Monday, 2013-07-01

*** tpb has joined #timvideos00:00
*** CarlFK has joined #timvideos00:37
*** mithro has joined #timvideos02:56
*** CarlFK has quit IRC03:49
*** CarlFK has joined #timvideos04:32
*** CarlFK has quit IRC08:10
*** bananadine has joined #timvideos08:44
*** bananadine has quit IRC09:38
*** mithro has quit IRC09:40
*** bananadine has joined #timvideos09:41
bananadinemithro-work weekly meeting10:06
mithro-workGoogle Hangouts okay?10:06
mithro-workokay, give me 5 minutes to find a room here10:06
bananadinealready there10:06
*** mithro has joined #timvideos10:08
mithrobananadine, your in a hangout?10:09
bananadinei sent you an invite just now10:10
mithroI don't see anything10:10
mithrowhich address did you send it too?10:10
bananadine[email protected]10:11
bananadinemithro, iiie this is what i've found out so far
tpbTitle: This doesn't get called - (at
mithro-workthat links to11:16
tpb<> (at
mithro-work    logCategory = 'httpauth'11:16
mithro-workdo you understand what this means?11:17
mithro-work    self.httpauth.setRequesterId(self.config['avatarId'])11:17
mithro-workexceptions.AttributeError: 'NoneType' object has no attribute 'setRequesterId'11:17
bananadineyea i know where it links to, but the whole module is untouched11:18
bananadinei tried raising exceptions and printing everywhere11:18
bananadineand i guess that means that11:19
bananadinethe object = None11:19
tpb<$3> (at
bananadineand yes that's true11:21
mithro-work"the object = None"11:21
mithro-work     self.httpauth.setRequesterId(self.config['avatarId']) <- on that line which is none?11:21
tpb<$3> (at
mithro-work     self.httpauth.setRequesterId(self.config['avatarId']) <- on that line which is none?11:24
bananadinei misread11:25
bananadinehttpauth is none11:25
mithro-workyes, correct11:26
bananadineis this some kind of an alias ?11:27
tpb<> (at
mithro-workbananadine, no11:28
mithro-workits setting the attribute "logCategory" to the string httpauth11:28
mithro-workno magic there11:28
bananadineinstead of None i gave httpauth "http.HTTPAuthentication"11:31
bananadineand got this11:32
tpbTitle: [email protected]:~/Downloads/flumotion$ trial flumotion.test.test_co - (at
mithro-workyou need to find out how httpauth is created on that instace11:33
mithro-workand make sure it is created11:33
bananadineit has to be none?11:34
mithro-workyour not listening11:37
mithro-workyou need to find out how httpauth is created on that instance11:37
mithro-workand make sure it is created11:37
bananadinewell there's not much11:39
bananadinethere's a initialization self.httpauth = None in streamer.py11:39
bananadineand there's     logCategory = 'httpauth' in http.py11:39
bananadinethe two are related11:40
mithro-worklogCategory = 'httpauth' is totally unrelated11:41
mithro-workyou are looking for things which set httpauth11:45
bananadinethen there's  no way that httpauth in has the attributes in http.HTTPAuthentication11:45
mithro-workthat sentence makes no sense11:54
bananadinewhat i meant is11:55
bananadinei don't understand how  self.httpauth.setRequesterId(self.config['avatarId']) is supposed to work11:56
bananadinewhen "self.httpauth = None"11:57
mithro-workit doesn't11:57
mithro-workso self.httpauth must be set to something other then none somewhere11:58
bananadinei added "self.httpauth = http.HTTPAuthentication(self)" in
tpb<$L> (at
bananadineand the test passed12:06
mithro-workbananadine, why did you do that?12:06
bananadinei don't know it just came to me12:07
mithro-workthat is a bad reason to do that12:09
bananadinehttpauth needed to be able to call the methods from http.HTTPAuthentication12:11
bananadineso first i wrote12:11
bananadine"self.httpauth = http.HTTPAuthentication"12:11
bananadinebut then i got12:11
bananadine"exceptions.TypeError: unbound method setRequesterId() must be called with HTTPAuthentication instance as first argument (got str instance instead)12:11
bananadineso i changed it to "self.httpauth = http.HTTPAuthentication(self)"12:12
mithro-workwhere else is http.HTTPAuthentication created?12:12
mithro-workI doubt the (self) argument is right12:13
bananadinei don't see it being created anywhere else12:14
bananadinedo you think the test result could be a false positive?12:15
mithro-workwhy would http.HTTPAuthentication exist if it wasn't used somewhere?12:16
bananadineok sorry12:17
bananadineit's actually used12:18
bananadinein other modules12:18
bananadineeverywhere it's used in this context "self.httpauth = http.HTTPAuthentication(self)"12:18
bananadinewait i'll link12:19
tpb<$_> (at
mithro-workso why are you doing self.httpauth yourself when there are a bunch of functions which do it for you?12:26
bananadinethat method which does that12:28
bananadinedoesn't get called12:28
bananadinei check with raising an exception there12:28
bananadineso now i called the method in the test12:30
bananadineand it works12:30
*** hyades has joined #timvideos12:31
bananadinei think it's fine now12:32
bananadinewell since i'm done with that12:39
bananadinei'm going to make a break now and later check out planet.xml files12:39
bananadinei saw an example .xml file on the documentation12:39
bananadinewhich is what i need for the basic pipeline12:40
hyadesmithro-work, i am working mostly offline till next two days12:41
mithro-workwe need to organise a weekly face to face meeting12:55
mithro-workwhen is good foryou?12:55
hyadesafter 3rd12:55
mithro-workI mean time wise13:06
hyadesaround this time?13:08
mithro-workcan you do any earlier, it's a bit late for me13:14
*** hyades has quit IRC13:34
*** CarlFK has joined #timvideos14:12
bananadineso i want to create a stream now without using the wizard and with this configuration
tpb<> (at
bananadinehow do i do that? :)14:14
mithro-workbananadine, this is where you should use your head :)14:43
CarlFKmithro-work: can you merge or what ever the hdmi2usb code into your git repo?14:44
*** CarlFK has quit IRC14:46
*** CarlFK has joined #timvideos14:50
CarlFKsorry if this is a double - wifi dropped, not sure if this made it:14:57
CarlFKmithro-work: can you merge or what ever the hdmi2usb code into your git repo?14:58
CarlFKin 10 hours I'll show my progress to Ste the local guy who owns his own Atlyus and has been looking for something to do with his14:58
mithro-workdid you get the atlys working with useful firmware?15:05
CarlFKI got that broken test pattern15:08
CarlFKdid you ever see the screen shot I posted?15:08
mithro-workCarlFK, you need to be more descriptive, what did you have plugged in, what firmware where you using, all that...15:09
CarlFKsec.. I'll put my script on the wiki15:12
mithro-workconfig.txt is part of the Raspberry Pi not the HDMI2USB15:23
CarlFKah.. thanks15:25
mithro-workthe raspberry pi doesn't do "hot plug" displays15:26
CarlFKI can power cycle the atlys, load the firmware and without unplugging the usb cable run camorama and get the test pattern15:29
tpb<> (at
mithro-workafter the fxload step you should pull the USB connector and plug it back in15:35
mithro-workthe tool you want to test with is15:35
mithro-workor mplayer15:36
mithro-workor vlc15:36
mithro-workor atleast something non-random15:36
*** CarlFK has quit IRC15:37
bananadinemithro-work, that .xml config in the documentation is probably way outdated15:39
bananadinei managed to get a flow to the muxer but it alternates from active to inactive15:40
bananadinep.s: i'm using the flumotion i got installed15:41
bananadinenot the one i'm porting15:41
bananadinei could paste you the output from the terminal and the config file i'm using15:42
bananadineif you like15:42
*** CarlFK has joined #timvideos15:43
CarlFKswitched to a usb wifi dongle... which seems to work, unlike my built in wifi which seems to have gone from bad to worse in the last 24 hours15:44
mithro-work<mithro-work> after the fxload step you should pull the USB connector and plug it back in15:45
mithro-work<mithro-work> the tool you want to test with is15:45
mithro-work<mithro-work> luvcview15:45
mithro-work<mithro-work> or mplayer15:45
mithro-work<mithro-work> or vlc15:45
mithro-work<mithro-work> or atleast something non-random15:45
CarlFKluvcview gave me a similar image, only the grey area is orange15:47
CarlFKare you sure about pulling the USB?15:47
mithro-workwhat are the LEDs on the board displaying?15:47
iiiebananadine: hey, sure is it giving errors in the log when it goes from active to inactive?15:48
CarlFK7 6 5 4 x 3 x x   ( number = on, x = off)15:48
bananadineiiie, i don't think i was saving a log15:49
iiiebananadine: ls /usr/local/var/log15:50
mithro-workSo Led 7 and Led 6 both being on means you are currently getting the test image15:50
iiieI'm actually unclear how to make it not log15:50
iiie(or log less)15:50
mithro-workLED7, LED615:50
mithro-workSelected Input15:50
mithro-workOff, Off15:50
mithro-workHDMI 015:50
mithro-workOff, On15:50
mithro-workHDMI 115:50
mithro-workOn, Off15:50
mithro-workReserved (for VGA extension board)15:50
mithro-workOn, On15:51
mithro-workTest pattern15:51
bananadineiiie, are yousure that's the right dir?15:51
bananadinei don't have var in local15:51
iiienot 100%, one second and I'll verify15:51
CarlFKmithro-work: should we move this to #hdmi2usb ?15:51
iiiebananadine: see if there is a flumotion directory in there15:53
iiiehm, well somehow mine is in /usr/local15:54
iiiewhat logs are in the flumotion directory?  (I have service.log)15:54
bananadinei have service.log too15:54
bananadinewell actually15:54
bananadinemanager.default.log  service.log.1  worker.default.log  worker.default.log.215:54
bananadineservice.log     service.log.2  worker.default.log.115:54
iiieAll the numbered ones are from logrotate, depending on the dates they may have useful information15:55
iiiewhen the muxer (or stream?) is alternating active / inactive, have a terminal up running "tail -f FILENAME"15:57
iiieyou'll probably have to use sudo15:57
iiie("sudo tail -f /var/log/flumotion/worker.default.log"15:58
bananadineyeah i figured15:58
bananadinejust a minute15:58
bananadinei have some trouble with my text editor15:58
bananadinecan't exit it15:58
iiiewhat editor?15:59
iiie(pico/nano, emacs, vi/vim)?15:59
bananadinecan't even kill it16:01
bananadinenever mind i'll open the log with another editor while you're here to help me16:01
bananadineand then i'll restart16:01
iiietail -f has the advantage of not crashing no matter the size16:02
iiielog files are particularly painful on that16:02
bananadinei opened the log files16:02
bananadinelooked at the dates16:02
bananadinethere is nothing past April 2116:03
iiierun "ls -la /var/log/flumotion" are any of those passed 4/21?16:03
CarlFKI am guessing a new version is logging somewhere else16:04
bananadinenope, none are past 4/2116:05
bananadinei remember i changed the version of flumotion then16:05
bananadinenow to find where it's saving16:05
bananadinei got recent logs in tmp16:06
bananadinebut that's not what we're looking for, right?16:06
iiiewe're looking for logs to watch while you're seeing the change in the muxer, presumably "something is wrong"16:07
CarlFK (11:07:11 AM) skay: live streaming europython
tpbTitle: EuroPython 2013: Florence, July 17 (at
CarlFKbut everyone is asleep now16:07
iiiebananadine: did you configure it to put logs in /tmp ?16:08
iiietail the logs and see what you get16:10
iiiea text editor is a poor choice for watching logs (even if it refreshes it has to read the whole file, while all you care about is the new end)16:10
bananadineshould i turn on the manager and worker?16:11
bananadineand then look at the log16:11
iiieyes, the only options are troubleshoot (investigate) while it's running, or record with great (excessive) accuracy when what happened and check in the logs.16:12
iiieand start tail -f first, so you can see what's happening16:13
bananadinecan i configure planet.xml to tell the worker where to put the logs?16:24
iiieI don't think so, I think it's in the service where the logs are set.  There might be an option to change it in /etc/default/flumotion, but I wouldn't know what it was.16:26
iiieI've got to run, back on hm, I'll be on at work, but my attention will mostly be lacking16:29
bananadineno problem16:31
*** bananadine has quit IRC16:31
*** iiie is now known as iiie216:32
*** iiie has joined #timvideos19:06
*** parx has left #timvideos19:26
*** CarlFK has quit IRC23:16
*** Jahanzeb has joined #timvideos23:19
*** Jahanzeb has joined #timvideos23:37

Generated by 2.5 by Marius Gedminas - find it at!