MythLogBot@irc.freenode.net :: #mythtv

Daily chat history

Current users (98):

abqjp, aloril, andreax, Anssi, anykey_, Beirdo, ben1066, brfransen, brtb, CaCtus491, Captain_Murdoch, cattelan_away, cesman, Chutt, clever, coling, Computer_Czar, Cougar, damaltor, dblain, dekarl1, dlblog, ElmerFudd, Ffseto, FinnTux, foxbuntu, ghoti, gigem, gregL, gregorcy, GreyFoxx, Guest81426, ikke-t, J-e-f-f-A, j-rod|afk, JackWinter, jafa, jams, jarle, jcarlos, jheizer, joe___, joki, jpabq, jstenback, jwhite, jya, k-man, kenni, knightr_, kurre2, kwmonroe, mag0o, markcerv, mike|2, mrand, MythBuild, MythLogBot, mzanetti, NightMonkey, obo, okolsi, peitolm, pepsiman, petefunk, poptix, purserj, rhpot1991, rsiebert_, sannes, Seeker`, skd5aner, Slasher`, SmallR2002, sphery, sraue, stichnot, stuarta, sunkan, superm1, sutula, taylorr, tgm4883, TheAsp, ThisNewGuy, toeb, tomimo, tris, Unhelpful, vallor, wagnerrp, wahrhaft, wseltzer, XDS2010_, xris, yb0t, zombor, _charly_
Tuesday, April 17th, 2012, 00:00 UTC
[00:00:10] ben1066_ (ben1066_!~quassel@host86-160-28-95.range86-160.btcentralplus.com) has joined #mythtv
[00:01:35] ben1066 (ben1066!~quassel@host109-152-54-11.range109-152.btcentralplus.com) has quit (Ping timeout: 260 seconds)
[00:07:58] stichnot: jya: now when I enter live TV with timestretch=1.2, I get no audio and timestretch=1 until I manually change timestretch, then audio comes back.
[00:08:44] jya: better than a crash :)
[00:09:18] jya: but what you describe make sense
[00:09:40] jya: you enter liveTV, you're already at the end of the recording, so you can't use timestretch at 1.2, it reverts to 1.0
[00:09:55] jya: isn't it the way it should be? did you expect anything else?
[00:10:09] jya: oh sorry
[00:10:16] jya: I just re-read.. you get no audio...
[00:10:22] jya: I have audio ...
[00:16:41] brockp (brockp!~brockp@108-65-202-7.lightspeed.livnmi.sbcglobal.net) has joined #mythtv
[00:17:31] zombor (zombor!~zombor_@kohana/developer/zombor) has quit (Remote host closed the connection)
[00:22:26] toeb (toeb!~tob@HSI-KBW-078-042-104-026.hsi3.kabel-badenwuerttemberg.de) has quit (Ping timeout: 272 seconds)
[00:23:53] jya: stichnot: just tried again… timestretch revert to 1.0 and audio start properly
[00:25:33] jya: if you could provide me with a full log of running mythavtest with -v audio and no other argument (this will start liveTV automatically)
[00:26:13] brockp (brockp!~brockp@108-65-202-7.lightspeed.livnmi.sbcglobal.net) has quit (Quit: brockp)
[00:26:32] stichnot: jya: I'll have to get that later tonight
[00:26:33] jya: stichnot: BTW, you were right, most of the time when the backend stopped responding seem to coincide with when liveTV failed with a error too many buffer…
[00:28:38] jk90090 (jk90090!~jk90090@76.167.224.151) has joined #mythtv
[00:28:52] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has quit (Quit: kormoc)
[00:29:11] toeb (toeb!~tob@HSI-KBW-078-042-104-026.hsi3.kabel-badenwuerttemberg.de) has joined #mythtv
[00:56:31] toeb (toeb!~tob@HSI-KBW-078-042-104-026.hsi3.kabel-badenwuerttemberg.de) has quit (Ping timeout: 264 seconds)
[00:57:16] stichnot: jya: http://code.mythtv.org/trac/attachment/ticket . . . 755.3560.log
[00:57:32] jya: it's tonight already?
[00:58:19] stichnot: close enough... I snuck in the run
[00:58:41] stichnot: now I'd better get default timestretch back to 1.0 before someone notices...
[00:58:43] jya: interesting log...
[00:59:27] jya: AC3 mono set as passthrough
[00:59:51] jya: the core issue is that the audio isn't configured properly to start with...
[01:00:07] jya: is livetv playback different to normal playback on how it start and is initialised?
[01:00:46] stichnot: the log is probably even more interesting since mythavtest was running at the same time as mythfrontend (main menu)
[01:01:07] stichnot: it was probably a news show, so I wouldn't be too surprised at AC3 mono
[01:01:09] jya: disabling passthrough when timestretch is activated is something done in avformatdecoder.cpp
[01:03:07] jya: the function AvFormatDecoder::SetDisablePassthrough is called from mythplayer
[01:03:08] toeb (toeb!~tob@HSI-KBW-078-042-104-026.hsi3.kabel-badenwuerttemberg.de) has joined #mythtv
[01:03:11] jya: I see the problem there
[01:03:50] jya: SetDisablePassthrough is called from mythplayer if the playback was normal (timestretch == 1.00) and changed to something else
[01:04:37] jya: but here timestretch was never 1.00 to start with… so passthrough isn't disabled
[01:05:45] jya: that's not something I can fix in the audio layer, it's in the playback code
[01:05:59] jya: it's wrong logic to start with
[01:07:47] jya: https://github.com/MythTV/mythtv/blame/cacada . . . thplayer.cpp
[01:08:38] jya: I need to find a LiveTV channel with AC3 to test… there are known at this time of the day here
[01:14:46] jya: stichnot: can you try this patch?
[01:14:47] jya: http://pastebin.com/itHVg93T
[01:21:54] jya: stichnot: cool, there's a channel with AC3 right now
[01:21:58] jya: can test myself
[01:23:46] jya: oh well… doesn't work :)
[01:27:02] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has joined #mythtv
[01:29:08] jk90090 (jk90090!~jk90090@76.167.224.151) has quit (Quit: jk90090)
[01:46:35] jya: hum… the issue is much more complex that I anticipated
[01:54:00] NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has quit (Read error: Connection reset by peer)
[01:58:42] anykey_ (anykey_!~anykey@46-126-245-147.dynamic.hispeed.ch) has quit (Ping timeout: 272 seconds)
[01:58:53] anykey_ (anykey_!~anykey@46-126-245-147.dynamic.hispeed.ch) has joined #mythtv
[01:59:41] stichnot: have you figured out why live TV has anything to do with it?
[02:00:29] jya: stichnot: I'm not even looking at that.. I'm only looking at mythplayer
[02:01:01] jya: and see how I can add a quick work around so passthrough is disabled if playback starts with a timestretch != 1.0
[02:02:38] jya: it's so convoluted that code…. I don't understand why everything is so complicated to start with
[02:02:44] stichnot: if it's this messed up, I'm fine disabling startup timestretch in live TV. though I'm still curious why it works in recorded playback but not live tv
[02:02:49] stichnot: me either
[02:04:11] jya: first, I'm trying to understand why there is a normal_speed and a play_speed
[02:07:09] jya: normal_speed is set to false when in pause mode. but there's a distinction done along the code if normal_speed = false, then no A/V sync is made, or just decode video and not audio.. yet set audio time stretch etc..
[02:09:13] Beirdo: blech
[02:16:49] jya: is there some kind of timeout when it comes to the new DBLogging in place?
[02:17:02] jya: when I run myth in gdb and do step by step
[02:17:44] jya: whenever I step over a LOG() I get an error about mysql and "Lost Connection to MySQL server during query"
[02:17:56] jya: it works find if I just continue in my breakpoint
[02:18:20] jya: now is MySQL server has gone away
[02:20:36] Beirdo: the connections always have timeouts
[02:20:47] Beirdo: don't think that has changed
[02:21:00] jya: I do my step over very quickly ...
[02:21:01] Beirdo: maybe we are just using it more?
[02:21:13] jya: like less than a second after a breakpoint
[02:21:24] Beirdo: step debugging does what for other threads though?
[02:21:52] Beirdo: if you are at a (gdb) prompt, no threads are running
[02:22:04] jya: I am at gdb prompt
[02:22:20] Beirdo: but anyways, might make it easier if you do a --nodblog while step debugging :)
[02:22:27] jya: no idea what it does for other threads… just trying to debug the damn thing to understand what that blob does
[02:23:22] Beirdo: OK, it's debugging time here too
[02:23:34] stichnot: I thought the debugger just let all the threads loose (except ones that you explicitly suspend in the debugger) and set a temporary breakpoint at the next instruction.
[02:23:43] Beirdo: I need to clear off the couch, this dead recliner's getting too uncomfortable
[02:23:44] jya: i don't know how to fix it.. Either do a work around for the specific case where it will fail ; or do the proper thing and make sure the situation incorrectly handled never occur to start with.. I'm afraid the 2nd type will just open a massive can of worms
[02:23:54] Beirdo: hehe
[02:24:07] Beirdo: well, we are at the right time in a dev cycle for fresh cans of worms
[02:24:50] jya: yeah, but that's a bug in 0.25 too.. I know myself, if I fix the bug in a dirty way, I will forget about it and not do it proper later
[02:25:56] jya: what is that "dummy" business used for ?
[02:26:02] jya: in mythplayer ?
[02:26:16] stichnot: you mean in live tv?
[02:26:20] jya: yes
[02:26:22] Beirdo: it's used during livetv while the tuner is tuning
[02:26:33] jya: only in livetv ?
[02:26:35] Beirdo: so the ringbuffer's reading from a dummy
[02:26:38] Beirdo: I think so
[02:26:54] Beirdo: not 100% sure, but that was my recollection
[02:27:00] jya: ah so , if I just make a special case for isDummy, really I'm fixing livetv startup ?
[02:27:14] jya: which make my fix not so dirty after all :)
[02:27:18] Beirdo: I think so
[02:27:22] stichnot: yeah, that would be nice
[02:27:39] Beirdo: danielk may remember in greater detail, but he's off at a tradeshow IIRC
[02:27:50] jya: I know… he's not even logged on IRC
[02:28:43] stichnot: speaking of live tv, why on earth is the live tv chain stuff stored in the DB?
[02:28:57] Beirdo: stored?!
[02:29:10] Beirdo: I don't recall seeing THAT
[02:29:18] jya: LOL… I love our in-code doc
[02:29:20] jya: http://code.mythtv.org/doxygen/mythplayer_8cpp.html
[02:29:24] Beirdo: I'd be wondering the same thing
[02:29:24] stichnot: yeah, let me find a reference
[02:29:52] Beirdo: jya: some is better than others ;)
[02:30:04] Beirdo: that one's... a project for a volunteer
[02:30:10] jya: so we have 5 functions appearing, for a 5110 lines source
[02:30:12] stichnot: e.g. LiveTVChain::AppendNewProgram()
[02:31:00] Beirdo: umm, the class is elsewhere
[02:31:06] stichnot: jya: well, now, I'm not so intimidated by mythplayer.cpp any more
[02:31:19] Beirdo: http://code.mythtv.org/doxygen/classMythPlayer.html
[02:31:36] Beirdo: the cpp file is only the non-class members that are left
[02:31:43] Beirdo: that's how doxygen works :)
[02:32:06] Beirdo: although the class ain't too documented either
[02:33:04] Beirdo: stichnot: no idea why that couldn't be in memory, offhand
[02:33:10] stichnot: Beirdo: just search for the uses of MSqlQuery in livetvchain.cpp. I guess this must predate the use of the event mechanism
[02:33:16] Beirdo: I'm sure there was SOME reason for it :)
[02:35:25] jya: Beirdo: ok… it's still a sad state of affair, there's barely any comment in there
[02:35:34] Beirdo: yep ;)
[02:36:09] Beirdo: we all need to take to the time to add documentation in our copious spare time :)
[02:36:37] jya: a few line of comments at the time your write the code to explain some fundamentals doesn't add much time
[02:36:44] jya: probably save heaps in the long term
[02:36:51] Beirdo: sure would
[02:38:55] Beirdo: but God help the person that suggests it... people will say it's too much like work :)
[02:40:53] jya: I find it quite frustrating that all the nasty broken stuff I've worked on in the past few weeks turned out to be always by the same people, who happens to not be part of the project anymore...
[02:41:18] Beirdo: heh.
[02:41:38] Beirdo: it's not always a case of "blame the guy who left"
[02:41:51] Beirdo: sometimes it IS their fault and all
[02:42:15] Beirdo: but ah well, we all have code that needs work, I'm sure
[02:44:53] Beirdo: ah come on valgrind
[02:50:11] jya: in gdb, how do you set a breakpoint on a specific line ?
[02:50:27] Beirdo: IIRC, br filename:linenum
[02:50:38] jya: thanks
[02:50:47] Beirdo: no prob, hope it works :)
[02:50:51] jya: haven't used gdb command line in years…
[02:51:15] jya: I tried kdbg before, absolute piece of crap
[02:51:51] stichnot: gdb-mode in emacs might be slightly more attractive than the command line
[02:52:29] Beirdo: ewww, you said the E-word
[02:53:02] jya: I remember using emacs with gdb at some point in my life.. but I forgot the magic word to start it properly
[02:53:26] Beirdo: now if you want a true religious war... emacs vs vi is the one.
[02:54:41] jya: wagnerrp: is there a way to access smolt output somewhere?
[02:57:07] ** jya found smolt **
[02:57:32] ** Beirdo upgrades his valgrind to that from precise **
[02:57:50] Beirdo: I think I hand-upgraded glibc earlier
[02:58:49] jya: interesting that the majority of myth users are in Europe/Berlin timezone
[03:00:21] Beirdo: how do you figure?
[03:00:39] brfransen (brfransen!~brfransen@64.179.142.146) has quit (Remote host closed the connection)
[03:00:54] stichnot: M-x gdb
[03:00:55] brfransen (brfransen!~brfransen@64.179.142.146) has joined #mythtv
[03:01:43] jya: Beirdo: when you run mythconvert, would audio.HasAudioOut() as checked in mythplayer be set?
[03:03:14] Beirdo: ummm
[03:03:26] Beirdo: you mean nuvexport?
[03:04:08] jya: What I'm really after is finding out under what circumstances would HasAudioOut() return false
[03:04:08] Beirdo: BTW, 15.8% of reports is not a majority... just the largest proportion :)
[03:04:47] Beirdo: hmmm, I dunno. one moment
[03:07:12] stichnot: Beirdo: the word you're looking for is plurality, I think
[03:10:15] Beirdo: jya: seems to be set when it's reinitializing audio in audioplayer
[03:10:36] Beirdo: ahhh, updated valgrind FTW
[03:11:28] Beirdo: soooo much slower when running valgrind :)
[03:12:30] jya: yeah, I get that… I wrote that part… it was more to do with knowing if there are cases where audio playback doesn't occur , like when used for something else that user playback (the visual type)
[03:12:39] jya: like commflag, transcode whatever
[03:12:57] Beirdo: maybe in commflag
[03:13:08] Beirdo: transcode uses audio, generally
[03:13:33] Beirdo: OK, what is taking so long here... finish the program already!
[03:15:07] wagnerrp: your personal output, you can call the script and get it to dump to terminal rather than upload it
[03:15:33] wagnerrp: besides that, theres only the data specifically available through the profile view
[03:15:45] Beirdo: ==00:00:04:40.623 31283== 3 errors in context 71 of 2082:
[03:15:48] Beirdo: snicker
[03:15:55] Beirdo: OK, maybe I should let it finish
[03:19:22] Quantum (Quantum!~quantum@c-24-1-48-163.hsd1.il.comcast.net) has joined #mythtv
[03:20:17] jya: ahhhhhhhh
[03:20:26] jya: I understand what's going on
[03:20:44] jya: SetDisablePassthrough is properly called
[03:21:27] jya: however , at the time it's disabled, it's done on the dummy videooutput (decoderbase.h) which does absolutely nothing for SetDisablePassthrough
[03:21:28] Quantum (Quantum!~quantum@c-24-1-48-163.hsd1.il.comcast.net) has quit (Client Quit)
[03:21:52] jya: decoderbase.h line 125
[03:22:54] jya: then the information that passthrough is disabled is never used again, and when the actual decoder AVFD is created, passthrough isn't disable anymore
[03:23:32] ** jya just thinking alound **
[03:30:17] stichnot (stichnot!chatzilla@mythtv/developer/stichnot) has quit (Ping timeout: 253 seconds)
[03:59:49] Computer_Czar (Computer_Czar!~Unknown@69.4.155.83) has joined #mythtv
[04:01:33] scott_z (scott_z!~scott@75-150-200-78-Illinois.hfc.comcastbusiness.net) has joined #mythtv
[04:02:10] scott_z: is there a way to watch netflix with mythtv on ubuntu?
[04:02:45] wagnerrp: this is the development channel, you want #mythtv-users
[04:02:47] scott_z: sorry wrong channel
[04:07:31] dblain: wagnerrp: regarding MythXMLTest... it still works for upnp services, but the MythXML tab only has a few working methods. I created it to help test and show off the functions... I guess with WSDL support/pages there is not too much of a need for it.
[04:08:15] dblain: if you feel strongly about removing it, I don't think too many people would notice it missing.
[04:08:41] ybot_ (ybot_!quassel@nat/nokia/x-gsmzoqwqqtqnukfd) has quit (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.)
[04:08:56] yb0t (yb0t!quassel@nat/nokia/x-sbbyjcvqliiqyjcq) has joined #mythtv
[04:10:35] Beirdo: oh look at that.
[04:10:38] Beirdo: heh
[04:11:01] wagnerrp: dblain: nah, was just going through contrib to produce a list of executable scripts for a thread on the -dev list
[04:11:07] wagnerrp: and noticed it was still in there
[04:11:20] Beirdo: I think I just found that double delete.
[04:12:16] Beirdo: valgrind proves its worth yet again
[04:19:22] Beirdo: now, how to clean it nicely...
[04:24:08] Seeker` (Seeker`!~cjo20@unaffiliated/seeker) has quit (Ping timeout: 240 seconds)
[04:25:55] scott_z (scott_z!~scott@75-150-200-78-Illinois.hfc.comcastbusiness.net) has quit (Quit: Leaving)
[04:25:56] Seeker` (Seeker`!~cjo20@host86-147-226-117.range86-147.btcentralplus.com) has joined #mythtv
[04:25:57] Seeker` (Seeker`!~cjo20@host86-147-226-117.range86-147.btcentralplus.com) has quit (Changing host)
[04:25:57] Seeker` (Seeker`!~cjo20@unaffiliated/seeker) has joined #mythtv
[06:09:38] wagnerrp: basically, within tens of seconds of the last entry on the mailing list
[06:10:49] wagnerrp: mysql wants you to upgrade its table formats
[06:12:17] wagnerrp: we appear to be on a blocklist for yahoo mail
[06:14:00] wagnerrp: Beirdo: check out that maillog
[06:14:06] wagnerrp: it was actually cut off mid-line
[06:14:18] wagnerrp: almost as if postfix caused the failure
[06:14:29] Beirdo: we are back
[06:15:08] Beirdo: heh
[06:15:24] Beirdo: not sure how it could
[06:16:00] wagnerrp: or it could just be coincidence, as that thing is probably servicing a dozen HTTP requests every second too
[06:16:51] wagnerrp: oh, and log rotation on the apache access logs... were running ~2.6GB currently
[06:16:56] andreax1 (andreax1!~andreaz@p5089EBF7.dip.t-dialin.net) has joined #mythtv
[06:18:05] Beirdo: yech
[06:18:29] Beirdo: OK, ran the mysql_update
[06:18:35] Beirdo: it should stop whining
[06:19:40] andreax (andreax!~andreaz@p5089EBF7.dip.t-dialin.net) has quit (Ping timeout: 265 seconds)
[06:36:54] dekarl: hmm, smolt gives "Service Temporarily Unavailable". is that expected?
[06:45:48] skd5aner (skd5aner!~skd5aner@cpe-071-071-242-134.carolina.res.rr.com) has quit (Ping timeout: 276 seconds)
[06:46:42] skd5aner (skd5aner!~skd5aner@cpe-071-071-242-134.carolina.res.rr.com) has joined #mythtv
[06:59:41] partoj (partoj!~par@94.127.35.115.c.fiberdirekt.net) has joined #mythtv
[07:02:49] Beirdo: it likely needs to be restarted
[07:02:58] Beirdo: the server had a power cycle
[07:03:05] Beirdo: jams: ^^ :)
[07:09:04] hashbang (hashbang!~alex@213-152-35-50.dsl.eclipse.net.uk) has quit (Quit: Leaving.)
[07:11:48] sannes (sannes!~ace@cm-84.209.72.206.getinternet.no) has quit (Ping timeout: 245 seconds)
[08:00:17] floh` (floh`!1148@ente.hawo.stw.uni-erlangen.de) has quit (Remote host closed the connection)
[08:20:54] sraue (sraue!~stephan@xbmc/staff/sraue) has quit (Read error: No route to host)
[08:22:20] sraue (sraue!~stephan@xbmc/staff/sraue) has joined #mythtv
[08:46:18] Beirdo: no wonder it was so quiet.
[08:46:25] Beirdo: mailman didn't get restarted :)
[08:46:33] Beirdo: ahhh, sweet silence
[08:47:43] highzeth (highzeth!~hz@hoiseth.no) has quit (Quit: Leaving)
[09:29:07] dekarl1 (dekarl1!~dekarl@p4FCEF719.dip.t-dialin.net) has joined #mythtv
[09:31:15] dekarl (dekarl!~dekarl@p4FCEEC63.dip.t-dialin.net) has quit (Ping timeout: 265 seconds)
[09:33:52] seeker (seeker!~seeker@unaffiliated/seeker) has joined #mythtv
[09:35:45] gregL (gregL!~greg@cpe-74-76-125-87.nycap.res.rr.com) has quit (Read error: Connection reset by peer)
[09:36:01] jya: http://smolt.mythtv.org/static/stats/stats.html is stil down
[09:43:48] seeker_ (seeker_!~seeker@unaffiliated/seeker) has joined #mythtv
[09:47:12] seeker (seeker!~seeker@unaffiliated/seeker) has quit (Ping timeout: 245 seconds)
[09:47:12] seeker_ is now known as seeker
[09:52:20] gregL (gregL!~greg@cpe-74-76-125-87.nycap.res.rr.com) has joined #mythtv
[10:01:21] seeker: jya: Btw, that email wasn't me complaini
[10:01:46] jya: seeker: no idea what you're referring to
[10:02:06] seeker: G, If there were a better method of advert detection in the UK I'd probably have a shot at implementing it myself
[10:02:42] jya: oh, I knew that… I wasn't going to bother replying to the original morons to start with
[10:02:52] seeker: Heh
[10:03:27] seeker: I could probably provide samples of adverts / program starts and ends if you were interested though
[10:03:42] jya: give this to Beirdo, he's the one with a hand on the matter
[10:03:53] jya: he asked for samples a while back… don't know if he got any
[10:04:00] jya: I know I never sent him some either
[10:05:00] Guest56046 (Guest56046!~mike@c-98-232-220-158.hsd1.or.comcast.net) has quit (Remote host closed the connection)
[10:05:31] seeker: I could well have missed it; my involvement in mythtv is somewhat sporadic
[10:05:52] mike|2 (mike|2!~mike@c-98-232-220-158.hsd1.or.comcast.net) has joined #mythtv
[10:14:44] ikonia (ikonia!~irc@unaffiliated/ikonia) has joined #mythtv
[10:37:44] ikonia (ikonia!~irc@unaffiliated/ikonia) has left #mythtv ()
[11:00:31] damaltor (damaltor!sbnc@h1889977.stratoserver.net) has quit (Ping timeout: 245 seconds)
[11:05:07] damaltor (damaltor!sbnc@h1889977.stratoserver.net) has joined #mythtv
[11:06:42] Guest81426 (Guest81426!~ricky@121-74-86-253.telstraclear.net) has quit (Remote host closed the connection)
[11:07:29] Guest81426 (Guest81426!~ricky@121-74-86-253.telstraclear.net) has joined #mythtv
[11:15:03] jya: seeker: are you Christopher Kerr ?
[11:16:45] zombor (zombor!~zombor_@kohana/developer/zombor) has joined #mythtv
[11:18:13] seeker: jya: No, Chris Oattes
[11:19:32] jya: ah ok.. Funny, two Chris, two Seeker
[11:19:48] seeker: Yeah :) I think that's why he pointed it out
[11:20:23] seeker: And not at all confusing :P
[11:29:06] zombor (zombor!~zombor_@kohana/developer/zombor) has quit (Remote host closed the connection)
[11:31:08] seeker (seeker!~seeker@unaffiliated/seeker) has quit (Remote host closed the connection)
[11:31:52] seeker (seeker!~seeker@unaffiliated/seeker) has joined #mythtv
[11:47:41] Computer_Czar (Computer_Czar!~Unknown@69.4.155.83) has quit (Read error: Connection reset by peer)
[11:48:00] Computer_Czar (Computer_Czar!~Unknown@69.4.155.83) has joined #mythtv
[12:08:16] j-rod|afk is now known as j-rod
[12:29:23] rsiebert_ (rsiebert_!~quassel@g225060030.adsl.alicedsl.de) has joined #mythtv
[12:32:51] rsiebert (rsiebert!~quassel@g225052173.adsl.alicedsl.de) has quit (Ping timeout: 276 seconds)
[12:50:42] seeker (seeker!~seeker@unaffiliated/seeker) has quit (Quit: Colloquy for iPhone - http://colloquy.mobi)
[12:51:57] Jordack (Jordack!~jordack@h69-131-44-221.plmomi.dedicated.static.tds.net) has joined #mythtv
[12:54:45] seeker (seeker!~seeker@unaffiliated/seeker) has joined #mythtv
[13:41:51] stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has quit (Remote host closed the connection)
[13:47:50] stichnot (stichnot!~chatzilla@192.55.54.40) has joined #mythtv
[13:48:01] stichnot (stichnot!~chatzilla@192.55.54.40) has quit (Changing host)
[13:48:01] stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has joined #mythtv
[14:08:23] cattelan_away (cattelan_away!~cattelan@c-66-41-26-220.hsd1.mn.comcast.net) has quit (Ping timeout: 244 seconds)
[14:30:55] cattelan_away (cattelan_away!~cattelan@c-66-41-26-220.hsd1.mn.comcast.net) has joined #mythtv
[14:35:12] gregL (gregL!~greg@cpe-74-76-125-87.nycap.res.rr.com) has quit (Quit: Leaving)
[14:45:37] jams: Beirdo- smoon is running again. For future reference /etc/init.d/smoon can be used to start/stop it. It didn't start on boot because of an old pid file.
[14:56:04] ben1066 (ben1066!~quassel@host86-160-26-249.range86-160.btcentralplus.com) has joined #mythtv
[14:58:26] ben1066_ (ben1066_!~quassel@host86-160-28-95.range86-160.btcentralplus.com) has quit (Ping timeout: 276 seconds)
[15:04:14] stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has quit (Ping timeout: 264 seconds)
[15:08:44] seeker (seeker!~seeker@unaffiliated/seeker) has quit (Read error: Connection reset by peer)
[15:09:38] ben1066 (ben1066!~quassel@host86-160-26-249.range86-160.btcentralplus.com) has quit (Changing host)
[15:09:38] ben1066 (ben1066!~quassel@unaffiliated/ben1066) has joined #mythtv
[15:25:35] partoj (partoj!~par@94.127.35.115.c.fiberdirekt.net) has quit (Quit: leaving)
[15:27:51] jya: how do I change my password in trac?
[15:29:00] skd5aner: !seen stichnot
[15:29:00] MythLogBot: stichnot was last seen 24 minutes 46 seconds ago
[15:30:52] stuarta: jya: i think it's an htpasswd change, so poke Beirdo when he wakes up
[15:31:13] jya: stuarta: thanks.. will do
[15:49:49] joki (joki!~joki@p54865478.dip.t-dialin.net) has quit (Ping timeout: 264 seconds)
[15:50:46] joki (joki!~joki@p54865376.dip.t-dialin.net) has joined #mythtv
[16:05:00] stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has joined #mythtv
[16:15:05] NightMonkey (NightMonkey!~NightrMon@pdpc/supporter/professional/nightmonkey) has joined #mythtv
[16:20:32] sannes (sannes!~ace@cm-84.209.72.206.getinternet.no) has joined #mythtv
[16:26:02] andreax1 (andreax1!~andreaz@p5089EBF7.dip.t-dialin.net) has quit (Ping timeout: 240 seconds)
[16:26:16] andreax (andreax!~andreaz@p5089ECAD.dip.t-dialin.net) has joined #mythtv
[16:30:19] Goga777 (Goga777!~Goga777@2.95.94.114) has joined #mythtv
[16:38:23] OPPluto (OPPluto!~jedi@cpe-76-185-75-39.tx.res.rr.com) has joined #mythtv
[16:38:37] pclark (pclark!~pclark@193.35.11.116) has joined #mythtv
[16:48:36] pclark (pclark!~pclark@193.35.11.116) has quit (Remote host closed the connection)
[16:53:26] skd5aner: stichnot: ping
[16:53:36] stichnot: hey
[16:54:36] skd5aner: stichnot: hey, was just perusing the -users list and stumbled across the report of weird duration times for recordings since upgrade... I'm seeing the same issue since the upgrade
[16:55:10] stichnot: what kind of recordings?
[16:55:11] skd5aner: stichnot: I noticed last night that several new recordings were displaying as significantly longer during playback
[16:55:21] skd5aner: stichnot: well, let me go see... brb
[16:55:45] stichnot: ok. also, are you on fixes/0.25, or Master?
[16:57:03] stichnot: I downloaded the 50MB sample from the reporter, but I won't have a chance to look at it for the next half day
[17:02:52] skd5aner: stichnot: 0.25 fixes, but haven't updated since the release was cut...
[17:03:51] skd5aner: stichnot: but I was able to narrow it down – any recording that was made PRIOR to upgrade, on my analog tuners (PVR-250s/500) – they all show widly exagerated duration times
[17:04:28] skd5aner: nothing consistent (like 4x, etc) – all seem to be almost... 4 hours, 8 hours, 10 hours... etc
[17:04:45] stichnot: interesting. That sounds different from the report on the users list
[17:04:56] skd5aner: all my recordings from these sources *since* the upgrade appear to have the correct duration reported
[17:06:02] skd5aner: probably why I didn't really notice it right away, since 90% of my content is HD (QAM and HDPVR) and I hadn't had a chance to watch anything more than 2 weeks old that also happened to be in SD recently
[17:06:25] stichnot: does the indication of the current position in the recording appear OK?
[17:06:44] skd5aner: stichnot: also, not that it probably makes much of a difference since it's analog capture, but I did check multiple recordings from different networks
[17:08:21] stichnot: This could have something to do with taylorr's changes to use the position and duration information from the mpeg file.
[17:08:22] skd5aner: stichnot: yes, the current position appears to act normal during playback and skipping, and will just end when it gets to the accurately reflected end-time. It's as if the progress bar is just lengthened to the wrong length of time
[17:08:46] kth (kth!~kth@unaffiliated/kth) has joined #mythtv
[17:09:17] skd5aner: want me to submit a bug report? I didn't know if it was the same as what the guy one the list was reporting or not
[17:09:24] skd5aner: s/one/on
[17:09:25] stichnot: Can you try running "mythcommflag --rebuild" on one of these recordings?
[17:09:34] skd5aner: yea
[17:10:34] stichnot: Based on some personal email from taylorr, I bet that will correct the problem you're seeing.
[17:11:14] skd5aner: stichnot: hmmm... well, the problem is I probably have a ton of recordings I'd have to fix manually that way :/
[17:11:16] stichnot: Not that that's what users should be subjected to...
[17:11:19] hashbang (hashbang!~alex@213-152-35-50.dsl.eclipse.net.uk) has joined #mythtv
[17:11:25] stichnot: yeah, what you said :)
[17:11:45] skd5aner: let me see if that at least fixes it or not
[17:12:02] stichnot: if that's the case, it still deserves a bug report
[17:12:14] skd5aner: btw – what's the easiest way to fnd the filename for that particular recording?
[17:12:33] skd5aner: I know I can dig it up – but always wondered if there's an "easy" way
[17:12:36] stichnot: press INFO twice in the Watch Recordings screen
[17:13:02] skd5aner: yea... that would probably be easiest, but the frontend is in the other room – any cli way to get it?
[17:13:08] skd5aner: (or mythweb?)
[17:13:33] stichnot: sure – what information do you have on hand?
[17:13:59] skd5aner: title, subtitle, programid, airdate, time ,etc
[17:15:09] stichnot: select basename from recorded where chanid=xxx and starttime=yyy;
[17:15:40] stichnot: then you'll have to hunt for that filename in all your storage group directories
[17:15:58] skd5aner: yup, thanks
[17:18:07] skd5aner: stichnot: uh... crashed out at ~7%
[17:18:08] skd5aner: 2012-04–17 13:17:49.081679 E decoding error
[17:18:08] skd5aner: eno: Broken pipe (32)
[17:18:31] stichnot: hmm
[17:18:54] skd5aner: restarted it, going much faster now... already to 50%??\
[17:18:54] stuartm (stuartm!~stuartm@mythtv/developer/stuartm) has quit (Remote host closed the connection)
[17:19:19] stichnot: just using --rebuild should go quite fast
[17:19:27] skd5aner: got to 100%, then gave me that error again
[17:19:34] stichnot: broken pipe?
[17:19:38] skd5aner: yea
[17:20:55] skd5aner: is there a verbose option I could throw in to see if it gives any more details on that error?
[17:21:01] stichnot: you usually see one of those errors at the very end due to EOF, but I didn't think it was broken pipe
[17:21:43] jcdutton (jcdutton!~jcdutton@5ac34c4d.bb.sky.com) has joined #mythtv
[17:22:03] skd5aner: actually, it got to 100%, then gave me one of those errors – not sure what happenedthe first time
[17:22:21] skd5aner: anyway, not sure if the error matters or not, I went and the duration is correct now after rebuilding the seektable
[17:22:41] jcdutton: Has anyone seen mythtv crashing on start up due to memory corruption. Somewhere inside QT
[17:23:08] skd5aner: stichnot: so, that's at least the manual fix – I'll go ahead and submit a bug report so you can take a look at it when you get the chance – feel free to reach out to me on here if you need any more details for troubleshooting... thanks!
[17:23:23] OPPluto (OPPluto!~jedi@cpe-76-185-75-39.tx.res.rr.com) has quit (Quit: Lost terminal)
[17:23:40] stichnot: so it works? cool, thanks.
[17:24:03] stichnot: but as you say, that needs to be fixed
[17:29:39] skd5aner: stichnot: #10623 – thanks in advance :)
[17:29:39] ** MythLogBot http://code.mythtv.org/trac/ticket/10623 **
[17:30:25] skd5aner: it does appear to be different than 10622, although in the same realm I guess
[17:46:02] natanojl (natanojl!~jonatan@c83-252-237-63.bredband.comhem.se) has joined #mythtv
[17:46:28] sraue (sraue!~stephan@xbmc/staff/sraue) has quit (Ping timeout: 250 seconds)
[17:48:02] sraue (sraue!~stephan@xbmc/staff/sraue) has joined #mythtv
[18:14:34] stichnot: skd5aner: are all problematic recordings being reported as the same (8+ hour) length?
[18:17:38] kth (kth!~kth@unaffiliated/kth) has quit (Read error: Connection reset by peer)
[18:45:14] Goga777 (Goga777!~Goga777@2.95.94.114) has quit (Remote host closed the connection)
[18:58:43] Beirdo: Mark Lord is Mr. Know-It-All as usual
[19:26:21] stuartm (stuartm!~stuartm@cpc1-derb9-0-0-cust441.8-3.cable.virginmedia.com) has joined #mythtv
[19:26:21] stuartm (stuartm!~stuartm@cpc1-derb9-0-0-cust441.8-3.cable.virginmedia.com) has quit (Changing host)
[19:26:21] stuartm (stuartm!~stuartm@mythtv/developer/stuartm) has joined #mythtv
[19:27:24] gregL (gregL!~greg@cpe-74-76-125-87.nycap.res.rr.com) has joined #mythtv
[19:27:34] stuartm: I've just received a DVD which causes the frontend to leak huge amounts of memory and so quickly that before I can do anything useful it's gone into swap and then the OOM killer goes on a murder spree
[19:31:59] gregL (gregL!~greg@cpe-74-76-125-87.nycap.res.rr.com) has quit (Remote host closed the connection)
[19:38:32] jcdutton (jcdutton!~jcdutton@5ac34c4d.bb.sky.com) has quit (Ping timeout: 276 seconds)
[19:58:18] Beirdo: whoah
[19:58:24] Beirdo: who the heck?
[19:58:29] Beirdo: herr how
[19:58:42] Beirdo: stupid fingers
[19:58:53] stuartm: testing a fix after I manage to run under valgrind for a second
[20:00:05] stuartm: use of delete instead of free() on a calloc'd integer array
[20:01:52] highzeth (highzeth!~hz@213.138.162.181) has joined #mythtv
[20:02:36] stuartm: I don't know that it's the only problem, the disc might be damaged or otherwise unnavigable leading to us calling this specific bit of code a lot
[20:02:50] stuartm: might well be stuck in a loop tbh
[20:04:13] stuartm: heh, yeah, that doesn't fix it
[20:12:35] Jordack (Jordack!~jordack@h69-131-44-221.plmomi.dedicated.static.tds.net) has quit (Quit: NSFW! http://www.youtube.com/watch?v=-6xENiFq0B4)
[20:13:56] Beirdo: good to fix that anyways :)
[20:14:35] stuartm: it's leaking in dvdnav_describe_title_chapters – disc reports 99 titles which is unusual for a start and might be relevant – a large number of chapters could cause a significant growth
[20:15:13] stuartm: http://pastebin.com/yZwPYxdR
[20:15:36] stichnot: 99 titles sounds like a copy protection thing
[20:18:06] stichnot: where each title except one represents some permutation of the chapters
[20:20:22] jcdutton (jcdutton!~jcdutton@5ac34c4d.bb.sky.com) has joined #mythtv
[20:21:52] mrand (mrand!~mrand@ubuntu/member/mrand) has quit (Ping timeout: 244 seconds)
[20:24:48] stuartm: hmm, this isn't our code leaking and I could spend hours getting up to speed on dvdnav (including understanding dvd structures) only to still not know enough to create a fix – might be better to a) Resync with upstream b) Don't attempt to grab chapter times if it reports a very high number of titles
[20:26:45] stichnot: Do you recall why we bother to grab all the chapter times in the initial disc scan?
[20:27:50] stuartm: I don't really understand what we're attempting to do with that code, it looks like we're trying to automatically establish which is the main title, but every DVD I've played recently has played from the beginning through the menus, so why do we need to know the main title if we're not going to skip straight to it?
[20:28:20] stuartm: stichnot: heh, I didn't notice you'd sent that, no, I don't
[20:28:34] stichnot: what you said rings a bell
[20:30:35] stichnot: once we guess the main title, do we actually do anything with it? if not, why bother?
[20:30:36] stuartm: commercial DVD players surely don't need to do this, they just start a pre-defined, or metadata defined start title surely?
[20:31:28] stichnot: yeah, I think they just jump straight into the root menu
[20:32:22] stuartm: stichnot: we call dvdnav_title_play() with that as the argument, but that implies behaviour which I've not actually witnessed, every single DVD I've played through mythfrontend in the last few months has started at the beginning, playing through the still frames/trailers/menus
[20:32:47] stuartm: so it's not working even if it's meant to skip those
[20:35:07] stuartm: commenting out that logic entirely to see what happens
[20:46:02] j-rod is now known as j-rod|afk
[20:46:54] jcdutton: stuartm, I wrote some of dvdnav, what is the problem?
[20:47:18] mrand (mrand!~mrand@ubuntu/member/mrand) has joined #mythtv
[20:47:19] stuartm: yeah, removing the 'guess the main title' bit allows the disk to be played without issue
[20:48:10] stuartm: jcdutton: we were calling dvdnav_describe_title_chapters for every title on the disc, for Lost Season 6 Disc 1 (PAL UK) it caused a _massive_ memory leak/growth
[20:48:35] stuartm: jcdutton: valgrind reported the following – http://pastebin.com/yZwPYxdR
[20:49:46] stuartm: jcdutton: fwiw I'm leaning to just dropping the code that called dvdnav_describe_title_chapters as it is flawed and doesn't really work as it was intended to
[20:50:40] jcdutton: stuartm, I did not know about that function, it must be new
[20:51:00] jcdutton: stuartm, which version of libdvdnav are you using?
[20:51:05] Seeker`: Sounds similar to the way blurays are processed, as bluray menus can't be displayed
[20:51:32] Seeker`: so there isn't a way of navigating the disc easily, so it starts on the longest title (which isn't always the right one )
[20:51:55] jcdutton: Seeker`, its complicated
[20:53:18] Seeker`: stuartm: any idea if it is carried over from a time where myth couldn't display menus?
[20:53:31] stuartm: Seeker`: only for DVDs we can display menus and it's working reasonably well, plus guessing the main title on any disc is difficult and error prone, I don't think it's better to play the wrong title 1 time in 10 just to avoid the still frame copywrite warnings etc
[20:54:02] natanojl: Beirdo: What permissions should I have on trac? I can't seem to modify tickets
[20:54:28] Beirdo: no?
[20:54:30] stuartm: Seeker`: oddly no, that code was recently re-factored and I think this was added in at that time, Stanley removed the skip to main title stuff years ago when he added menu support to MythTV
[20:54:31] Beirdo: pne sec
[20:54:33] jcdutton: stuartm, Yes, just looked at that function. Its a mess. Leaks all over the place
[20:55:11] Seeker`: stuartm: are there some formats of menu that can't be displayed?
[20:55:15] Beirdo: natanojl: so true, one sec, let me fix that
[20:55:34] taylorr: stichnot: I don't have time to look into it but prior to 0.25 the total_duration in the database for a recording shouldn't exist.... so my best guess is that the fps detected might be wrong or the number of total frames is off... dunno
[20:55:39] stuartm: jcdutton: ok, glad to have that official confirmation that it's leaky – it makes it easier to rip out this code from mythtv without arguments
[20:56:02] jcdutton: stuartm, It can be fixed quite easily.
[20:56:10] Beirdo: natanojl: try now
[20:56:34] taylorr: stichnot: the ticket submitter needs to provide a -v playback --loglevel debug logfile for the bad duration recordings
[20:56:41] stuartm: jcdutton: it's not clear what version we're using, it was re-synced last June though
[20:56:46] jcdutton: stuartm, the problem is that vm_ifo_close() is not being called in all cases
[20:57:15] jcdutton: stuartm, What do you mean re-synced? Why not just link to libdvdnav?
[20:57:58] natanojl: Beirdo: Yup, looks like I can do more stuff than 'leave as new' now :) Thanks!
[20:58:15] Seeker`: http://code.mythtv.org/trac/attachment/ticket . . . g-leak.patch ?
[20:58:27] Beirdo: sorry, I'd missed adding you to the devs group in trac itself
[20:59:22] stuartm: jcdutton: that's historical, I'm not sure anyone who is around remembers the original reason – could be that when we started to include our own copy it wasn't yet stable enough or available on all distributions, or maybe we needed to make some customisations
[20:59:58] jcdutton: stuartm, The problem is the if(!retval && ifo)
[20:59:58] jcdutton: 793 vm_ifo_close(ifo);
[21:00:14] jcdutton: change that to if (ifo)
[21:00:19] jcdutton: and it will probably be fixed
[21:00:26] stuartm: there are advantages to using a copy rather than linking, namely that everyone is using the same version of the lib and if there are bugs it's easier to identify whether they are in our code or libdvdnav
[21:00:28] natanojl: Beirdo: no problem
[21:01:20] jcdutton: stuartm, make that change, and see if it still leaks
[21:02:19] stuartm: jcdutton: thanks, I'll give it a go
[21:03:13] stichnot: taylorr: yeah, my guess is there is something weird in the PVR-250 (or whatever) recording that confuses ffmpeg. If I'm lucky, I will have some old PVR-150 recordings that might show the problem.
[21:04:04] stuartm: jcdutton: always helps to have someone familiar with the code to find/fix these bugs quickly
[21:04:30] taylorr: stichnot: in the past ffmpeg has used a non-av track to determine duration which is wrong... I actually wrote a function to fix the issue in avformatdecoder.cpp
[21:05:01] stichnot_ (stichnot_!~chatzilla@mythtv/developer/stichnot) has joined #mythtv
[21:05:12] jcdutton: stuartm, also, myth will have to free *times when it has finished with it.
[21:05:56] stuartm: jcdutton: it does (at least in master as of an hour ago)
[21:06:52] stuartm: for some reason it was calling delete instead of free, but it's now correctly free()'d
[21:07:19] jcdutton: stuartm, have you tried my suggested change to if(!retval && ifo) ?
[21:07:24] stichnot (stichnot!~chatzilla@mythtv/developer/stichnot) has quit (Ping timeout: 245 seconds)
[21:07:26] stichnot_ is now known as stichnot
[21:08:47] stuartm: jcdutton: doing so, just recompiling/installing now
[21:09:17] jcdutton: stuartm, the only bits that allocate memory in that function are "ifo = vm_get_title_ifo(this->vm, title);" and "tmp = calloc(1, sizeof(uint64_t)*parts);"
[21:09:56] jcdutton: The "tmp = calloc(1, sizeof(uint64_t)*parts);" is cleaned up by freeing *times
[21:10:33] stuartm: yeah, it was leaking at both, I added a fix for the latter but I didn't spot where we were leaking ifo
[21:11:19] jcdutton: stuartm, the ifo is cleaned up but just making sure all paths through the function call "vm_ifo_close(ifo);" <- The equivalent of free() for ifo
[21:11:48] pepsiman (pepsiman!~mparsons@secure.linguamatics.com) has joined #mythtv
[21:12:07] pepsiman: I'm running mythtv 2:0.25.0+fixes.20120417.00a2d87–0ubuntu0mythbuntu3. I use mythwelcome, I've got a mythshutdown process using 100% which is stopping mythwelcome from painting. It looks like http://code.mythtv.org/trac/ticket/10032 which is supposed to be fixed. Any ideas?
[21:12:12] jcdutton: stuartm, there might be other mem leaks due to bugs in the vm_ifo_close(), but that is another story
[21:12:49] stuartm: jcdutton: changing it from if (!retvat & ifo) to if (ifo) doesn't fix the leak
[21:13:14] stuartm: at least not the big leak I'm seeing
[21:13:20] zombor (zombor!~zombor_@kohana/developer/zombor) has joined #mythtv
[21:15:33] jcdutton: stuartm, another thing to change, change the "if(!retval && tmp)" to "if (tmp)"
[21:15:58] jcdutton: Can I ask a question: how do I get mythbackend to change "Using runtime prefix =..."
[21:16:15] stuartm: k – won't be able to test that tonight now, I want to get on and watch this dvd so I can post it back in the morning
[21:16:47] jcdutton: stuartm, dvdbackup is your friend
[21:18:14] stuartm: I'll make sure to test before I send the disc back, but for now I'd rather just watch it ;)
[21:19:01] jcdutton: Ok, I still use xine to watch DVDs
[21:19:20] jcdutton: being that I wrote the dvd menu playing part of xine
[21:20:16] jcdutton: stuartm, one thing to note, libdvdnav caches a lot of information, so what you see as a leak might just be caching
[21:21:03] jcdutton: stuartm, but the size of the cached info should not be much bigger that the total size of all the .IFO files on the DVD
[21:21:04] stuartm: jcdutton: hence why I described it as a leak/growth, it might just be caching but 2GB of caching is excessive ;)
[21:22:28] jcdutton: stuartm, 2GB is probably too much. Are you sure the problem is that one function?
[21:22:36] stuartm: anyhow, as I said at the top, the code we had calling GetChapterTimes on every single title wasn't worth keeping, I've just pushed a commit which comments it out and renders the leak/caching moot (at least for MythTV)
[21:23:30] stuartm: jcdutton: that's what valgrind was pointing to and I don't have any better tools at my disposal right now, I do know that not calling that function prevents the problem
[21:23:51] jcdutton: Ok, good suspect then. ;-)
[21:24:38] jcdutton: Does anyone know how to change "Using runtime prefix = /usr" to something else?
[21:24:57] skd5aner: stichnot: no, it seems to be sporadic (re: reported length) – want more specific reports?
[21:25:08] stuartm: dvd playback startup is much faster with our code to guess the main title gone
[21:28:31] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has quit (Ping timeout: 244 seconds)
[21:29:16] taylorr: skd4aner: you need to provide a logfile with -v playback --loglevel debug
[21:30:28] stuartm: Beirdo: so you know when I said a purely numerical shortened sha1 would be rare? https://github.com/MythTV/mythtv/commit/12372544
[21:30:58] stuartm: I guess not _that_ rare
[21:34:05] skd5aner: taylorr: is skd4aner like my younger brother?
[21:34:06] skd5aner: ;)
[21:34:32] Beirdo: hehe, still rare, but yeah.
[21:35:24] Beirdo: not as rare as one might think (there are more numeric digits than alpha digits in hex)
[21:35:53] stuartm: aye
[21:36:44] skd5aner: dumb question – but in the new logging system, can I specific a specific debug file to output to so that I can easily find it amongst all the others in my logpath dir?
[21:36:51] stuartm: hmm, since Mark isn't here to argue, maybe I should just delete that code rather than commenting it out
[21:37:17] Beirdo: probability of x numeric-only digits in a shortened SHA1 of length n is = (10/16)^n
[21:37:47] Beirdo: skd5aner: no, it will be app.date.pid.log ;)
[21:38:12] Beirdo: I'm considering extending it to allow overrides for debugging
[21:38:16] Beirdo: but it's not coded yet
[21:40:45] Beirdo: let me go add that to the TODO list so it's not forgotten
[21:49:03] Beirdo: stuartm: if you are in the mood to delete code, go nuts :)
[21:49:24] skd5aner: taylorr, stichnot: attached the log as requested to #10623 – thanks! Let me know if you need any more details
[21:49:24] ** MythLogBot http://code.mythtv.org/trac/ticket/10623 **
[21:49:28] Beirdo: one of us has to maintain it, and if (for now) that's you... have a machete!
[22:05:50] stuartm: Beirdo: wouldn't be the first time, a long time ago I was the de-facto maintainer for dvd stuff, although all I ever did was to maintain it in the state I found it with the exception of a few simple bugs
[22:07:21] stuartm: I use it so it's in my interest that it works, but that's about as far as it goes, I've no desire to spend any time re-factoring or fixing bugs that don't affect me directly
[22:08:14] Beirdo: probably more interest than most of us at the moment :)
[22:08:33] stuartm: just as well really, I think half the bugs we deal with each release stem from one re-factor or another, would be nice to see fewer of them in future
[22:08:59] Beirdo: we really need to do some regression testing
[22:09:09] Beirdo: easier said than done, I do realize
[22:14:06] highzeth (highzeth!~hz@213.138.162.181) has quit (Ping timeout: 246 seconds)
[22:14:59] stuartm: I'd love to see some standardised testing, but yeah, much easier to say than to do – someone has to devise those tests, document the expected results and overcome issues such as the fact that no two developers have the same hardware, nor can we say that the same hardware will be available across releases (we all upgrade periodically or hardware fails)
[22:22:08] natanojl (natanojl!~jonatan@c83-252-237-63.bredband.comhem.se) has quit (Ping timeout: 272 seconds)
[22:22:25] stuartm: heh, Ohloh want to interview me for some sort of article on how Open Source projects use and benefit from Ohloh
[22:22:28] jcdutton (jcdutton!~jcdutton@5ac34c4d.bb.sky.com) has quit (Quit: Leaving)
[22:24:34] stuartm: think I'll decline, since my first reaction to that question is 'Huh? Uh ... hmm. Dunno."
[22:29:56] stuartm: good job I'm not in PR or communication
[22:29:58] hashbang (hashbang!~alex@213-152-35-50.dsl.eclipse.net.uk) has quit (Quit: Leaving.)
[22:32:09] Beirdo: hehe. Wonder if we have a good candidate to do that
[22:32:51] jya: I like Ohlo… it made me feel like , oh cool, I didn't waste my life away for nothing in the past 2 years...
[22:33:00] Beirdo: "we'd benefit if someone paid us the $45M you say our project is worth"
[22:33:30] Beirdo: the one use I have for it is to show a commit count
[22:52:03] stichnot: taylorr: do you see anything useful from the 10623 logs? seems to me we'd really like to know what av_estimate_timings() is computing.
[22:58:39] taylorr: stichnot: no, nothing helpful... probably need to add some extra debug logging
[22:59:11] taylorr: if the seektable exists it shouldn't use ffmpeg for anything other that the fps
[22:59:15] stichnot: taylorr: probably easiest to just get hold of an entire recording to play with
[22:59:35] taylorr: right... doesn't #10622 have a sample
[22:59:35] ** MythLogBot http://code.mythtv.org/trac/ticket/10622 **
[23:00:12] stichnot: 10622 is probably a different issue
[23:00:36] stichnot: apparently its problem is only for post 0.25 recordings
[23:00:51] stichnot: I'll look at the sample tonight
[23:01:07] skd5aner: taylorr, stichnot: keep in mind, that on 16023, the issue is with recordings PRIOR to 0.25 (as stichnot mentioned, different than 10622, which is issues with recordings post upgrade)
[23:01:52] skd5aner: on 16023, new recordings via my pvr-250/500s show the correct duration
[23:01:56] stichnot: I may have to dust off my old PVR-150 and generate test recordings...
[23:04:29] skd5aner: stichnot: also, sorry for playing back 2 recordings in that log... the second playback is the one that's messed up
[23:04:40] skd5aner: forgot I had already "fixed" the first one
[23:04:46] skd5aner: :)
[23:04:53] stichnot: ok, thanks
[23:05:23] gregL (gregL!~greg@cpe-74-76-125-87.nycap.res.rr.com) has joined #mythtv
[23:05:50] stichnot: skd5aner: are you in a position to upload an entire "short" problematic recording if needed?
[23:06:12] stichnot: not now, but just in case
[23:10:22] skd5aner: sure, but don't really have an easy way to host it
[23:17:24] stichnot: ok, hopefully I can reproduce the problem without a gigantic file
[23:45:55] taylorr: stichnot: weird... I just ran the video for #10622 through mythcommflag and avformatdecoder finds it to be 38.136 seconds which is correct... when I playback it's 17 seconds
[23:45:55] ** MythLogBot http://code.mythtv.org/trac/ticket/10622 **
[23:46:23] taylorr: playback should be using the 38.136 secs strored in the database
[23:47:27] stichnot: you mean playback will display something like "0:33 of 0:17" ?
[23:48:11] taylorr: no it displays "0:17 or 0:17" the entire time
[23:48:25] stichnot: wow
[23:48:30] taylorr: so position and duration are busted
[23:49:25] taylorr: time to play with the kids... I'll add some debug to mythplayer later tonight
[23:49:58] stichnot: ok, thanks
[23:57:20] jya: trac code browser is quite slow :(
[23:57:50] jya: compare to github that is

IRC Logs collected by BeirdoBot.
Please use the above link to report any bugs.