Back to Home / #mythtv / 2007 / 11 / Prev Day | Next Day
#mythtv IRC Logs for 2007-11-22

---Logopened Thu Nov 22 00:00:27 2007
00:09|-|Shmattie [] has quit ["Leaving"]
00:25|-|purserj [] has joined #mythtv
00:39|-|JoeXBorn changed nick to JoeBorn
00:43|-|czth [i=dbrobins@nat/microsoft/x-b114c727221c6f3c] has joined #mythtv
01:00|-|Ava- [i=aval0n@] has quit [Remote closed the connection]
01:24|-|harryk [] has joined #mythtv
01:28|-|Chutt [] has quit [Remote closed the connection]
01:46|-|harryk [] has left #mythtv ["Leaving"]
01:49|-|BleedAway [] has quit [Read error: 110 (Connection timed out)]
02:16<xris>hmm, backend doesn't seem to be properly cleaning up after itself...
02:17<xris>ran out of disk space with lots of "deleted" recordings around to remove
02:18<xris>extra odd because that storage group should have spilled over to another partition, too
02:33<xris>ah, looks like the storage group wasn't set up properly.. but deleted recgroup should have kicked in
02:46|-|cmoates [] has quit [Read error: 113 (No route to host)]
03:01|-|_gnome42 [] has quit ["Leaving"]
03:25|-|Cougar [] has quit []
03:25|-|Cougar [] has joined #mythtv
03:35|-|justinh [] has joined #mythtv
03:35|-|xris [] has quit ["Leaving."]
03:39|-|mzb_d800 [] has quit [Read error: 110 (Connection timed out)]
03:48|-|clever_ [] has joined #mythtv
03:53|-|czth_ [i=dbrobins@nat/microsoft/x-5229917a60c3091b] has joined #mythtv
04:00|-|o_cee [] has joined #mythtv
04:03|-|clever [] has quit [Read error: 110 (Connection timed out)]
04:03|-|JoeBorn [] has quit [Read error: 110 (Connection timed out)]
04:08|-|czth [i=dbrobins@nat/microsoft/x-b114c727221c6f3c] has quit [Read error: 110 (Connection timed out)]
04:19|-|onixian [] has joined #mythtv
04:20|-|beavis [] has joined #mythtv
04:45|-|lucas changed nick to lucas2
04:46|-|lucas2 changed nick to lucas
04:46|-|lucas changed nick to ljmeijer
06:08|-|justinh [] has left #mythtv []
06:18|-|czth__ [i=dbrobins@nat/microsoft/x-02cd2f055f1541f6] has joined #mythtv
06:34|-|czth_ [i=dbrobins@nat/microsoft/x-5229917a60c3091b] has quit [Read error: 110 (Connection timed out)]
06:53|-|justinh [] has joined #mythtv
07:01|-|czth [i=dbrobins@nat/microsoft/x-d522ff73484ff687] has joined #mythtv
07:14|-|czth__ [i=dbrobins@nat/microsoft/x-02cd2f055f1541f6] has quit [Read error: 110 (Connection timed out)]
07:32<gbee>have to hand it to MS, they make it impossible to take the MCE remote apart without damaging it
07:32<gbee>plastic is _very_ soft
07:42<gbee>having got the Windows logo removed, I'm now wondering how I replace it with something more mythtv-esque, no colour printer here (least not one with a colour cartridge)
07:46<janneg>argh, creating new capture cards segfaults
07:53<janneg>it might be onlty a merge error in myultirec
07:54|-|tzanger [] has left #mythtv []
08:01|-|janneg [] has quit [Remote closed the connection]
08:38|-|askehil1 [n=askehill@] has joined #mythtv
08:38|-|askehil1 [n=askehill@] has left #mythtv []
08:41<gbee>anyone know why we take so long to change inputs? switching between a DVB-T card and a PVR-150 takes over 10 seconds here during which the screen goes black
08:41|-|BleedAway [] has joined #mythtv
08:42<justinh>using the card on demand eh?
08:53<gbee>I was, but changing it doesn't make a difference
08:53<justinh>think that setting needs a backend restart to change
08:53<gbee>takes forever to initialise the PVR when switching from DVB
08:53<gbee>justinh: I did
08:54<gbee>watching the IR blaster, it doesn't even send the signal until the first 5 or more seconds of the transition
08:54<gbee>and it takes a further 5 seconds for any video
09:00<gbee>never seen people complain of a transition time like that, it's a slow machine but I don't see why that would have such a great effect
09:18|-|Chutt [] has joined #mythtv
09:39<gbee>Would someone mind reviewing this patch? My solution for getting around the chicken and egg problem of the font scaling isn't as pretty as I'd have liked --
09:43<stuarta>that actually looks not too bad. bit of a reformatting change crept in but apart from tha
09:46<gbee>main change was moving the initialisation of most things from the constructor to the new Init(), but that allows us to keep the theme info related stuff in OSD where it belongs and still scale the fonts etc using the correct aspect ratios
09:48<gbee>the osdtypes change snuck into the patch, doesn't actually form part of the themeinfo changes
09:48<stuarta>does it work okay?
09:48[~]stuarta needs a cup of tea
09:49<gbee>works great, which I've got to say suprised me, I'd expected that there would be some bugs
10:26|-|PointyPumper [i=Pintlezz@] has quit [Read error: 110 (Connection timed out)]
10:26|-|mzb [] has quit [Read error: 113 (No route to host)]
10:42|-|Dabian [n=M0RTEN@fsf/member/dabian] has joined #mythtv
10:43|-|splat1 [] has joined #mythtv
10:44<Dabian>What is kvadivo trying to tell me? He was autoignored by my client, and I haven't figured out how to unignore him just yet! (Sbeen a while since I IRC'ed for real)
10:44|-|foxbuntu [] has quit [Read error: 110 (Connection timed out)]
10:46<Dabian>OK, anyhow .. I have a question for you, here it goes:
10:46<Dabian>OK .. I am trying to set up videosources (mythbuntu 7.10 or similar), but the grapper I need is configured wrong .. would you possibly hint me where to fix it, either in mythbuntu or default install?
10:47<gbee>Dabian: kvandivo was trying to tell you that this isn't the correct channel for support requests
10:48<Dabian>gbee: OK .. this question is related to development though, isn't it?
10:48<Dabian>A bug?
10:49<gbee>doesn't sound like a mythtv bug from your description
10:50<Dabian>gbee : I am not a native english speaker ... what doesn't make it sound like dev?
10:50<gbee>a mythbuntu one maybe, possibly an XMLTV one or just user error
10:50<Dabian>gbee: Lemme be more specific then
10:52<Dabian>gbee: The grapper used to fetch channels for Denmark .. calls tv_grap_dk or similar, and it gives sensible arguments for it. However, this script asks the user questions about which channels to include, username and so. This causes mythtv-setup to hang at 50% and never proceed.
10:53<Dabian>gbee : Well, I guess the grabber *is* tv_grap_dk ... but it needs a frontend to enter username, and per default accept all channels, I guess.
10:54<Dabian>(If you run it without parameters, it wont require username or stuff .. only when called with the config option like mythtv-setup uses)
10:58<Dabian>gbee : Honestly I dont know if its fixed in the latest version, I only have mythbuntu .. but I doubt that the issue is limited to mythbuntu.
10:59<gbee>Dabian: it's not a bug, you need to proceed with configuration by switching to the console that you stated mythtv-setup from
10:59<gbee>it will be improved in 0.21, but we don't consider it a mythtv bug
11:00<gbee>stated = started
11:00<Dabian>gbee : For some reason the console doesn't show these questions.
11:01<Dabian>gbee: So as it stands, I need a fix so I can run the program manually, and tell myth where the xml file is .. or similar.
11:05<Dabian>gbee : Well, actually the problem is that I cannot switch to the terminal, as mythtv-setup grabs my keyboard.
11:06<gbee>Dabian: that's a user issue, no-one else seems to have the same problem - ask in #mythtv-users
11:08<Dabian>gbee: Possibly, or a bug in mythbuntu I guess. I talked to another user with a similar problem, but they didn't need to switch to the terminal.
11:23|-|capt-rogers [] has joined #mythtv
11:36|-|gregor__ [] has joined #mythtv
11:37|-|gregor__ [] has quit [Client Quit]
11:37|-|gregor__ [] has joined #mythtv
11:37|-|gregor__ [] has quit [Client Quit]
11:38|-|gregor__ [] has joined #mythtv
11:38|-|gregor__ [] has quit [Client Quit]
11:38|-|gregor__ [] has joined #mythtv
12:09|-|Dabian [n=M0RTEN@fsf/member/dabian] has quit ["nvm then"]
12:09|-|PointyPumper [i=Pintlezz@] has joined #mythtv
12:12|-|JoeBorn [] has joined #mythtv
12:18|-|melunko [n=hmelo@] has joined #mythtv
12:29|-|jimmygoon [] has joined #mythtv
12:29<jimmygoon>I'm building/buying a frontend pc... What stats would I need on a VIA EPIA MiniITX motherboard to properly use it as a Media Director?
12:31<jimmygoon>woops. I always miss the "dev channel" bits, my apologizes
12:39|-|hasse_ [] has joined #mythtv
12:40<hasse_>hey installed the mythbuntu pack, and everything was fine, until i added the weekly builds, now i miss the libmythtv-0.20.2 the package installed i 0.20.0 how to change that ?
12:40[~]justinh points at the channel topic
12:41|-|Cardoe [n=Cardoe@gentoo/developer/Cardoe] has joined #mythtv
12:57|-|gnome42 [] has joined #mythtv
12:59|-|hasse_ [] has quit ["leaving"]
13:07|-|JoeBorn [] has quit [Read error: 110 (Connection timed out)]
13:09|-|JoeBorn [] has joined #mythtv
13:41|-|justinh [] has left #mythtv []
13:44|-|cmoates [] has joined #mythtv
13:49|-|stiev3 [] has joined #mythtv
13:49|-|stiev3 [] has left #mythtv ["Leaving"]
13:53|-|splat1 [] has quit [Remote closed the connection]
13:53|-|splat1 [] has joined #mythtv
13:57|-|PointyPumper [i=Pintlezz@] has quit [Read error: 110 (Connection timed out)]
14:05|-|omry_ changed nick to omry
14:19|-|mattwire [] has joined #mythtv
14:24|-|jimmygoon [] has quit [Read error: 110 (Connection timed out)]
14:28|-|mattwire [] has quit ["Leaving"]
14:37|-|janneg [] has joined #mythtv
14:38<gbee>janneg: as I think you already know, there is an issue with auto-expiry since your change
14:40<janneg>I suspected it
14:41<janneg>my server was unfortunately offline until now, heven't read any mails the whole afternoon
14:48|-|onixian [] has quit ["This computer has gone to sleep"]
14:53|-|onixian [] has joined #mythtv
14:53|-|onixian [] has quit [Client Quit]
14:54|-|cmoates [] has quit ["Leaving"]
15:16|-|janneg_ [] has joined #mythtv
15:17<janneg>gbee: was xris online today?
15:17<gbee>janneg: no
15:19<janneg>it was reproduceable. I recorded this afternoon on 9 channels backt to back
15:20<gbee>I saw the problem on one of my backends
15:21<janneg>the error seems to be that the autoexpirer couldn't dispatch the delete messages after some time
15:21|-|JoeBorn [] has quit [Read error: 110 (Connection timed out)]
15:21<gbee>it ran out of space as I was testing OSD changes by watching livetv
15:22<janneg>gbee: did you saw "SendResponse: Unable to write to client socket, as it's no longer there" messages in the backend output
15:23<GreyFoxx>SendResponse is part of the upnp stuff I think
15:24<gbee>janneg: let me check, I didn't stop to look at the logs, I just manually deleted a few things and restarted it
15:25<gbee>2007-11-22 20:13:08.528 AutoExpire: Found max recording rate of 0 MB/min2007-11-22 20:13:08.529 AutoExpire: CalcParams(): Required Free Space: 0.0 GB w/freq: 15 min
15:25<janneg>GreyFoxx: it is the MythEvent dispatching
15:25<gbee>2007-11-22 15:12:17.421 AutoExpire: ERROR: Filesystem Info cache is empty, unable to calculate necessary parameters.2007-11-22 15:12:17.440 AutoExpire: ERROR: Filesystem Info cache is empty, unable to determine what Recordings to expire
15:26<janneg>gbee: that is ok, if no recording is runnen (the first line)
15:26<gbee>janneg: yeah, I meant to past the second two lines
15:29<janneg>gbee: that is a different error and might be bogus
15:29<gbee>janneg: I've looked back to just before that that error appears and it looks like a sockets issue
15:30<janneg>ah that makes sense
15:32<janneg>I suspect that the changes cause somewhere a deadlock
15:37|-|melunko [n=hmelo@] has quit [Remote closed the connection]
15:37|-|PointyPumper [i=Pintlezz@] has joined #mythtv
15:59<gbee>do we composite the OSD, then scale the video to the screen res, or scale the video, then composite the OSD? Appears to be the latter but I don't pretend to understand all the video rendering stuff so I'm not sure why it's done that way.
15:59|-|xris [] has joined #mythtv
16:01<gbee>xris: I've not updated to the latest trunk version of mythweb, so forgive me if I'm pointing out something which was fixed but shouldn't there be some text next to the top checkbox?
16:04<xris>gbee: I don't see that extra checkbox on my system
16:05<hads>Hmm, neither.
16:07|-|gregor__ [] has quit ["Leaving"]
16:08<xris>anyone know who here is paul g? I got a linkedin request from him, regarding this channel, but don't know many people's real names.
16:10<janneg>xris: do you see "SendResponse: Unable to write to client socket, as it's no longer there" or other socket related messages in the backend log?
16:11<xris>janneg: what I pasted was the only thing there. it had been going on too long before I noticed, and was FULL of that stuff
16:13<xris>looks like the file version of the log didn't get updated at all during that period. weird.
16:29|-|cmoates [] has joined #mythtv
16:33|-|janneg_ [] has quit [""tschuess""]
16:37|-|capt-rogers [] has quit ["Ex-Chat"]
16:44|-|eYeLez [] has joined #mythtv
16:57|-|gbee [] has left #mythtv ["Gone"]
16:57|-|gbee [] has joined #mythtv
17:00<gbee>before I commit it, let me just ask one more time if anyone has a problem with this patch?
17:01<gbee>ignoring the osdtypes.cpp changes which have nothing to do with it
17:03|-|gregor [] has joined #mythtv
17:06|-|mzb_d800 [] has joined #mythtv
17:13<hads>gbee: I just applied it here. Can't see any issues. Anything specific to test?
17:14<gbee>hads: just that there are no problems with OSD graphics and font scaling
17:14<hads>OSD could be a little more jittery although could be my imagination.
17:15<hads>Actually that only seems to be on my PVR150 input, on DVB it's fine.
17:15<gbee>I've tested it pretty thoroughly, it works well, my concerns are mainly about the code rather than whether it works
17:16|-|andreasm [] has joined #mythtv
17:16<gbee>doesn't mean that additional testing is a bad thing though
17:16<hads>OK, I'm not going to be help there then, I'll shutup :)
17:28<gbee>hads: nah, I appreciate your input, whatever it is - it's not possible to over test something, there are hundreds of possible combinations of video, theme and screen aspect ratios ands resolutions, it's not impossible that I've simply not tried the combination that will break things
17:28<laga>it's svn trunk
17:28<laga>people will tell you if it breaks :)
17:30<hads>gbee: Cool. Well nothing breaks here anyway.
17:30<gbee>laga: sure and as I said, that's not my major concern, I just wasn't completely happy with the solution I chose for a particular problem and I just thought I'd check whether anyone else had an issue with it
17:32<gbee>it _can_ be changed at a later date, but I'd rather get it right first time and then forget about it
17:33<gbee>on the other hand, I don't want to wait forever, I want the patch out of my tree so I can move on
17:51<gbee>I've only got a couple of HD test vids and my screen doesn't support resolutions above 1280x800, but here is a screenshot anyway:
17:52<gbee>if you compare it to the SD screenshots, the difference is pretty obvious:
17:53<hads>Indeed, very nice.
18:01<janneg>argh, it happened again and I forgot to start the backend under gdb
18:01<janneg>damn, I could have attached gdb to the pid
18:02<clever_>did what again?
18:05<janneg>the the autoexpirer problem I'm debugging atm
18:05<clever_>ive found a similar one but havent confirmed it
18:06<clever_>auto expire moves things to the deleted recgroup
18:06<clever_>then it takes another 10mins to actualy expire it
18:06<gbee>well I was going to show a screenshot taken at HD res but with a standard theme, only mythfrontend is now refusing to play the same video again
18:07<gbee>2007-11-22 23:06:38.676 [h264 @ 0x2b2137a3d970]no frame!
18:08<janneg>gbee: have you restarted the frontend?
18:08<gbee>janneg: running mythtv directly because I haven't imported that video into the recording list
18:09<gbee>I wouldn't worry about it, I can't play back h264 smoothly on this machine anyway, but it was working well enough before that I could grab a screenshot
18:10<janneg>strange, libavcodec has known problems with some samples but since it played once it should play always
18:20|-|splat1 [] has quit []
18:24<janneg>hah. found something. the socket error happens when there is a reschedule while expiring programs
18:25<janneg>clever_: I would guess the autoexpirer needs to call delete recording with a parameter really_delete
18:25|-|beavis [] has quit ["Verlassend"]
18:25<clever_>yeah that would do
18:26<clever_>auto expire from the deleted group realy does delete i beleive
18:29<janneg>or the autoexpirer has to change the recgroup to deleted group before deleting
18:29<janneg>that's probably easier to fix
18:29|-|mzb [] has joined #mythtv
18:38|-|gregor [] has quit ["Leaving"]
18:48|-|CrevilO [] has joined #mythtv
18:49|-|CrevilO [] has left #mythtv ["going to support"]
19:01<clever_>janneg: yeah wont cause a change to the function definition
19:05<janneg>clever_: was there a ticket for that problem? I'm about to commit a fix
19:08<clever_>i didnt make one for my problem
19:08<clever_>didnt even test to see if it realy happened
19:09<clever_>its tricky to test auto expire since you need to eat up enough space to triger it
19:10<janneg>it's not a problem with multirec, I have atm 9-12 recordings running on my dev system
19:11<clever_>didnt think multirec supported undelete
19:11<janneg>clever_: did you sent a mail to the dev list? I remember reading about that problem
19:11<clever_>nope just the irc channel
19:12<janneg>multirec is just some commits behind trunk
19:12<clever_>im on 14799 right now
19:12<janneg>ok, then it was probably here
19:12<clever_>14800 doubles my recording bitrate
19:12<clever_>and some time latter everything started to seg fault on startuo
19:13<janneg>last multirec merge is iirc 14925
19:16<janneg>clever_: please describe your recording profile
19:17<clever_>frame grabber
19:17<clever_>software encoding to mpeg4
19:17<clever_>some1 said before that the problem is only on software encoders
19:18<clever_>my bitrate was set to 2200 and when i calc it out i got 4400
19:18<clever_>and the files also doubled in size(1ghg/hour to 2gig/hour)
19:19<janneg>ffmpeg merge can only affect software recorders in regard to the recording bitrate
19:19<clever_>hardware encoders dont record thru ffmpeg
19:19<clever_>but whenyou transcode it may become a problem
19:20<janneg>I want more details: resolution and all settings on the mpeg4 codec options page
19:20<clever_>i think it was 400x400 res
19:20<clever_>im not at home right now but have access to mysql
19:23<janneg>yes, transcoding should be affected too
19:23|-|mzb_d800 [] has quit [Remote closed the connection]
19:24|-|mzb_d800 [] has joined #mythtv
19:24<clever_>i know for shure that 14800 itself is affected
19:24<clever_>might have allready been fixed by now but i havent recently upgraded
19:25<janneg>unlikely that it has been fixed
19:29<janneg>clever_: SELECT * FROM recordingprofiles as r, profilegroups as p WHERE r.profilegroup = AND cardtype = "V4L" AND videocodec = "MPEG-4";
19:30<clever_>2 rows in set (0.01 sec)
19:30<janneg>clever_: then "SELECT * FROM codecparams WHERE profile = X;" where X is the first id in the output of the first query
19:31<janneg>please paste that to a pastebin
19:32<clever_>i cut my bitrate down to 1000(its actualy obeying it atm) so i can get more space out when not home
19:37<janneg>clever_: have you tried what happens if you change the bitrate with 14800? is the ratio constant or ...
19:38<clever_>its a perfect 200% i think
19:38<clever_>when i changed it to 1100 i got 2200 recordings
19:38<janneg>ok, that sounds like a bug
19:41<GreyFoxx>I've run into a new one since the last ffmpeg sync with .mpg playback. I've got 12 files that played perfectly (I reverted to test it) with 14799 of multirec, but after 14800 I get a stuff and the output keeps talking about the video being more than 30 frames behind the audio and it's dropping frames to keep up
19:41<GreyFoxx>And of course they are my daughters favourite movies :)
19:44<clever_>ive got a few shows which cause mythtranscode to segfault
19:44<clever_>but they play fine
19:49<janneg>so many problems so little time :(
19:51<GreyFoxx>I've also noticed some of my HDTV recordings are stuttering during playback, but not all . So I'm installing 14799 again to see if it was the ffmpeg sync or something else
19:52<clever_>i found the chunk of myth code that read the bitrate from the db and set one of the structs
19:52<clever_>i could just /2 the bitrate there instead of fixing it elsewhere
19:52<janneg>that's an ugly workaround
19:53<clever_>yeah but i lost track of where that variable is used so it would be more work for me to fix it properly
19:53<janneg>but I could have told you that it is in libs/libmythtv/NuppelVideoRecorder.cpp
19:53<clever_>thats about where i found it i beleive
19:53<clever_>that just inits a struct/class from the coded stuff in the db
19:53<clever_>(the function i read over)
19:55<janneg>the scheduler also runs amok once the SendResponse error happened. it reschedules once again if it's finished
19:56<clever_>and my frontend segfaults after editing recording profiles(it was fixed but i downgraded)
20:06|-|mattwire [] has joined #mythtv
20:08|-|mattwire [] has quit [Client Quit]
20:54<janneg>xris: [14935] might have fixed the auto expire issue, at least it isn't as fast reproduceable as before
21:04<janneg>good night
22:13|-|czth_ [i=dbrobins@nat/microsoft/x-e9c19a052120821b] has joined #mythtv
22:31|-|czth [i=dbrobins@nat/microsoft/x-d522ff73484ff687] has quit [Read error: 110 (Connection timed out)]
22:54|-|purserj [] has quit [Read error: 104 (Connection reset by peer)]
22:56|-|PointyPumper [i=Pintlezz@] has quit [Read error: 110 (Connection timed out)]
23:27|-|wladston [n=wladston@] has joined #mythtv
---Logclosed Fri Nov 23 00:00:04 2007