#mer log for Tuesday, 2012-03-27

*** tilgovi_ has quit IRC00:04
*** JRepin has joined #mer00:09
*** JLP has quit IRC00:09
*** jluisn has quit IRC00:11
*** clopez has joined #mer00:12
*** E-P has joined #mer00:13
*** himamura has joined #mer00:14
*** JRepin has quit IRC00:16
*** raignarok has quit IRC00:16
*** JRepin has joined #mer00:17
*** tilgovi has joined #mer00:17
*** dijenerate has quit IRC00:29
*** KaIRC has quit IRC00:44
*** dijenerate has joined #mer00:45
*** tilgovi has quit IRC00:53
*** ALoGeNo has quit IRC01:10
*** zmc has joined #mer01:13
*** eg81 has quit IRC01:16
*** eg81 has joined #mer01:17
*** ALoGeNo has joined #mer01:21
*** Joye has joined #mer01:28
*** Joye has quit IRC01:38
*** Joye has joined #mer01:46
*** sandroandrade has quit IRC02:09
*** clopez has quit IRC02:17
*** Joye has quit IRC02:20
*** new_ has joined #mer02:35
*** new_ is now known as xuc_02:37
*** xuc_ is now known as xuchao_02:38
*** xuchao_ is now known as xucdsp_02:38
*** clopez has joined #mer02:40
*** M4rtinK has quit IRC02:50
*** clopez has quit IRC03:12
*** tilgovi has joined #mer03:21
*** Jade has quit IRC03:31
*** clopez has joined #mer03:37
*** clopez has quit IRC03:43
*** clopez has joined #mer03:44
*** Jade has joined #mer04:08
*** dijenerate has quit IRC04:12
*** dijenerate has joined #mer04:24
*** clopez_ has joined #mer04:51
*** jpwhiting has quit IRC04:57
*** clopez_ has quit IRC05:00
*** otep has quit IRC05:17
*** otep has joined #mer05:21
*** tomeff1 has quit IRC05:23
E-PMorning05:30
*** clopez has quit IRC05:30
Stskeepsmorn E-P05:31
*** wubudubudubudu has quit IRC05:38
*** wubudubudubudu has joined #mer05:46
E-Per, what is this: http://wiki.merproject.org/wiki/Diet_program_and_Excessive_fat_Burn05:48
E-PI searched a work "quality", and one result was that :)05:49
Bostiklovely spam, wonderful spam...05:51
E-Pactually, lots of spam: http://wiki.merproject.org/wiki/Special:Deadendpages05:52
*** tomeff has joined #mer06:04
vgrademorniing06:05
*** u1106 has joined #mer06:10
*** niqt has joined #mer06:20
*** furikku has joined #mer06:22
*** tomeff has quit IRC06:32
*** jukkaeklund has joined #mer06:35
*** pohly has joined #mer06:39
xucdsp_hello everyone,i find the examples in connman source code is written in python, but i want to use connman in my c++ program,could you help me?06:44
*** jpwhiting has joined #mer06:45
*** jukkaeklund has quit IRC06:47
xucdsp_or are there any document of c api of  connman?06:49
xucdsp_by the way, i am a member of sonach's team06:55
arfollxucdsp_: you probably should be use the dbus api of connman06:56
Bostikc api? I thought connman was exposed over dbus06:56
*** cxl000 has quit IRC06:56
arfollxucdsp_: http://wiki.meego.com/D-Bus/ConnMan06:56
xucdsp_shoud i use dbus to control connman?06:57
arfollyes06:58
xucdsp_oh thanks, i'll read this doc first~06:58
*** leinir has joined #mer07:09
*** leinir has quit IRC07:09
*** leinir has joined #mer07:09
*** jukkaeklund has joined #mer07:09
*** ericlr has joined #mer07:10
jukkaeklundmarquiz, hi07:10
marquizhello07:10
marquizwhat's on your mind? ;)07:11
*** dcthang has joined #mer07:11
jukkaeklundyeah, meant to ask what is the correct way of removing moslo from n9?07:12
jukkaeklundnow we have just re-flash whole thing in wiki page07:13
Stskeepsxucdsp_: any examples that exist out there for connman with dbus under C (glib) should be good07:13
jukkaeklundwrong channel, I know.. hope they don't kick me :p07:13
xucdsp_yes, document get me confused...07:15
*** raignarok has joined #mer07:16
Stskeepsxucdsp_: is your C++_app Qt-based or not?07:17
xucdsp_not qt-based~07:17
Stskeepsok07:17
*** slaine has joined #mer07:19
slainemorning all07:19
jukkaeklundmarquiz, see above07:20
*** ahma has joined #mer07:20
ahmalo07:21
Bostikmorning07:21
ahmaback from my well ear..spent vacation07:21
StskeepsBostik: did you slay the openid beast?07:21
BostikStskeeps: nope, had no time on weekend and certainly no time yesterday :(07:21
Bostikwhen it's not work issuing demands, it's the family07:22
Stskeepssure, just checking up status07:22
Bostikonce I get my hands free from current burden, I'll try again07:22
slaineBostik: I'm in a similar position07:23
slainewith regards demands07:23
*** andre__ has joined #mer07:25
*** andre__ has joined #mer07:25
BostikStskeeps: it might be reasonable to wait with qtbase integration in any case, the most recent release list mails indicate they're going to freeze qtbase (or at least one branch) for Qt5-alpha-real in very near future07:26
Bostikas in days07:26
*** ALoGeNo has quit IRC07:26
Stskeeps:nod:07:26
Stskeepseglibc is proving a bigger headache than intended, but should yield a lot of improvements07:26
niqtvgrade: has mer x86 (for plasma) fix wifi problem ( i see your twit for vivaldi) ?07:27
*** harbaum has joined #mer07:27
Bostikahma: I'll nudge you up to speed with specworks tomorow afternoon... I hope07:27
Bostiktook over for a while and did some minor cleanups07:28
StskeepsBostik: i think we should put things to review early anyway, doesn't have to be today, sage's off :P07:28
Bostikexcellent07:28
BostikI have a deeply technical spec to read and digest :)07:28
ahmaBostik: nice07:29
ahmaminor or "minor"? :)07:29
Bostikminor, mostly07:29
Bostikjust check my git repo for commits07:29
ahmaI'll do that07:30
*** JRepin has quit IRC07:31
*** wmarone_ has joined #mer07:33
*** wmarone has quit IRC07:33
* Stskeeps waits for tizen conference mails07:36
timophsent a proposal?07:36
timophyou should redo the tranparency talk there07:37
* timoph liked that07:37
timophand good morning07:37
marquizjukkaeklund: there's no use in flashing moslo to n907:38
Stskeepstimoph: i proposed one on cross compilation07:40
jukkaeklundmarquiz, yeah.. my bad I did that07:40
marquizjukkaeklund: but, if you've (for some reason) flashed it (to the moslo partition), you can remove it in similar way as for n9007:40
marquizn950, that is07:40
Stskeepstimoph: if i had to do the transparency talk for tizen, it'd right now be "what transparency?07:40
Stskeeps:P07:40
slainehehe07:41
marquizjukkaeklund: one thing that is missing is how to remove the Nemo partition, too (without re-flashing)07:41
jukkaeklundyep07:41
jukkaeklundso flash with -L and empty .img?07:41
marquizjukkaeklund: one can do that from moslo console (by telneting in), but no instructions, yet07:41
marquizjukkaeklund: yeah, flash empty.img with -L07:42
jukkaeklundour wiki would benefit from instructions how to remove the altos partition07:42
marquizjukkaeklund: true, i could write initial instructions today (evening)07:42
marquizjukkaeklund: there's a nice command, already (but that has to be run from the telnet console)07:43
jukkaeklundthats nice, send to me or put straight to wiki :)07:43
*** ALoGeNo has joined #mer07:43
*** ALoGeNo has joined #mer07:43
marquizjukkaeklung i don't remember  that out of my head ;)07:43
marquizjukkaeklund: well, telnet in the device07:44
marquizand run autopartition.sh07:44
marquizjukkaeklund: that is, see it's help07:46
*** pirut has joined #mer07:47
* Stskeeps is looking forward to getting closer to linaro's toolchain07:51
Stskeepsless surprises in toolchain bugs07:51
*** notmart has joined #mer07:52
*** phdeswer has quit IRC07:55
*** jonnor_work has joined #mer07:58
timophStskeeps: :D07:58
*** norayr has joined #mer07:59
*** Joye has joined #mer08:04
*** jukkaeklund has quit IRC08:05
Joyehelp08:07
StskeepsJoye: no need to say help, just ask your question :)08:09
*** Eren has joined #mer08:09
Joyesorry, type wrong command. new for irc08:10
Stskeepshehe, it's okay :)08:11
*** bergie has joined #mer08:21
Stskeepslbt: i think we're OK for a prerelease08:25
*** sonach has joined #mer08:30
*** cxl000 has joined #mer08:32
*** smoku has joined #mer08:34
sonach恩,这个问题别人可能没法真回答。。。如果有具体问题就好了:) 对了,你问的是Stskeeps吗?08:34
Stskeepssorry, my babelfish must be malfunctioning today :)08:35
sonach:( sorry for mis-typing08:35
Stskeepshehe, no problem, i was making a joke ;)08:35
*** slx is now known as swerden08:36
Stskeepsbabelfish is a fictious creature fish from book "hitchhikers guide to the galaxy", which you put in your ear and it translates all the world languages to you :)08:36
sonachhehe,interesting!!08:36
*** mdfe_ has joined #mer08:39
*** jukkaeklund has joined #mer08:43
*** dazo_afk is now known as dazo08:46
*** tsdedst has quit IRC08:48
*** bergie has quit IRC08:48
alteregolol08:49
* alterego just saw babelfish jokes08:49
AardStskeeps: s/world/universe/08:51
*** bergie has joined #mer08:51
Stskeepspoint08:51
alteregoWell galaxy08:51
alteregoThere's no scientific evidence to suggest it'll work outside of the milkyway.08:52
Aardget funding for a research project?08:53
*** JRepin has joined #mer08:54
sonachStskeeps: One basic question. Why did modern linux components, such as connman and bluez,  use d-bus instead of classic Posix IPC message queue?08:56
Stskeepssonach: ah, now that's a good question..08:57
Stskeepsi don't actually know :)08:57
sonachah?08:57
BostikI would suspect the built-in marshaling of variable types has something to with it08:57
sonachBostik: what is the advantage of "built-in marshaling of variable types"? programmers can ignore byte-alignment based on this?08:59
Bostiksonach: have you ever transferred a dictionary over IPC? ;)09:00
*** tsdedst has joined #mer09:00
Bostik'dictionary' as in python dict() or perl hash09:01
*** wool has quit IRC09:02
*** Timmy has joined #mer09:02
*** Timmy has joined #mer09:02
sonachBostik: great explanation!  so when we need to transfer complex data structures, or we don't really know what the data struct is,  we use d-bus instead of classic IPC.09:03
Stskeepssonach: there's also other parts such as launching processes to receive dbus information09:03
Stskeepsinstead of with posix ipc where they have to be up all the time09:04
Bostiknot to mention some bloody handy bindings for multiple languages that "just work [TM]"09:04
Bostikso for anything which doesn't need high-bandwidth or low-latency IPC, dbus is usually a lot simpler and allows more flexibility without much thought09:05
sonachcool:)09:05
sonachBostik: but if we don't use bindings, just use the low-level dbus APIs, it seems that it is a lot complex than classic IPC.09:06
*** notmart has quit IRC09:07
*** notmart has joined #mer09:10
*** notmart has joined #mer09:10
*** JRepin has quit IRC09:16
*** jonnor_work has quit IRC09:16
*** arcean has joined #mer09:16
Bostiksonach: oh yeah, and the final thing which is really neat on desktops (and mobiles) is that dbus has this neat concept of "broadcast messages" (signals) that may be used by one or many applications09:16
niqtsomeone has installed mer on trim slice?09:16
harbaumimho vgrade has09:21
slaineniqt: grade has, I think cxl000 is the one that's done most of the HA work09:22
slainevgrade09:22
slainedamn spell checker09:22
niqt I'm considering if buying one09:24
niqtbut maybe I do too many things for the time that i have09:25
slainenice little device09:25
slaineniqt: that's exactly my problem, lol09:25
niqt:)09:25
*** KaIRC has joined #mer09:26
*** GeneralAntilles has quit IRC09:27
*** GeneralAntilles has joined #mer09:28
slaineRight, so, back to OBS install09:38
slaineI did a copypac of acl into my server yesterday09:38
slainesomething interesting happened09:38
slaineIt seemed to work from a copying point of view, but I don't seem to be able to get the acl package to build09:38
slaineNo option to schedule a rebuild etc.09:39
slaineit's just listing it as a source package09:39
Stskeepsdid you add the repository ?09:39
slaineYup, possibly incorrectly09:40
slaineStskeeps: http://pastebin.com/CDsyzTun09:40
*** sonach has left #mer09:41
notmartany idea about this during image creation from the sdk? Error <mount>: Error creating ext3 filesystem on disk /dev/loop0209:41
Stskeepsnotmart: do a reboot, does it continue?09:41
Stskeepsslaine: looks correct09:41
notmartie i don't have /dev/loop02, i guess is just a distro difference, so making it use /dev/loop2 should be the same?09:41
notmartStskeeps: yes, i did a reboot09:41
Stskeepsnotmart: hmm09:41
Stskeepswe discussed this problem some days ago09:42
Stskeepscheck log from 03-2309:42
Stskeeps(is in topic)09:42
Stskeepsslaine: osc ls MDS:Core:i586 works i presume?09:43
* notmart checks09:43
slaineStskeeps: yup09:43
Stskeepsslaine: is your OBS scheduler running?09:44
slaineWondering of the worker isn't working properly09:44
slainenod09:44
Stskeepsi586 one too? ps aux | grep bs_sched on where it's supposed to be09:44
slaineyup09:45
Stskeepsokay09:45
slainesigner is marked as off09:45
Stskeepscheck it's log?09:45
slaineYeah, problems09:46
slainethanks09:46
*** tsdedst has quit IRC09:53
slaineStskeeps: I'm seeing a 503 in the fakeobs logs09:55
slaineobsbevm.labs.lincor.com - - [27/Mar/2012 10:54:30] "POST /public/lastevents HTTP/1.1" 503 -09:55
Stskeepsthat one is OK09:55
slaineok09:55
Stskeepslong story short the OBS has retarded remote link behaviour09:55
Stskeepsand i work around it with giving it a 50309:55
slaineSo all the 503 errors in the schedular logs are expected09:56
slaineresponse from watcher for http://mdsvm.labs.lincor.com:8001/public09:56
slaine503 remote error: Service Unavailable09:56
Stskeepsthat's ok09:56
Stskeepsyes09:56
slaineHmm, can't see anything else10:02
Stskeepsok10:03
Stskeepsls -l obs-repos/ for me?10:03
*** Eren has quit IRC10:06
*** Joye has quit IRC10:07
slaineon the mds box ?10:07
slaineI'm thinking it's config related10:07
slaineon the backend server10:07
slaineI should see the option to rebuild acl10:07
Stskeepson the mds box10:07
slainehttp://pastebin.com/CNMkgtrw10:10
Stskeepslooks correct10:11
slainenod, like I said, my hunch is something isn't right on the backend server10:11
Stskeepsslaine: scheduler_i586.log10:12
slainelike I said, it's only got those 503 errors10:13
slaineeverything else looks fine10:13
Stskeepsok10:13
Stskeepstry to remove your <repository> and watch the scheduler10:14
Stskeepslog10:14
slaineI'll tail that log, remove the acl project and reimport it10:14
Stskeepsand re-add it10:14
Stskeepsacl project?10:14
Stskeepspackage you mean10:14
slainesorry, yeah10:14
slaineStskeeps: yeah, something not right there10:19
slainehttp://pastebin.com/dRr17tJf10:19
Stskeepsyour fakeobs down?10:19
slainehope10:19
* lbt turns up10:19
slainea few 200 ok and 50310:20
Stskeepsmorn lbt, late night?10:20
lbtyeah - so I had a lie-in with a new book10:20
slaineStskeeps: I meant nope10:21
slainemy osc ls still works against MDS:Core:i58610:21
Stskeepsawkward10:22
slainenod10:24
* Stskeeps ponders idly10:24
slaineand still no obs fe option to schedule a rebuild10:24
slaineso something is up with the scheduler10:24
Stskeepscurl of mdsvm.labs.lincor.com:8001 works?10:24
slaineworks in that it gives a 40410:26
Stskeepsok10:26
Stskeepswhy the heck does it say connection refused..10:26
slainenod10:26
slaineespecially when I see "obsbevm.labs.lincor.com - - [27/Mar/2012 11:15:50] "GET /public/source/Core:i586/acl?expand=1&rev=latest HTTP/1.1" 200 -" in the fakeobs console10:27
Stskeepsright, but you should also see /public/build/ too10:27
*** dcthang has quit IRC10:27
slainecould be onto something, no public/build at all10:27
Stskeepslbt, and others, RM in 30 minutes?10:29
lbtwow - I *did* get up late!10:29
slaineStskeeps: got an example public/build/ url I can curl ?10:30
*** lizardo has joined #mer10:31
Stskeepsslaine: try with a second package10:38
jussiSage_: o/10:39
Sage_jussi: o/10:39
slaineStskeeps: weird10:40
Stskeepsworks?10:40
slaineyeah, I pulled in zlib and now acl and zlib are scheduled for rebuild10:42
Stskeepsmust have been an old inconsistent restate10:42
slainebuild is blocked though10:43
Stskeepswhat release are you building against?10:44
Stskeepsdoes it say scheduled, or blocked10:44
slaineit says scheduled10:46
slainemy worker appears to have disappeared.10:46
Stskeepstry to start it then10:47
*** jonnor_work has joined #mer10:48
slaineNot sure why it wasn't running10:51
vgradelbt, a couple more reports of the loop2 issue on SDK. Reboot and unmount/remount worked for one10:51
*** DocScrutinizer has quit IRC10:52
lbtvgrade: OK, so one is stuck ?10:52
*** DocScrutinizer has joined #mer10:52
vgradeother is on a forum discussion10:53
*** clopez has joined #mer10:53
vgradeI've asked him to stop by here10:54
lbtok - we have RM meeting now for ~1hr10:54
vgradeok, np. I'll look out for him10:55
*** jluisn has joined #mer10:55
rippel_Hi everyone, just a quick question. How is Mer dealing with licenses?11:00
*** Timmy_ has joined #mer11:00
Stskeepsrelease management meeting in #mer-meeting now11:00
Stskeepsrippel_: elaborate on what specifics you're interested in?11:01
*** Timmy has quit IRC11:02
Stskeeps(licenses is a wide term)11:04
*** u1106 has left #mer11:04
rippel_Stskeeps, yes you're right, I'll try to explain myself...11:04
rippel_If I want to use Mer in a commercial product, I must take into account the license of each single package. I guess that you're leaving those licenses unchanged (right?) so I should be able to pick just the LGPL ones (in example)11:05
Stskeepscorrect - Mer is made up of a wide range of licenses, all of them open source though11:06
slainelbt, Good setup for my OBS Workder, How many CPU's and how much RAM11:06
rippel_is there any required/fundamental package that may cause "comemrcial problems", e.g. GPLv311:06
slainethis the VM11:06
rippel_or for which I should release the source code (or patches)11:07
Stskeepsrippel_: we're actively trying to leave out GPLv3 for the moment in images, though you can't avoid them in build phase11:07
Stskeepsrippel_: you'd have to release source code for the parts in mer itself that are governed under GPL11:07
Stskeepsand your modifications to those11:07
slaineI also found another unused server which I'll make act as a worker or perhaps a phost of an obsbe11:07
Stskeepsbut there's nothing saying you should publish your entire OS image with commercial packages on top11:07
Stskeepsframeworks are properly lgpl as well, for example gstreamer11:08
rippel_Stskeeps, ok thank you. So there isn't any "umbrella" license for Mer? Something embracing the whole platform I mean11:08
Stskeepsrippel_: correct, it's almost impossible to do in any linux based stack due to the components in use11:09
rippel_does the same apply to the SDK?11:09
StskeepsSDK is made up of same license types as mer is11:09
rippel_Stskeeps: I mean, Tizen's SDK has some sort of "Samsung" license11:09
Stskeepsie, you can edit it to your hearts extent, it's open source11:10
Stskeepsour SDK is practically a mini Mer :)11:10
rippel_Stskeeps: cool, thank you :)11:10
Stskeepsthe idea is too that you can make your own product SDK11:10
Stskeepsso, a closed source license would be counterproductive11:11
rippel_good, so I should be able to customise the SDK for my products right?11:11
Stskeepsyep11:11
rippel_but does it still have to be based on chroot?11:12
Stskeepsjust as long as you follow the individual licenses, naturally, but that goes for entire stack11:12
Stskeepsit doesn't have to be, but we've selected this method as it's the easiest to deploy11:12
Stskeepsthere was already work on Qt Creator based SDK as well11:12
rippel_oh that would be cool11:12
Stskeepswe just wanted to get a 'platform' sdk working to solve a bunch of typical problems11:13
Stskeepsthat blocked people's ability to start working and contributing to mer, and to use it11:13
Stskeepsso taking that part first :)11:14
*** himamura has quit IRC11:16
alteregoI think chroot SDKs are awesome :p reduces my urges to use VMs for everything :D11:17
alteregoMaybe we should call it PDK though :)11:17
Stskeepspdk makes symbian people shiver11:18
alteregoHeh11:19
*** tilgovi has quit IRC11:19
* w00t has a slight VM insanity at the moment11:21
w00tI have an OS X machine with two windows VMs, and a fedora VM with a debian chroot with scratchbox inside it11:21
w00tI need to set up an ubuntu VM sometime this week to put a Qt network test server on, too11:22
w00tall in all, I am extremely thankful I have a 480gb hard drive nowdays11:22
lbtw00t: I'm tryng to setup a rapidly redeployable VM test network .... they are fun eh :)11:26
Stskeepsrippel_: we also have tools like this: http://wiki.meego.com/File:Kernel5.png11:26
Stskeepsrippel_: so you can review licensing yourself11:26
w00tlbt: apart from forgetting which machine I'm on sometimes... :-)11:26
*** Timmy_ has quit IRC11:27
lbtw00t: I thought it would be nice to name the main Mer admin VM the same as the main MeeGo admin VM ... has occasionally led to WTF moments....11:27
lbtohhhhhh .... I'm on *that* zathras11:28
w00tI sometimes forget which machine has what name altogether11:28
lbtI did consider naming them all zathras in the true spirit of Bab511:28
w00tI've started using less abstract naming just so I can recognise them (i.e. my lenovo is called "dublin" - guess why? :-))11:28
lbtyeah - functionally named vms for me ... boss, be, fe, wiki, bz1, bz211:29
rippel_Stskeeps: that's COOOOL (excuse my shouting, just very surprised)11:29
rippel_what tool is it?11:30
lbthehe ... "ooh,,,, shiney!"11:30
rippel_oh AgileBrowser... I'll have a look :-)11:30
lbtrippel_: I'd prefer something a little more DB oriented11:31
lbtthough AB does have some license tracking utility11:32
lbtbut .... -> do stuff11:32
dm8tbrlbt: for license stuff there is this new format pushed by e.g. yocto. spx? stx? I like the idea11:34
E-Ptimoph, Stskeeps: Can I create "Quality" page to Mer wiki? We can collect todos, ideas, processes etc. to there11:39
StskeepsE-P: we have a 'just do it' attitude, if the area isn't there, go forth and create :)11:40
E-Palright, will do11:40
*** nj3ma has joined #mer11:41
*** simh_ has joined #mer11:41
Stskeepsdm8tbr: that sounds like a good task bug11:42
Stskeepsrippel_: warning, big svg, http://releases.merproject.org/~carsten/mer-core10.svg is a simplified version of Mer generated from AgileBrowser11:43
Stskeeps(missing base essentials in that diagram)11:43
Stskeepsso that's what it can navigate inside, with metadata per package, etc11:44
*** lamikr has joined #mer11:53
Stskeepsdm8tbr: https://events.linuxfoundation.org/events/embedded-linux-conference/flanagan11:53
dm8tbryes, I saw beth's talk in SF :)11:53
dm8tbrIIRC it works on tarball hashes, so pristine tar or whatever should be a good combination?11:54
*** clopez has quit IRC12:01
*** rippel_ is now known as rippel|Away12:01
*** enc0de has joined #mer12:02
*** enc0de has left #mer12:04
*** enc0de has joined #mer12:04
enc0de@vgrade hi, rebooted and re-mounted the mersdk stuff... but it is still the same error12:07
*** tomeff has joined #mer12:07
*** Macer has quit IRC12:07
*** rippel|Away has quit IRC12:10
slainelbt, re the obs worker vm, what kinds stats should I allocate it for RAM and no. CPU's ?12:12
slaineI'm thinking of perhaps moving the OBS fe/be to another box and have the 16GB server act as phost worker12:12
lbtslaine: so we've found the best use of resources is RAM as tmpfs12:13
*** ericlr has quit IRC12:14
slaineSo, lots of RAM and 4 cpu's then ?12:14
*** jluisn has quit IRC12:14
lbtit depends what you need to build12:14
*** jluisn has joined #mer12:14
slainethe job I'm kicking off is a full rebuild of Mer Core12:15
lbtI'd go for 2 workers, each with 4Gb tmpfs, 4Gb RAM and 2 CPUs12:15
slaineSo allocate 8GB to the VM and allocate 4GB to tmpfs12:16
slainewithin the vm12:16
vgradeenc0de, hi. lbt enc0de has the loop2 issue12:16
slainetakes me back, used to build our old red hat stuff on 500MB tmpfs to speed it up12:16
slaineback when 500MB was a lot12:17
lbtslaine: the tmpfs is on the host12:17
slaineI'm confused then12:17
lbtsince the host needs to prepare the fs for the worker12:17
slainewhy would the phost need more tmpfs12:17
lbtit makes a loopback device12:17
lbtnb ... this is mainly taken care of by OBS12:18
slaineI'm not clear on the division of work here12:18
lbtsysconfig/obsworker12:18
slainephost or vmhost os12:18
lbtso tmpfs is a kernel thing12:18
*** rippel has joined #mer12:18
lbtkinda hard to use as / :)12:19
lbtunless initrd did some kind of prep12:19
lbtso we make a tmpfs on the phost12:19
lbtthen make a loopback-based large file12:20
lbtthen we pass that file or loop to the VM to use as a block device12:21
lbt(not sure which)12:21
slaineI seem to recall Domique Le Foll doing something like this12:21
slainefor the MeeGo production servers, to optimize IO12:21
lbtyeah - we all did it and it got scripted into the OBS worker setup12:22
slaineSo, the OBS_VM_USE_TMPFS="" setting reflects this ?12:22
lbtyes12:22
*** arcean_ has joined #mer12:23
*** arcean has quit IRC12:24
*** tomeff1 has joined #mer12:25
*** tomeff has quit IRC12:27
enc0desorry was distrected... for the others: everytime i switch into mersdk and want to build a image i get 'Error <mount>: Error creating ext3 filesystem on disk /dev/loop02'12:28
enc0devgrade told me that i have to reboot and remount the stuff but it is still the same error12:28
*** toscalix has joined #mer12:29
*** norayr has quit IRC12:33
slaineok, importing MDS:Core:i586 into my OBS VM's, lets see what happens12:34
enc0devgrade means that this is a known issue?12:35
lbtenc0de: yes, it is12:35
enc0deand normally it is solved by rebooting12:35
lbtit has been, yes12:36
lbtI'm not really sure what's happening12:36
enc0deok then i will give it a second try12:36
slaineright, off to get some foodage while that's importing12:36
lbtfor sure - we can also do some diagnostics12:36
slaineMDS is getting hammered atm, 4 worker threads working away12:37
lbtgood12:37
*** jukkaeklund has quit IRC12:38
dm8tbrlbt: you saw my discussion with Stskeeps about SPDX?12:40
lbtno12:40
lbtlooking12:41
dm8tbrhttps://events.linuxfoundation.org/events/embedded-linux-conference/flanagan12:41
enc0demh no the same issue... btw i have ubuntu and i forgot to say that there is also python stuff after the error... some libproxy exception12:42
enc0decan i post it here?12:42
lbtuse pastie - yes12:42
enc0deException AttributeError: "'NoneType' object has no attribute 'px_proxy_factory_free'" in <bound method ProxyFactory.__del__ of <libproxy.ProxyFactory object at 0xaa474ec>> ignored12:42
lbtyeah - not an issue... mic is already dead by then I think12:43
*** Sage_ is now known as Sage12:45
enc0deok12:45
*** Sage has quit IRC12:45
*** Sage has joined #mer12:45
Sagejussi: got it back ;)12:45
lbtenc0de: just having lunch12:54
*** araujo has quit IRC12:56
*** Macer has joined #mer12:56
enc0deyep no problem :) enjoy your lunch12:57
*** araujo has joined #mer12:59
*** rektide has quit IRC13:02
*** rektide has joined #mer13:02
rippelHi, I'm building a Mer image from the SDK. Is it possible to specify a different pkg manager than yum?13:09
rippelIn the 'mic create' command I mean13:09
*** slx has joined #mer13:11
enc0derippel: the sdk told me only zypp and yum13:12
*** jluisn has quit IRC13:12
*** jluisn has joined #mer13:13
*** swerden has quit IRC13:15
rippelthank you enc0de13:17
*** drussell has quit IRC13:19
*** drussell has joined #mer13:21
Stskeepsrippel: it's possible to say zypp too but it's a bit unreliable, we use zypp13:22
rippelStskeeps, you mean yum? :-)13:24
Stskeepserr.. yes13:24
rippel:)13:24
slainelbt, minor annoyance, the Status monitor page, the plots/graphs don't have any data.13:25
slaineany idea where to look ?13:25
*** himamura has joined #mer13:26
*** wmarone has joined #mer13:36
*** wmarone_ has quit IRC13:36
*** lamikr has quit IRC13:38
*** gabrbedd has joined #mer13:39
*** lamikr has joined #mer13:41
*** arcean_ has quit IRC13:46
*** clopez has joined #mer13:52
*** slx is now known as swerden13:52
*** dazo is now known as dazo|mtg13:53
swerdenlbt: Did you got my question regarding the community machine?13:55
*** dazo|mtg is now known as dazo13:59
*** clopez has quit IRC14:00
*** pirut has quit IRC14:02
*** clopez has joined #mer14:03
vgradeswerden, hi14:05
vgradeI think David popped out for lunch14:06
slainelol, I was being a spa14:14
*** jluisn has quit IRC14:14
*** jluisn has joined #mer14:14
slainethe status plot/graphs where showing armv7l14:14
slaineonce I pointed them to i586, it all became clear14:14
*** jluisn has quit IRC14:14
*** M4rtinK has joined #mer14:15
slaineStskeeps: I need an ssse3 capable box to build i586, isn't that right ? (trying to recall what we said last week)14:20
*** mlfoster has joined #mer14:22
*** toscalix has quit IRC14:30
*** kurtul has joined #mer14:36
*** rippel has quit IRC14:39
*** tsdedst has joined #mer14:47
*** niqt has quit IRC14:48
*** M4rtinK has quit IRC14:49
harbaumAny news from the archos g9/spark side? Is any of these making progress?14:50
slaineharbaum: the spark is now called vivaldi14:51
slainemight help in searching youtube for recent demos14:51
slainevgrade put one up this morning iirc.14:52
*** steff_ is now known as steff14:52
harbaumPeople should use names like sfgslkaaffdsasadv and not something that gives millions of other hits in search engines14:56
vgradehttp://opentablets.org/14:56
vgradeg9 still waiting on sde but the latest image is flying according to mdfe_14:57
harbaum"vgrade vivaldi" doesn't return anything on youtube14:58
*** KaIRC has quit IRC14:59
*** leinir has quit IRC15:01
*** KaIRC has joined #mer15:01
*** clopez has quit IRC15:02
vgradehttp://www.youtube.com/watch?feature=player_embedded&v=Z1N1_-RHvTY15:06
harbaumvgrade: thanks, found it via opentablets.org15:09
enc0devgrade: someone whispered to me that you implemented the wireless driver in vivaldi successfully15:10
enc0dewhen can I expect with a little how-to ;)15:10
vgradesoon just making sure its working on 2m and 1f models now15:11
enc0deah ok if you want i can help (i have 2m)15:12
*** leinir has joined #mer15:14
*** leinir has quit IRC15:14
*** leinir has joined #mer15:14
*** lamikr has quit IRC15:14
*** clopez has joined #mer15:17
*** harbaum has quit IRC15:29
*** gabrbedd has quit IRC15:36
slaineIn OBS, do I use armv7l or armv7el, for the scheduler and arch names ?15:40
Stskeepsarmv7el15:41
slainethat's what I've used15:41
*** kurtul has quit IRC15:42
*** kavurt has joined #mer15:42
slaineStskeeps: this look right, http://pastebin.com/ghwrFUjw15:43
slainesimilar problems as to earlier, where the worker doesn't seem to kick in15:44
Stskeepslooks fine15:45
slaineand the obs scheduler settings I have are i586 armv7el armv8el15:52
*** gabrbedd has joined #mer15:52
*** rippel_ has joined #mer15:52
slainewhere armv7el pulls from MDS:Core:armv7l and armv8el pulls from MDS:Core:armv7hl15:52
slaineStskeeps: when setting up Arm repo's for a project, do I have to include one for i586 also, to get the build env working ?15:53
*** thopiekar has joined #mer15:55
*** dazo is now known as dazo_afk15:56
vgradeenc0de, thanks for testing16:00
slainelbt, Stskeeps thanks for all your help getting this setup working16:02
*** simh_ has quit IRC16:02
slaineIt's up and running now16:02
slaine2 VM's, 1 with 4CPU's and 8GB RAM, another with 2CPU's and 4 GB RAM.16:03
slaineand I've i586 and armv7el schedulers building away.16:03
*** NIN101 has joined #mer16:05
*** bergie has quit IRC16:08
*** pohly has quit IRC16:08
rippel_Hi, I'm trying to load Mer's rootfs from redboot (via NFS) but I get this error: Failed to open /dev/kmsg for logging: No such file or directory16:08
rippel_after this, it hangs16:08
rippel_unfortunately I don't have the 'CONFIG_DEVTMPFS_MOUNT' option in my kernel's menuconfig16:08
rippel_any idea?16:08
vgraderippel_ I think you only get it once you select CONFIG_DEVTMPFS option16:15
vgradewhat version is it?16:15
*** nj3ma has quit IRC16:15
rippel_it's 2.6.35 but it's been customised for an embedded device (intel atom)16:16
Stskeepswhat vgrade said, it appears after enabling devtmpfs in general16:16
*** tilgovi has joined #mer16:16
Stskeepshttp://cateee.net/lkddb/web-lkddb/DEVTMPFS_MOUNT.html16:16
rippel_I've created (mknod) the /dev/kmsg manually, I'll see if it fixes the problem in a while16:17
rippel_thank you Stskeeps, I can probably try to recompile my kernel, adding that option to .config manually (it's not included in the menuconfig)16:18
vgraderippel_ you may get it once you try and compile16:19
Stskeepsor oldconfig16:19
*** onekenthomas has quit IRC16:20
vgradeStskeeps, wifi running nice on various models of vivaldi now.  Need to add the magic ./test-connman enable wifi to the wiki somewhere16:21
Stskeepscool16:21
vgradetime to have a look at those amlogic gles libs16:22
rippel_ok mknod fixed it, but still the boot process hangs with no message16:22
rippel_:-(16:22
Stskeepsrippel_: how far in it?16:23
StskeepsNFS is a bit of an unknown for us, so it's good you're feeding back16:23
rippel_very early I would say16:23
rippel_but can't get a log... can I?16:23
rippel_except the redboot output16:23
Stskeepsdo you have 'quiet' in your kernel command line?16:24
rippel_no. My command is: console=ttyS0,115200 root=/dev/nfs rw nfsroot=192.168.3.167:/srv/export ip=dhcp mem=exactmap memmap=1024K@0 memmap=199M@1M16:24
rippel_I'll try without the memmap stuff...16:24
Stskeepsok, add systemd.log_level=debug and systemd.log_target=kmsg16:25
rippel_ok16:25
*** smoku has quit IRC16:28
*** norayr has joined #mer16:28
*** lamikr has joined #mer16:29
vgradeis wiki locked?16:32
vgradeok now16:33
*** gimli has joined #mer16:45
*** gimli has joined #mer16:45
*** mdfe_ has quit IRC16:46
vgradelbt, do we have such a thing a a template bugzilla vm16:47
*** clopez has quit IRC16:48
*** gabrbedd has quit IRC16:49
*** slaine has quit IRC16:51
*** clopez has joined #mer16:55
*** wala has joined #mer16:56
*** wool has joined #mer16:57
*** the-gibson has joined #mer17:01
*** gabrbedd has joined #mer17:01
*** nsuffys has joined #mer17:02
rippel_no luck with my kernel issue: I still don't get any output afterNFS-mounting the rootfs17:05
rippel_VFS: Mounted root (nfs filesystem) on device 0:1317:06
rippel_I've also added the option init=/sbin/init17:06
rippel_my original system uses busybox. Is there anything I should to on Mer in order to make it similar?17:07
*** tmzt_ has quit IRC17:11
*** enc0de has quit IRC17:13
*** rippel_ has quit IRC17:15
*** tilgovi has quit IRC17:34
*** ALoGeNo has quit IRC17:45
*** marquiz_ has joined #mer17:48
*** afiestas has quit IRC17:51
*** wool has quit IRC17:52
*** thopiekar has quit IRC17:53
*** trbs has joined #mer17:53
*** NIN102 has joined #mer17:57
*** NIN101 has quit IRC17:58
*** ALoGeNo has joined #mer17:59
*** ALoGeNo has joined #mer17:59
*** wool has joined #mer18:00
* Stskeeps yawns18:02
*** NIN101 has joined #mer18:05
*** clopez has quit IRC18:06
*** NIN102 has quit IRC18:07
*** M4rtinK has joined #mer18:13
*** enc0de has joined #mer18:15
*** furikku has quit IRC18:15
*** gabrbedd has quit IRC18:16
*** afiestas has joined #mer18:22
*** netweaver_ has joined #mer18:23
netweaver_stskeeps: call me stupid but I'm trying to find a kernel config param ("CONFIG_MMC_DEBUG") in config-generic and config-adaptation-n900 but I can't find it in either. Where should it be ?18:26
Stskeepsnetweaver_: google for it, i'm sure you'll find it deps on some other flag18:26
*** NIN101 has quit IRC18:26
netweaver_stskeeps: if I add it simply to the config-adaptation-n900 the build script collapses mentionning this CONFIG_MMC_UNSAFE_RESUME18:29
netweaver_+ exit 1. BUt that key is also not there :( In fact, I cant find ANY mmc key in either of the 2 to-be-merged config files. So where does this result .config come from ? Somewhere it must pick up some defaults for the entries not redefined...18:29
netweaver_it depends on CONFIG_MMC18:30
netweaver_so 'walking up the tree' and manually also defining the parent as well should be enough?18:32
Stskeepsnetweaver_: basically the scripts verify you have all settings18:33
*** gabrbedd has joined #mer18:34
Stskeepsso enabling debug means you also have to choose on config mmc unsafe resume18:38
netweaver_stskeeps: I though I understood the hierarchical reaplacement of flag but I was missing the knowledge on the requirements on even not-specified flags. Thanks :)18:39
Stskeepsour packaging is a bit special18:40
netweaver_stskeeps: ok, I tried again, now specifying CONFIG_MMC=y, CONFIG_MMC_DEBUG=y and CONFIG_MMC_UNSAFE_REMOVAL=y, added to the end of  config-adaptation-n900. Fingers crossed.18:44
netweaver_stskeeps: can we redirect the kernel boot log via usb to a PC for capturing? Or is that not enabled in the Nemo kernel?18:45
Stskeepsnetweaver_: sadly not18:45
netweaver_stskeeps: ok ...mer-next build happily continues but the normal build target say 'building' but the progress doesn't show any logs, also trigger rebuild doesn't work. can it be stuck ?18:50
Stskeepsnetweaver_: obs is a bit odd at times18:50
*** florin_ has joined #mer18:52
florin_hello18:52
netweaver_stskeeps: all things I'm aksing recently I'm getting the 'funny' OBS answer :D I'll do a dummy file change and hope it starts rebuilding on the real target. Or should the mer-next target also be fully usable ?18:52
florin_I've read some logs from this room just now and saw a discussion about some pandaboard gpu drivers18:52
Stskeepslo florin_ :)18:53
florin_anybody have any idea where I could get RN_4_04_00_01 from ? somebody mentioned that particular release here http://mer.bfst.de/logs/%23mer/%23mer.2012-03-06.log.html18:53
Stskeepsnetweaver_: it should be, that one is older than "current" though18:53
florin_and I would really need that if anybody had any idea18:53
Stskeepsflorin_: http://software-dl.ti.com/dsps/dsps_public_sw/sdo_sb/targetcontent/gfxsdk/18:54
florin_thank you very much; I've searched on that site but didn't find it ... must be too tired at this hour18:57
*** vivijim has quit IRC18:57
Stskeepsflorin_: it is a bit hidden admittedly18:57
Stskeepsi couldn't find it through google18:57
florin_thanks again18:58
Stskeepsnp - out of curiousity, what do you hope to use them for?18:58
florin_need opengl working on omap 443018:59
Stskeepsalright18:59
*** agi has quit IRC19:04
*** evoc has joined #mer19:04
Stskeepshello evoc19:05
evochi Stskeeps19:06
*** lamikr has quit IRC19:06
Stskeepswelcome :) so what brings you here to #mer ?19:06
netweaver_stskeeps: does the Nemo kernel has all the KDB patches in? Sorry for being a pain in the backside, learning here :)19:07
Stskeepsnetweaver_: KDB? the patches there are the patches there is :P19:07
*** jstaniek has joined #mer19:08
*** vivijim has joined #mer19:11
*** vivijim has joined #mer19:11
netweaver_stskeeps: I didn't see any KDB patch in the patchlist, but maybe these days it's part of the standard kernel tree. I couldn't find anything for above 2.6.33 version on the SGI site and also nothing for arm. So doesn't it exist for arm and/or for more recent kernels?19:11
Stskeepsnetweaver_: i'll be daft and not know what kdb is..19:13
_av500_a defunct secret service19:14
Stskeepsnetweaver_: we have version 2.6.37 in nemo for n900, check that19:14
netweaver_stskeeps: KernelDeBugger : http://www.ibm.com/developerworks/linux/library/l-kdbug/ and http://kernel.org/doc/htmldocs/kgdb.html19:14
Stskeepsnetweaver_: ok, don't know if we have that19:15
netweaver_stskeeps: it's anyway depending on proper keyboard access during boot, with MAGIC_SYSRQ ... mmmm I need to read/dig a bit deeper on that.19:16
Stskeepsnetweaver_: yea, doubt you'll get that on a n900 :)19:17
*** evoc has quit IRC19:21
w00tyou never know!19:22
*** marquiz_ has quit IRC19:24
*** peuhkura has quit IRC19:26
*** raignarok has joined #mer19:33
*** dionet has joined #mer19:33
*** peuhkura has joined #mer19:33
*** meegoexperts has joined #mer19:46
*** clopez has joined #mer19:51
Sagelbt, Stskeeps: how does the new mer release on friday looks?19:53
*** Timmy has joined #mer19:55
StskeepsSage: well, we were going to push out a prerelease today with new eglibc as that warrants new testing20:01
Stskeepsand postpone final to 2012040520:02
*** himamura has quit IRC20:06
*** jonnor_work has quit IRC20:09
lbthey - sorry - I've been asleep all day - real bad headache20:12
*** netweaver_ has quit IRC20:13
vgradelbt, sounds like you're coming down with something20:14
lbtcould be - not feeling great20:14
*** tsdedst has quit IRC20:15
vgradei'd say get an early night but you've just got up :)20:15
*** bergie has joined #mer20:16
*** trbs has quit IRC20:19
*** s1gk1ll has joined #mer20:19
*** sigkill_ has quit IRC20:23
*** niqt has joined #mer20:31
*** tilgovi has joined #mer20:36
*** cxl000 has quit IRC20:38
*** peuhkura has quit IRC20:39
*** phdeswer has joined #mer20:42
*** niqt has quit IRC20:47
*** peuhkura has joined #mer20:51
*** nsuffys has quit IRC20:57
*** koo4 has quit IRC20:58
*** koo0 has joined #mer20:58
*** lizardo has quit IRC21:00
*** notmart has quit IRC21:01
*** koo0 has quit IRC21:02
*** sirdancealot has joined #mer21:04
*** andre__ has quit IRC21:21
*** gabrbedd has quit IRC21:22
*** dionet has quit IRC21:28
*** gabrbedd has joined #mer21:33
*** Timmy has quit IRC21:44
*** Termana has quit IRC21:50
*** gedw99 has joined #mer21:58
*** gabrbedd has quit IRC22:07
*** gedw99 has quit IRC22:10
*** gimli is now known as gimli_sleep22:13
*** Guest43061 has joined #mer22:29
*** kavurt has quit IRC22:47
*** mlfoster has quit IRC22:53
*** shanem has quit IRC22:54
*** norayr has quit IRC22:55
*** Anssi138 has quit IRC22:57
*** selesnie has quit IRC22:58
*** selesnie has joined #mer22:58
*** ahma has quit IRC22:59
*** ahma has joined #mer22:59
*** raignarok has quit IRC23:03
*** cxl000 has joined #mer23:04
*** jstaniek has quit IRC23:38
*** ahma has quit IRC23:48
*** xucdsp_ has quit IRC23:50
*** stochastic has quit IRC23:55
*** stochastic has joined #mer23:57
*** stochastic has joined #mer23:57

Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!