19:43
thumper<o>19:43
thumper|-o-|19:43
bigjools!19:44
bigjools+119:44
bigjoolsperfect :)19:44
rockstar_thumper, the first is an interceptor19:44
Rinchenwell, I was using the vader model19:44
bigjoolslol19:44
rockstar_Only Vader flew one of those until RotJ19:44
Rinchentie interceptor maybe19:44
rockstar_Yup19:44
* Rinchen has trouble remembering imperial ship names19:44
rockstar_|-oo-|  <-- TIE Bomber19:45
bigjoolsI'm not nearly geeky enough to remember19:45
* rockstar_ played too much Rogue Squadron19:45
thumperme neither19:45
* thumper hasn't played a game in ages19:45
rockstar_Those movies are all older than me.19:45
bigjoolsnot me :(19:45
rockstar_thumper, your kids don't game?19:45
thumperrockstar_: oldest is 719:46
bigjoolsWii bowling is my limit these days19:46
thumperso they play Clifford games, and my little pony19:46
thumpernot something that I get involved with19:46
rockstar_thumper, I was 4 when I started playing.  Got into computers/electronics because my dad would unhook the stuff from the TV, so I taught myself to hook it back up.19:46
bigjoolscome now thumper don't be shy19:46
thumperI'd be happy to get a games console later19:47
thumperI remember with fondness the an old playstation game that I can't remember the title to19:47
thumpera problem solving one where you were some skeleton19:48
bigjoolsRealistic Breast Physics III ?19:48
rockstar_thumper, our LUG here gets together for Warhawk pretty often (everyday lately...)19:48
thumperlol19:48
* thumper has never heard of warhawk19:48
rockstar_New PS3 game.19:48
* thumper wonders if we should really have a #launchpad-offtopic channel19:50
rockstar_:)19:50
matsubaraCOD4 ftw20:05
=== salgado_ is now known as salgado
=== salgado is now known as salgado-afk

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!