MythLogBot@irc.freenode.net :: #mythtv

Daily chat history

Current users (90):

MythLogBot, bbc581, caelor, Captain_Murdoch, Computer_Czar, dagar, darkdrgn2k, elmojo, ghoti, gregL, highzeth, holomntn, ikonia, J-e-f-f-A, jcarlos, jpabq, jpabq-, kc, knightr, kurre, leprechau, mike|2, PointyPumper, purserj, skd5aner, sphery, Splat1, stuartm, ThisNewGuy, tomimo, wagnerrp, xris, aloril_, Anduin, Anssi, anykey__, beata_, Beirdo, brfransen, cesman, Chutt, clever, coling, Cougar, dblain, eharris, elvum, foobum, foxbuntu, Gibby, gigem, GreyFoxx, hads, iamlindoro, jams, jannau, joe__, justpaul, jwhite, kormoc, laga, mag0o, markk_, mrand, okolsi, pheld, poptix, reynaldo, rhpot1991, rooaus, Snow-Man, stuarta, superm1, sutula, tgm4883, tris, weta, ybot, _charly_, jarle, h7251, j-rod|afk, jstenback, JEDIDIAH__, gbutters1, jya, Dave123, Diverdude, XChatMav, MythBuild
Saturday, January 29th, 2011, 00:38 UTC
[00:38:00] kormoc is now known as kormoc_afk
[00:39:04] Kunalagon1 (Kunalagon1!~Kunalagon@212.200.240.190) has quit (Quit: Leaving.)
[01:06:46] elmojo (elmojo!~elmojo@unaffiliated/elmojo) has quit (Remote host closed the connection)
[01:09:17] elmojo (elmojo!~elmojo@cpe-075-177-177-233.nc.res.rr.com) has joined #mythtv
[01:09:18] elmojo (elmojo!~elmojo@cpe-075-177-177-233.nc.res.rr.com) has quit (Changing host)
[01:09:18] elmojo (elmojo!~elmojo@unaffiliated/elmojo) has joined #mythtv
[01:18:51] kormoc_afk is now known as kormoc
[01:25:09] hashbang (hashbang!~alex@213-152-35-50.dsl.eclipse.net.uk) has quit (Quit: Ex-Chat)
[01:31:31] gigem_ (gigem_!~david@mythtv/developer/gigem) has joined #mythtv
[01:35:18] davide_ (davide_!~david@mythtv/developer/gigem) has quit (Ping timeout: 250 seconds)
[01:48:24] Captain_Murdoch: stuartm, I agree on getting rid of mythwelcome as well. I don't know all of what it does, but it always struck me as odd that the things I do know of were put into there instead of directly into mythfrontend.
[01:56:52] iamlindoro: I think it's odd too, but in the absence of that getting done, giving some feedback that the FE is loading is a good thing
[01:57:07] iamlindoro: (this presumes stuartm's comment was spurred by my taking the MythWelcome ticket)
[02:03:30] sphery: IIRC, the main reason MythWelcome exists as a separate app is that a frontend's connecting to the backend blocks backend auto-shutdown. So, if we had better tracking of what's happening and whether the frontend is idle or doing something, it shouldn't be missed.
[02:03:56] kormoc: couldn't we have the frontend not stay connected to the backend on the main menu?
[02:04:20] iamlindoro: Frontend still has to theoretically handle events coming from the BE, even when idle
[02:04:45] kormoc: Ahh, true enough
[02:04:50] iamlindoro: Granted not a *whole* lot right now, but hopefully lots more going forward
[02:05:16] iamlindoro: ie theme updates, notification that new videos have been scanned by the backend, maybe some other niceties that I think about adding sometimes :)
[02:05:44] iamlindoro: ie having fun little themable "what's new" video widgets that cycle through the newest recorded content, etc.
[02:05:57] iamlindoro: That's the kind of stuff I'd find neat, anyway
[02:07:12] iamlindoro: But even with that kind of thing on offer, we'd still be stuck with ShitCenter-wide ;)
[02:09:25] iamlindoro: I have fantasies about having all optical and removable media accessible on all systems, too, that would be neat
[02:09:41] iamlindoro: I think Captain_Murdoch has thought about stuff up that alley, ie every frontend is a limited SG host as well
[02:10:13] iamlindoro: So drop a DVD in in the Living room, watch it from the bedroom without taking it along, etc.
[02:10:31] iamlindoro: but that would mean catching mount/media events across frontends and backends
[02:10:32] kormoc: Yeah, that'd be really neat
[02:10:38] iamlindoro: What were we talking about again?  :)
[02:10:50] kormoc: iamlindoro, well, not overly, you could have it update on the MBE and everyone pull from there
[02:10:58] kormoc: %s/pull/poll/
[02:11:27] iamlindoro: Right, but point being you still need to be ready to handle backend events from the menu
[02:11:37] iamlindoro: anyway, just pie in the sky at this point
[02:11:38] kormoc: yeah, fair 'nuff
[02:16:13] wagnerrp: iamlindoro, kormoc: backend shutdown is controlled by the type of connection you make to the backend
[02:16:48] wagnerrp: if you announce as 'Playback', backend shutdown is blocked
[02:16:59] wagnerrp: if you announce as 'Monitor', the backend can shutdown
[02:18:23] wagnerrp: you can still receive events if registered as a monitor, but you would need to reconnect to re-announce
[02:27:03] iamlindoro: Do we have a solution for the fact that mythfrontend won't shut down (while tearing down the uPnP client) yet?
[02:27:17] iamlindoro: Think this is one of the outstanding issues w/ MythSystem, maybe?
[02:29:33] wagnerrp: not one im aware of, whats it doing?
[02:29:58] iamlindoro: just sits at
[02:30:10] iamlindoro: Deleting UPnP client...
[02:30:12] iamlindoro: forever
[02:30:57] wagnerrp: never noticed that... but to be honest, whenever im testing, i just ctrl+c the console
[02:31:16] iamlindoro: yeah, I can't test MythWelcome changes since I can't ever exit the FE
[02:31:33] iamlindoro: and attempting to kill a process that MythSystem has spawned crashes everything
[02:31:55] iamlindoro: (which makes one wonder what happens when you run something with MythSystem and that process crashes)
[02:32:14] iamlindoro: Here's what the FE is up to: http://mythtv.pastebin.com/brmkK9PT
[02:32:25] kormoc: I wonder if that's why everything sigfaults on exit now
[02:32:40] kormoc: (everything being mythfrontend, mythfilldatabase, mythcommflag, etc)
[02:32:50] iamlindoro: yeah, I've noticed that too
[02:33:20] iamlindoro: This may actually be related to MDM, it seems like
[02:33:37] iamlindoro: Captain_Murdoch, Can you look at the BT of the hung FE process above?
[02:33:56] iamlindoro: It sort of looks like the MDM thread is stuck on wait()
[02:34:21] iamlindoro: presumably in the dtor
[02:35:43] iamlindoro: yeah, stuck at m_queueWaitCond.wait(&m_queueWaitLock, 200);
[02:35:57] iamlindoro: It thinks it's downloaing *something*, anyway, and waits forever
[02:38:22] ** wagnerrp is happy hes not at fault **
[02:39:00] iamlindoro: Heh, I've seen the segfault thing that kormoc has seen, too, but that's likely unrelated to this hang
[02:39:05] bbc581: I would love a solution for my android phone to build a watch live tv. have you guys seen any blog entries showing how to configure mythweb? I don't think any of wiki articles do any justice compated to a complete walkthrough. maybe something with troubleshooting suggestions?
[02:39:10] iamlindoro: the segfault might still be your cault ;)
[02:39:16] iamlindoro: bbc581, Wrong channel, see topic
[02:39:33] wagnerrp: as long as its not my fault, im fine with caults
[02:39:42] iamlindoro: a cault is so much worse
[02:40:06] iamlindoro: It's where people all all worship some guy with long hair as the second coming
[02:40:09] iamlindoro: you know, a cault
[02:40:46] wagnerrp: oh, the kind of thing you start for money and power
[02:40:54] wagnerrp: yeah, ill take one of those
[02:43:52] kormoc: wagnerrp, you drink the poisoned koolaid first, okay?
[02:44:05] wagnerrp: sure
[02:44:23] wagnerrp: luckily, ive been drinking the poison every day for the past five years
[02:44:30] wagnerrp: first in small amounts
[02:44:51] wagnerrp: slowly increasing to ever larger amounts
[02:48:34] kormoc: Curses to the man in black and his iocane powder!
[02:51:08] knightr (knightr!~knightr@mythtv/developer/knightr) has quit (Ping timeout: 250 seconds)
[02:51:21] knightr (knightr!~knightr@mythtv/developer/knightr) has joined #mythtv
[02:52:07] wagnerrp: mythcommflag segfaulting?
[02:53:02] kormoc: on exit, aye
[02:53:11] wagnerrp: i dont think the mythsystem stuff can be causing it
[02:53:21] wagnerrp: since it doesnt even spawn its worker threads until the first use
[02:53:47] wagnerrp: and AFAIK, mythcommflag never calls any external executables
[02:53:54] kormoc: hrm, fair 'nuff
[02:54:00] iamlindoro: Does anything in mythcontext/mythcorecontext
[02:54:01] iamlindoro: ?
[02:54:13] wagnerrp: ill look though the destructor though
[02:54:17] iamlindoro: Since that's likely where the issue is, since it's in all apps
[02:54:20] wagnerrp: maybe its deleting something twice
[02:54:58] sphery: could also be more of #7714
[02:57:45] kormoc: mythdb* was renamed mythbase*, any other changes?
[02:58:00] wagnerrp: couple other files were shifted around
[02:58:20] kormoc: drat, time to dig the github mines
[03:49:22] Beirdo: kormoc: you do have the latest latest code, right? :)
[03:55:22] kormoc: Nah
[03:57:16] Beirdo: ah. anyways, if you guys have backtraces for me to dig into, I'd be happy to investigate, especially if it's in stuff I've messed with
[03:57:53] ** Beirdo goes off to order from Zeek's **
[04:08:55] kenni (kenni!~kenni@mythtv/developer/kenni) has quit (Read error: Operation timed out)
[04:13:50] jpabq (jpabq!~jpabq@mythtv/developer/jpabq) has quit (Remote host closed the connection)
[04:14:51] jpabq (jpabq!~jpabq@mythtv/developer/jpabq) has joined #mythtv
[05:02:05] kormoc: hrm
[05:02:17] kormoc: should we allow the backend to stop while mythfilldatabase is running?
[05:02:30] kormoc: tho I guess it doesn't really matter
[05:07:04] wagnerrp: the only thing mfd needs the backend for is to call a reschedule
[05:07:12] wagnerrp: which will be done automatically when the backend starts up anyway
[05:07:44] kormoc: well, the backend starts up with a bunch of the program table gone, etc
[05:07:52] kormoc: it gets a little weird until the fill finishes
[05:08:05] Captain_Murdoch: iamlindoro, that BT doesn't indicate that MDM is hung. it shows that MDM is still inside its "while (m_runThread)" loop which indicates that the dtor hasn't been called yet. when the dtor is called, m_runThread is set to false and anything waiting on the QWaitCondition is awakened and the dtor waits for the thread to exit so the dtor can finish. if it was hung in the dtor, you'd see that in the BT as well. currently it looks l
[05:08:05] Captain_Murdoch: ike I never even call the dtor or delete the MDM, so it just dies when the program exits. trying to see if I can put in atexit() hander now, but am getting a weird "QObject::startTimer: QTimer can only be used with threads started with QThread" error after the dtor completes whether I call it the teardown function with atexit() or via MythCoreContext's dtor.
[05:09:27] Captain_Murdoch: kormoc, same thing applies if the BE was running the whole time.
[05:10:08] kormoc: If a resched was run during that time, true enough
[05:10:28] Captain_Murdoch: If you're asking should we allow the MBE to shutdown if I hit CTRL+C during a MFDB run, then yes, cause I do that often on my dev box. :)
[05:11:50] kormoc: it was more a WTF moment when things didn't appear to come up correctly after a update
[05:17:26] markk_: has something changed recently with FE<->BE communications? certain sockets connections seem to be failing as the frontend is now trying to look for masterbackend:6543 rather 192.168.0.2:6543
[05:20:34] wagnerrp: markk_: sounds like you put a hostname in the ip box in mythtv-setup
[05:24:29] markk_: wagnerrp: just checked – all still plain ip's. and it's only certain requests – I only noticed because the disk utilisation figures are missing from the Watch Recordings screen
[05:25:50] markk_: ah – something to do with the themedownloader and the the tmp storage group.
[05:26:22] markk_: Captain_Murdoch: ^^
[05:27:50] markk_: the theme path is stored as myth://Temp@masterbackend/remotethemes/trunk/Arclight rather than @192.168.0.2
[05:29:30] Captain_Murdoch: I can change that to use MBE's IP from MasterServerIP setting
[05:29:45] Captain_Murdoch: right now it calls gCoreContext->GetMasterHostName() instead.
[05:30:27] markk_: Captain_Murdoch: might explain why I get theme download errors on remote fe's
[05:34:09] Captain_Murdoch: probably. I'll have a fix in in a few. trying to get MythDownloadManager to shutdown without any warnings right now. think I tracked the issue down and am testing a fix for that now.
[05:37:45] bbc581 (bbc581!~bbc581@74-129-133-90.dhcp.insightbb.com) has quit (Read error: Connection reset by peer)
[05:37:53] xris: grumble. rpm package builder has install issues. looks like it still needs a little help
[05:49:16] superm1: stuartm, i'm assuming only master and later though that's happening, won't be backported to fixes right?
[05:54:09] chelz (chelz!~iridesenc@unaffiliated/chelz) has joined #mythtv
[05:54:53] gbutters1 (gbutters1!~Gary@ip68-11-58-116.no.no.cox.net) has joined #mythtv
[05:55:04] simonckenyon (simonckenyon!~simoncken@195.7.61.12) has joined #mythtv
[05:55:34] simonckenyon (simonckenyon!~simoncken@195.7.61.12) has quit (Remote host closed the connection)
[05:55:58] gbutters1 (gbutters1!~Gary@ip68-11-58-116.no.no.cox.net) has quit (Read error: Connection reset by peer)
[05:56:22] gbutters1 (gbutters1!~Gary@ip68-11-58-116.no.no.cox.net) has joined #mythtv
[05:57:53] gbutters (gbutters!~Gary@ip68-11-58-116.no.no.cox.net) has quit (Ping timeout: 240 seconds)
[06:00:47] simonckenyon (simonckenyon!~simoncken@195.7.61.12) has joined #mythtv
[06:05:24] h7252 (h7252!~h725@62-47-246-91.adsl.highway.telekom.at) has quit (Quit: Leaving.)
[06:09:04] Captain_Murdoch: iamlindoro, what more could you expect from a user with a silly name.
[06:19:04] dougt_ (dougt_!~dougt@c-24-130-158-216.hsd1.ca.comcast.net) has joined #mythtv
[06:37:57] dougt_ (dougt_!~dougt@c-24-130-158-216.hsd1.ca.comcast.net) has quit (Quit: dougt_)
[06:45:57] okolsi: build seems to be broken due to the atexit change
[07:07:53] Kunalagon1 (Kunalagon1!~Kunalagon@212.200.241.68) has joined #mythtv
[07:10:40] simonckenyon (simonckenyon!~simoncken@195.7.61.12) has quit (Remote host closed the connection)
[07:11:08] simonckenyon (simonckenyon!~simoncken@195.7.61.12) has joined #mythtv
[07:30:08] markk_: okolsi: fix coming in a second or two...
[07:32:45] okolsi: :)
[07:34:59] innatech (innatech!~DF@cpe-76-173-194-208.socal.res.rr.com) has joined #mythtv
[07:35:04] innatech: exit
[07:35:18] innatech: (errp. wrong window.)
[08:07:26] Chutt (Chutt!~ijr@cpe-24-166-52-87.neo.res.rr.com) has quit (Ping timeout: 255 seconds)
[08:17:32] natanojl (natanojl!~jonatan@c83-252-237-63.bredband.comhem.se) has joined #mythtv
[08:22:26] Chutt (Chutt!~ijr@cpe-24-166-52-87.neo.res.rr.com) has joined #mythtv
[08:41:27] NightMonkey (NightMonkey!debian-tor@pdpc/supporter/professional/nightmonkey) has quit (Ping timeout: 240 seconds)
[08:51:40] okolsi: :q
[08:52:28] innatech (innatech!~DF@cpe-76-173-194-208.socal.res.rr.com) has quit (Quit: Leaving)
[08:59:45] okolsi: Beirdo: did some quick testing last night, latest patch didn't seem to help. Testing gets bit tricky with three patches which apply no-more (libmythdb changes)
[09:19:19] okolsi: markk_: looks like LiveTV is now broken here.. don't have any useful logs etc. yet
[09:20:04] mianos (mianos!~mrfish@124-171-45-127.dyn.iinet.net.au) has joined #mythtv
[09:20:11] mianos (mianos!~mrfish@124-171-45-127.dyn.iinet.net.au) has left #mythtv ()
[09:27:56] MaverickTech (MaverickTech!~MaverickT@111.86.233.220.static.exetel.com.au) has joined #mythtv
[09:39:00] stuartm: Captain_Murdoch: 0.24 build is broken
[09:39:09] stuartm: http://code.mythtv.org/buildbot/waterfall
[09:42:38] stuartm: iamlindoro, Captain_Murdoch, kormoc, sphery: The solution to replacing mythwelcome is just to have an 'idle' status for the frontends which the backend can track before deciding to shut down
[09:44:16] Goga777 (Goga777!~Goga777@shpd-92-101-150-59.vologda.ru) has joined #mythtv
[09:44:18] stuartm: when the backend shuts down the frontends can go into an offline mode and any activity triggers a backend wakeup
[10:10:59] stuartm: Beirdo: can we have MythBuild join here too? That seems easier than getting everyone into -commits and events like build failures are relevant in here
[10:27:56] chelz: i'm new so i don't know if it already does this, but maybe if the bot just reports failures and not every build it would be good in here
[10:37:43] stoffel (stoffel!~quassel@p57B4D2B1.dip.t-dialin.net) has joined #mythtv
[10:45:09] Goga777 (Goga777!~Goga777@shpd-92-101-150-59.vologda.ru) has quit (Remote host closed the connection)
[10:49:40] Goga777 (Goga777!~Goga777@shpd-92-101-150-59.vologda.ru) has joined #mythtv
[11:05:02] Guest50921 (Guest50921!~mike@c-24-21-63-118.hsd1.or.comcast.net) has quit (Remote host closed the connection)
[11:05:55] mike|2 (mike|2!~mike@c-24-21-63-118.hsd1.or.comcast.net) has joined #mythtv
[11:10:34] jya (jya!~avenardj@mythtv/developer/jya) has joined #mythtv
[11:20:44] okolsi: markk_: livetv log: http://mythtv.pastebin.com/FHNaxt9d
[11:22:31] markk_: okolsi: thanks – but I'm going to need a backtrace :) the logs aren't helping. was that working prior to the eof refactor
[11:22:34] markk_: ?
[11:23:26] okolsi: it was working like 2–5 days ago or so
[11:24:11] okolsi: local frontend and livetv works, this is remote frontend which fails
[11:24:54] markk_: okolsi: have you make dist cleaned ?
[11:25:38] okolsi: markk_: yes, i was wondering if it's better to start fe under gdb or attach it when playback hangs?
[11:26:40] markk_: okoldi: in this case, it probably makes no difference
[11:27:12] okolsi: markk_: okay. need to make debug build and see if I can get the bt
[11:27:48] markk_: okolsi: :) sorry – been a beery sort of afternoon (and mojitos!)
[11:28:22] okolsi: markk_: sounds like fun! :)
[11:37:26] stuartm: http://www.reghardware.com/2011/01/28/seesaw_boxee/
[12:01:03] okolsi: markk_: is this okay? http://mythtv.pastebin.com/ecj496F0
[12:31:16] gigem_ (gigem_!~david@mythtv/developer/gigem) has quit (Read error: Operation timed out)
[12:32:19] gigem_ (gigem_!~david@host137.12.intrusion.com) has joined #mythtv
[12:32:20] gigem_ (gigem_!~david@host137.12.intrusion.com) has quit (Changing host)
[12:32:20] gigem_ (gigem_!~david@mythtv/developer/gigem) has joined #mythtv
[12:47:57] Goga777 (Goga777!~Goga777@shpd-92-101-150-59.vologda.ru) has quit (Remote host closed the connection)
[12:48:02] danielk22 (danielk22!~danielk@14.96.215.146) has joined #mythtv
[12:49:40] andreax (andreax!~andreaz@p57B92EA4.dip.t-dialin.net) has quit (Ping timeout: 255 seconds)
[12:50:07] andreax (andreax!~andreaz@p57B9221A.dip.t-dialin.net) has joined #mythtv
[13:11:16] Kunalagon1 (Kunalagon1!~Kunalagon@212.200.241.68) has quit (Quit: Leaving.)
[13:17:42] SteveGoodey (SteveGoodey!~steve@host109-158-210-119.range109-158.btcentralplus.com) has joined #mythtv
[13:19:29] Dave123 (Dave123!~dave@cpe-74-74-222-96.rochester.res.rr.com) has joined #mythtv
[13:21:42] danielk22 (danielk22!~danielk@14.96.215.146) has quit (Quit: Leaving.)
[13:26:20] natanojl (natanojl!~jonatan@c83-252-237-63.bredband.comhem.se) has quit (Read error: Operation timed out)
[13:26:30] Kunalagon1 (Kunalagon1!~Kunalagon@212.200.241.68) has joined #mythtv
[13:30:26] stuartm: Captain_Murdoch: I'm seeing a loop in the download manager that's causing load for that thread to hit 100%
[13:31:08] stuartm: and staying there, for over 20 minutes now
[13:39:19] stuartm: Captain_Murdoch: it's spinning over and over trying to process yth://Temp@QUERY_HOSTNAME/remotethemes/trunk/Terra/themeinfo.xml
[13:39:34] stuartm: it fails, but keeps trying again
[13:43:11] SteveGoodey (SteveGoodey!~steve@host109-158-210-119.range109-158.btcentralplus.com) has quit (Remote host closed the connection)
[13:45:32] SteveGoodey (SteveGoodey!~steve@host109-158-210-119.range109-158.btcentralplus.com) has joined #mythtv
[14:03:10] stuartm: who is maintaining the perl bindings these days?
[14:05:51] simonckenyon (simonckenyon!~simoncken@195.7.61.12) has quit (Ping timeout: 240 seconds)
[14:07:56] Captain_Murdoch: stuartm, does that file exist? ~/.mythtv/remotethemes/trunk/Terra/themeinfo.xml
[14:10:54] stuartm: Captain_Murdoch: no
[14:12:13] stuartm: there isn't even a remotethemes directory
[14:17:08] Captain_Murdoch: QUERY_HOSTNAME comes from MythCoreContext::GetMasterHostName() failing to talk to the backend. so that would be fixed by last night's commit, it should be using the master's IP now. I don't know why it would be looping. do you have a BT?
[14:17:45] Captain_Murdoch: should be using master's IP, so it doesn't even call MythCoreContext::GetMasterHostName() although I'm not sure why that would fail unless your MBE wasn't up.
[14:23:36] stuartm: are we caching the value? The backend wasn't up when the frontend started
[14:23:57] stuartm: and I'm not sure whether I have last night's commit ...
[14:24:54] Captain_Murdoch: yeah, the value is cached, I'll fix that to check to make sure we got a valid result. I'll have to try to reproduce the loop. we only check that URL on the FE every hour via a QTimer. the MDM shouldn't be looping over it.
[14:24:59] stuartm: I've been stepping through with gdb, but a bt is difficult, it's a question of timing to catch useful information given that it only prints out the last ~30 steps
[14:25:44] stuartm: top points to that thread, so it's definitely the MDM
[14:27:19] Captain_Murdoch: I'll force my hostname bad and see if I can reproduce. I don't think markk was seeing anything like this and he had a 'bad' hostname as well since his hostname wasn't in DNS.
[14:35:08] stuartm: it's spinning in MythDownloadManager::run() – downloading is always true, m_downloadInfos always contains the above url
[14:36:55] stuartm: yeah, seems we only remove the url from the list when the download completes, in the event of an error it stays in the list?
[14:38:10] stuartm: just a guess
[14:41:04] chelz (chelz!~iridesenc@unaffiliated/chelz) has quit (Quit: chelz)
[14:41:50] stuartm: Captain_Murdoch: ok interesting, restarted the frontend and this time it found and successfully downloaded the file
[14:45:38] markk_: okolsi: I can't see from that bt where the deadlock is occurring (the playback thread is waiting for a lock and the decoder thread is sleeping ) – can you roll back to pre 679b668e3669a38bd08e (Refactor 'eof') and see if that fixes it?
[14:56:07] Captain_Murdoch: even in case of error, downloadInfos entry should get removed by ::downloadFinished() which is called inside RemoteFileDownloadThread::run(). that gives me more info to go on. I can replicate a failure and verify what happens. I'll look at it later today. the hostname issue wouldn't happen anymore, and I have a patch in-tree to not cache that bad value in case that happens anywhere else.
[14:59:22] SteveGoodey (SteveGoodey!~steve@host109-158-210-119.range109-158.btcentralplus.com) has quit (Remote host closed the connection)
[14:59:24] Captain_Murdoch: stuartm, ^^ just in case you didn't see that. I'll be back later.
[15:02:27] SteveGoodey (SteveGoodey!~steve@host109-158-210-119.range109-158.btcentralplus.com) has joined #mythtv
[15:06:34] fa2k (fa2k!~fa2k@muon.lan.fa2k.net) has joined #mythtv
[15:08:35] ** Captain_Murdoch still isn't sure (or just plain doesn't know) why his compile doesn't catch those missing "static" keywords so he blames it on old CentOS 5.x. sorry markk_ and thanks for the commits **
[15:12:13] natanojl (natanojl!~jonatan@c83-252-237-63.bredband.comhem.se) has joined #mythtv
[15:19:56] stuartm: Captain_Murdoch: I can't see how it happens within mythdownloadmanager, is there something higher up which would keep feeding the same url back?
[15:21:38] simonckenyon (simonckenyon!~simoncken@195.7.61.12) has joined #mythtv
[15:36:03] simonckenyon (simonckenyon!~simoncken@195.7.61.12) has quit (Remote host closed the connection)
[15:37:43] stuartm: Captain_Murdoch: build failures are reported by MythBuild in #mythtv-commits in the event that you don't actively monitor http://code.mythtv.org/buildbot/waterfall
[15:50:43] dougt_ (dougt_!~dougt@c-24-130-158-216.hsd1.ca.comcast.net) has joined #mythtv
[15:54:54] MavT (MavT!~MaverickT@111.86.233.220.static.exetel.com.au) has joined #mythtv
[15:57:43] h7251 (h7251!~h725@62-47-246-190.adsl.highway.telekom.at) has joined #mythtv
[15:57:46] MaverickTech (MaverickTech!~MaverickT@111.86.233.220.static.exetel.com.au) has quit (Ping timeout: 272 seconds)
[15:59:11] andreax (andreax!~andreaz@p57B9221A.dip.t-dialin.net) has quit (Quit: Leaving.)
[15:59:43] andreax (andreax!~andreaz@p57B9221A.dip.t-dialin.net) has joined #mythtv
[16:05:03] Diverdude (Diverdude!~Diverdude@1709ds1-vbr.0.fullrate.dk) has joined #mythtv
[16:05:05] Diverdude: I have some movie-files (atriology) called [01]moviename, [02]moviename and [03]moviename...But for some reason, when viewed in mythTV the [01], [02] and [03] parts of the filenames are ignored...Why?
[16:05:05] MythLogBot: SVN 02: (branch master) https://github.com/MythTV/mythtv/commit/654df684
[16:05:05] MythLogBot: No match for SVN revision 03
[16:05:24] Diverdude: hi
[16:18:58] okolsi: markk_: I'm trying further back, looks like the eof refactor is not causing it
[16:21:05] dougt_ (dougt_!~dougt@c-24-130-158-216.hsd1.ca.comcast.net) has quit (Ping timeout: 264 seconds)
[16:24:21] sn9 (sn9!lZcYW3@gimpelevich.san-francisco.ca.us) has joined #mythtv
[16:24:39] sn9 (sn9!lZcYW3@gimpelevich.san-francisco.ca.us) has left #mythtv ("This is a /part message.")
[16:29:51] stoffel (stoffel!~quassel@p57B4D2B1.dip.t-dialin.net) has quit (Ping timeout: 240 seconds)
[16:42:58] dougt_ (dougt_!~dougt@c-24-130-158-216.hsd1.ca.comcast.net) has joined #mythtv
[16:43:44] dougt_ (dougt_!~dougt@c-24-130-158-216.hsd1.ca.comcast.net) has quit (Client Quit)
[16:55:15] okolsi: markk_: livetv problem might have started before recent db schema upgrade, a bit problmatic to test older version since i'm playing with a production system..
[17:09:22] stuartm: markk_: any idea what might be causing the following – QGLContext::makeCurrent(): Failed
[17:14:11] stuartm: and my DVD playback issue with 0.24 is the following – "Waited 100ms for video buffers" ... what's that about?
[17:16:32] XChatMav (XChatMav!~MaverickT@111.86.233.220.static.exetel.com.au) has joined #mythtv
[17:19:10] MavT (MavT!~MaverickT@111.86.233.220.static.exetel.com.au) has quit (Ping timeout: 240 seconds)
[17:23:10] ** stuartm suspects broken drivers in Ubuntu **
[17:31:38] gigem_ (gigem_!~david@mythtv/developer/gigem) has quit (Remote host closed the connection)
[17:32:04] gigem_ (gigem_!~david@host137.12.intrusion.com) has joined #mythtv
[17:32:05] gigem_ (gigem_!~david@host137.12.intrusion.com) has quit (Changing host)
[17:32:05] gigem_ (gigem_!~david@mythtv/developer/gigem) has joined #mythtv
[17:52:58] stuartm: why would we be starved for frames decoding a SD DVD from disc?
[18:09:36] cattelan_away (cattelan_away!~cattelan@c-24-118-127-188.hsd1.mn.comcast.net) has quit (Ping timeout: 265 seconds)
[18:10:14] SteveGoodey (SteveGoodey!~steve@host109-158-210-119.range109-158.btcentralplus.com) has quit (Remote host closed the connection)
[18:11:35] Kunalagon1 (Kunalagon1!~Kunalagon@212.200.241.68) has quit (Ping timeout: 240 seconds)
[18:15:49] h7251 (h7251!~h725@62-47-246-190.adsl.highway.telekom.at) has quit (Read error: Operation timed out)
[18:17:10] Steve_Goodey (Steve_Goodey!~steve@host109-158-210-119.range109-158.btcentralplus.com) has joined #mythtv
[18:18:50] bbc581 (bbc581!~bbc581@74-129-133-90.dhcp.insightbb.com) has joined #mythtv
[18:19:05] bbc581 (bbc581!~bbc581@74-129-133-90.dhcp.insightbb.com) has quit (Remote host closed the connection)
[18:19:12] hno (hno!~hno@squid/developer/hno) has joined #mythtv
[18:19:35] hno (hno!~hno@squid/developer/hno) has left #mythtv ()
[18:19:56] bbc581 (bbc581!~bbc581@74-129-133-90.dhcp.insightbb.com) has joined #mythtv
[18:25:46] Kunalagon1 (Kunalagon1!~Kunalagon@212.200.241.42) has joined #mythtv
[18:25:58] Goga777 (Goga777!~Goga777@shpd-92-101-150-59.vologda.ru) has joined #mythtv
[18:52:14] Captain_Murdoch: stuartm, yeah, I'm normally in -commits but forgot to rejoin the other day after the power came back on.
[18:53:32] Beirdo: okolsi: I'll rebase the patches for you and combine them so you can try again easier :)
[18:54:03] Beirdo: stuartm: we could have the buildbot here too, that should be doable
[18:54:57] Captain_Murdoch: the only thing higher up than MDM is the ThemeUpdateChecker::checkForUpdate() in mythfrontend/themechooser.cpp that should only get called once an hour when m_updateTimer fires. There aren't any loops at all in ThemeUpdateChecker, just the ctor, dtor, and checkForUpdate().
[19:05:56] Beirdo: okolsi: new combined and rebased patch posted.
[19:07:02] kormoc: Captain_Murdoch, Know offhand a -v level I should use to poke at themechooser issues?
[19:07:20] Captain_Murdoch: file,extra and maybe network.
[19:07:27] kormoc: kk, thanks
[19:07:45] Captain_Murdoch: if you're having an issue downloading, and your MBE's hostname is not in DNS, update to HEAD to fix a bug that was fixed last night.
[19:08:43] kormoc: Kk, I'll give that a go but I think it's different. First download from MBE always fails, second one works
[19:09:35] Captain_Murdoch: yeah, add 'network' for something like that.
[19:11:19] MythBuild (MythBuild!~MythBuild@184-106-209-209.static.cloud-ips.com) has joined #mythtv
[19:11:25] Beirdo: there
[19:12:03] Beirdo: it's setup for only reporting exceptions and success->fail, fail->success
[19:15:27] Steve_Goodey (Steve_Goodey!~steve@host109-158-210-119.range109-158.btcentralplus.com) has quit (Ping timeout: 240 seconds)
[19:25:20] Captain_Murdoch: stuartm, if you see that happen again, adjust your running EE's log level via the NetworkControl socket with "set verbose network,file,extra" and see what messages it is printing. I don't see how it would be looping unless it didn't "finish" the download when RemoteFile couldn't open the connection, but I don't see how that could happen either (yet?)
[19:26:34] andreax (andreax!~andreaz@p57B9221A.dip.t-dialin.net) has quit (Quit: Leaving.)
[19:35:48] stuartm: I suspect that it got into the looping state when it failed because of the hostname issue, but I can't see how and now that the ip fix has gone in the chances of reproducing are reduced
[19:35:49] Captain_Murdoch: stuartm, I do see one other potential hang that's possibly what you ran into, but I'm not 100% sure. potentially for a RemoteFile download, if the download hangs, we could loop forever in MDM::downloadNow() since RemoteFile doesn't update m_lastStat. I'm inserting another check in there to handle RF hangs.
[19:36:00] Captain_Murdoch: timing... :)
[19:36:29] Captain_Murdoch: so if there is a bug somewhere in RemoteFile or MythCoreContext, it could have hung MDM::downloadNow()
[19:37:40] Captain_Murdoch: later tonight I'll get that fix into git along with the one for the caching issue in MythCoreContext::GetMasterHostName() where it caches bad values.
[19:38:09] ** Captain_Murdoch is afk again **
[19:43:52] stuartm: the backtrace fragments all pointed to run() but then most of those I captured were at moments it was in QUrl or QString, I could never get a clear picture of exactly where it was looping and run() with it's while loop was the best suspect
[19:48:31] Steve_Goodey (Steve_Goodey!~steve@host86-145-233-30.range86-145.btcentralplus.com) has joined #mythtv
[20:08:57] bbc5812 (bbc5812!~bbc581@32.163.34.83) has joined #mythtv
[20:10:30] bbc581 (bbc581!~bbc581@74-129-133-90.dhcp.insightbb.com) has quit (Ping timeout: 240 seconds)
[20:10:30] bbc5812 (bbc5812!~bbc581@32.163.34.83) has quit (Client Quit)
[20:35:22] kormoc: OS X builds are broken with "../libmythbase/mythmedia.h:36: error: conflicting declaration ?typedef enum MediaType MediaType?"
[20:35:54] sphery: heh, the build bot is using the same nick that kormoc uses
[20:46:10] MythBuild (MythBuild!~MythBuild@184-106-209-209.static.cloud-ips.com) has quit (Quit: buildmaster reconfigured: bot disconnecting)
[20:46:17] MythBuild (MythBuild!~MythBuild@184-106-209-209.static.cloud-ips.com) has joined #mythtv
[20:48:52] stuarta: kormoc: i'm hacking on the OSX builds tonight
[20:49:08] stuarta: once i finish securing a few cupboards from small children
[20:53:02] kormoc: Awesome :)
[20:54:58] stuarta: my aim is to have my mac do a variety of builds as part of the buildbot network
[21:17:16] fa2k (fa2k!~fa2k@muon.lan.fa2k.net) has quit (Quit: leaving)
[21:17:52] stuartm: kormoc: hmm, that's helpful – I wonder what it's conflicting with
[21:19:10] jya: jannau: when is the next ffmpeg resync scheduled? we had first talked about something being done before Christmas ..
[21:19:59] stuartm: jya: that was before jannau lost all his myth work to a hard drive failure
[21:20:07] jya: ah :(
[21:20:10] jya: that sucks..
[21:20:22] jannau: jya: I lost it unfortunately during the ubuntu installer event
[21:20:36] dekarl (dekarl!~deKarl@e180158177.adsl.alicedsl.de) has joined #mythtv
[21:22:06] jya: jannau: it's the ubuntu installer that cleaned your disk ?
[21:22:29] kormoc: stuarta, I'm planning to get my box into the buildbot network as well
[21:22:37] jannau: just overwrote the luks header
[21:22:42] stuartm: jya: it overwrote his encryption key header
[21:23:20] dekarl1 (dekarl1!~deKarl@e180142086.adsl.alicedsl.de) has quit (Ping timeout: 276 seconds)
[21:23:29] jya: ah
[21:23:45] kormoc: stuartm, /System/Library/Frameworks/QuickTime.framework/Headers/Movies.h:208: error: ?struct MediaType? has a previous declaration as ?struct MediaType?
[21:24:55] kormoc: stuartm, it's a system header that we bring in for the quartz rendering
[21:25:40] stuartm: kormoc: hmm, odd, it didn't conflict when MythMedia was in libmyth and we're including fewer frameworks in libmythbase
[21:26:27] stuartm: I don't know the best way to fix that, probably best left for an OSX dev
[21:27:45] jya: jannau: is there plan to do it anytime soon? I rely on some of the late changes, I have backported some of the DTS related changes ; but using a complete ffmpeg would be easier
[21:35:00] kormoc: stuartm, yeah, no worries, just letting folks know :)
[21:41:20] Goga777 (Goga777!~Goga777@shpd-92-101-150-59.vologda.ru) has quit (Remote host closed the connection)
[21:41:59] stuartm: kormoc: well it's my fault, I like to be the one to fix it, I'm just missing whatever the solution was previously
[21:58:57] MythBuild (MythBuild!~MythBuild@184-106-209-209.static.cloud-ips.com) has quit (Quit: buildmaster reconfigured: bot disconnecting)
[22:00:38] MythBuild (MythBuild!~MythBuild@184-106-209-209.static.cloud-ips.com) has joined #mythtv
[22:01:32] MythBuild (MythBuild!~MythBuild@184-106-209-209.static.cloud-ips.com) has quit (Client Quit)
[22:01:36] MythBuild (MythBuild!~MythBuild@184-106-209-209.static.cloud-ips.com) has joined #mythtv
[22:05:22] MythBuild (MythBuild!~MythBuild@184-106-209-209.static.cloud-ips.com) has quit (Client Quit)
[22:05:28] MythBuild (MythBuild!~MythBuild@184-106-209-209.static.cloud-ips.com) has joined #mythtv
[22:06:27] Beirdo: OK. If I did that right, we should get doxygen run every 2h on master, then rsynced to http://code.mythtv.org/doxygen/
[22:06:38] Beirdo: first run should be in about 9min
[22:17:37] MythBuild (MythBuild!~MythBuild@184-106-209-209.static.cloud-ips.com) has quit (Quit: buildmaster reconfigured: bot disconnecting)
[22:17:42] MythBuild (MythBuild!~MythBuild@184-106-209-209.static.cloud-ips.com) has joined #mythtv
[22:22:30] MythBuild (MythBuild!~MythBuild@184-106-209-209.static.cloud-ips.com) has quit (Client Quit)
[22:22:35] MythBuild (MythBuild!~MythBuild@184-106-209-209.static.cloud-ips.com) has joined #mythtv
[22:22:59] Beirdo: does help if I actually put the scheduler into the array of schedulers
[22:24:18] Kunalagon1 (Kunalagon1!~Kunalagon@212.200.241.42) has quit (Quit: Leaving.)
[22:26:15] Steve_Goodey (Steve_Goodey!~steve@host86-145-233-30.range86-145.btcentralplus.com) has quit (Remote host closed the connection)
[22:30:11] Beirdo: I think I'll put it to every 6h (but only if there are changes)
[22:31:59] dougt_ (dougt_!~dougt@c-24-130-158-216.hsd1.ca.comcast.net) has joined #mythtv
[22:32:49] dekarl (dekarl!~deKarl@e180158177.adsl.alicedsl.de) has left #mythtv ()
[22:33:37] dougt_ (dougt_!~dougt@c-24-130-158-216.hsd1.ca.comcast.net) has quit (Client Quit)
[22:36:45] MythBuild (MythBuild!~MythBuild@184-106-209-209.static.cloud-ips.com) has quit (Quit: buildmaster reconfigured: bot disconnecting)
[22:36:51] MythBuild (MythBuild!~MythBuild@184-106-209-209.static.cloud-ips.com) has joined #mythtv
[22:38:39] stuartm: existing mirrors are only updated daily, it's not really critical that they be updated any faster
[22:39:55] bbc581 (bbc581!~bbc581@74-129-133-90.dhcp.insightbb.com) has joined #mythtv
[22:40:53] MythBuild (MythBuild!~MythBuild@184-106-209-209.static.cloud-ips.com) has quit (Client Quit)
[22:40:57] MythBuild (MythBuild!~MythBuild@184-106-209-209.static.cloud-ips.com) has joined #mythtv
[22:41:04] stuartm: no-one is going to complain that the code documentation is 23 hours old :)
[22:41:54] stuarta: it's far more likely that the comments which are used as the source of the documentation are out of date by 12months or more.
[22:42:23] stuartm: yup, quite a lot of the documentation is no longer accurate
[22:42:45] ** stuarta thinks of creative ways to un-bork mac build **
[22:43:36] stuartm: referring parameters which don't exist any more, behaviour which has changed or code that is dead (but not marked as such)
[22:45:43] wagnerrp: Beirdo: seems i dont have permission to access /doxygen/ on this server
[22:48:24] stuartm: it's exclusive
[22:49:57] Beirdo: you will have to wait until the rsync is done :)
[22:50:13] Beirdo: at least until the index.html gets there
[22:51:26] Beirdo: stuartm: good point. We can run it daily. :) I have it using the "Nightly" schedule, which lets you set in much like cron
[22:51:59] Beirdo: I want to see it run a few times with a 2h interval, then it can go to 12h or 24h or something
[22:55:45] stuarta: well renaming MediaType to MythMediaType seems to do the trick
[22:56:18] Beirdo: fun borked builds :)
[22:57:55] stuarta: haven't even had a chance to start coding up the changes i need todo to osx-packager
[22:58:11] stuarta: let alone merge Use-Git back to master
[23:00:21] Beirdo: aye, fair enough
[23:00:51] stuarta: it's depressing
[23:00:58] stuarta: set aside tonight for all that
[23:01:33] stuarta: and end up fixing crap at work as well as making cupboards safe for kiddie
[23:01:38] stuarta: don't mind the last part
[23:01:54] Beirdo: the former... is just annoying. I had that too today
[23:02:10] stuarta: but when you sit down with 30m left on a 3hr hacking session it seems a bit pointless
[23:03:55] Beirdo: hmmm, seems we'll have a perms problem, need another step here :)
[23:15:29] stuartm: stuarta: if we can figure out why it was able to build without conflict when mythmedia.h was in libmyth that might lead to a nicer solution
[23:18:13] stuarta: yeah, agreed
[23:18:34] stuarta: i'll go back to pristine sources and grab the full error message
[23:23:20] jya: Beirdo: is it possible to do cherry-picking but on one file only from a particular commit?
[23:24:35] stuarta: bah this push is going to be noisy
[23:25:26] stuartm: stuarta: a MythTV namespace based solution might be appropriate if we can find a good place to stick a global 'using' declaration – that would fix it with just 3 loc
[23:25:30] stuarta: it's got 25 commits to merge in
[23:26:06] stuarta: heh, i briefly tried that only for it to bitch about MythTV::QObject not being known
[23:26:08] stuartm: I'm not sure what our policy is on namespaces, since they are infrequently used in the code
[23:26:40] stuartm: stuarta: heh :/
[23:27:31] Beirdo: jya: no, cherry pick pulls commits.
[23:27:57] stuartm: elmojo, markk_: http://mythtv.pastebin.com/sGspYVSG << After 15 minutes I think we've waited long enough and we should have exited playback?
[23:28:02] Beirdo: you CAN cherry-pick the full commit, then git reset HEAD^, and then commit only the part you wanted
[23:28:05] jya: Beirdo: see, we need to go back to svn, we can do that with svn :)
[23:28:19] Beirdo: whatever
[23:28:39] stuarta: you can do that with git too
[23:28:48] jya: Beirdo: how cna you do a cherry-pick followed by a reset, doesn't that remove all the changes?
[23:28:49] stuarta: git diff a..b <file>
[23:29:11] Beirdo: jya, it uncommits them, but they will still be in your working copy
[23:29:15] jya: stuarta: but then you loose the merging tracking.. So a later merge will create a conflict
[23:29:16] Beirdo: unless you do reset --hard
[23:29:40] stuarta: never had any merge tracking with svn, so no loss there :-p
[23:29:50] jya: Beirdo: I've been using cherry-pick – n for that
[23:29:58] Beirdo: and it is smart enough to see that a partial merge happened, only minor conflict at most
[23:29:58] jya: it does the cherry-pick, but doesn't commit
[23:30:04] Beirdo: ah
[23:30:15] Beirdo: well that's pretty much the same thing :)
[23:30:31] Beirdo: -n is likely less error-prone
[23:31:12] stuartm: stuarta: for the last few months of svn we did actually have merge tracking
[23:31:45] stuartm: never got the chance to give it a workout though
[23:31:52] stuarta: and i'm sure that there's a way that git will track what jya wants
[23:31:54] jya: Beirdo: where is it smart enough to detect a partial merge? using cherry-pick or patch & diff
[23:32:10] Beirdo: I need to go buy another power bar so I can get the dev box up again
[23:32:12] stuarta: have you tried a merge on the file?
[23:32:47] jya: stuartm: when I did a partial merge from a file only, it created hundred of mods by removing all the svn-merge-info file .. absolute mess
[23:33:11] stuartm: Beirdo: for a moment there I was wondering what an energy rich candy bar had to do with your dev box
[23:33:29] stuarta: you have to feed the mice powering it something
[23:34:00] stuartm: jya: sounds a lot like git
[23:34:09] Beirdo: ehhehe
[23:35:14] jya: think I'm just going to manually edit the file and deal with a future conflict when required
[23:35:28] stuartm: btw, when we get an svn repo up and running again, I assume we'll go with the most recent version possible?
[23:35:34] Beirdo: s/when/if/
[23:35:46] Beirdo: that would be the plan
[23:35:59] Beirdo: and the line there was to jya, not you, stuartm :)
[23:37:48] h7251 (h7251!~h725@62-47-234-54.adsl.highway.telekom.at) has joined #mythtv
[23:38:17] jya: Beirdo: you can run a merge on a file only with a git command ?
[23:38:55] Beirdo: probably
[23:38:58] Beirdo: I've never tried
[23:39:08] Beirdo: man git-merge (that would be my first guess)
[23:39:50] Beirdo: wow, this first rsync is sure taking forever :)
[23:40:15] jya: Beirdo : http://stackoverflow.com/questions/449541/how . . . th-git-merge , similar to what you suggest
[23:43:40] jya: or git checkout source_branch <path>
[23:48:41] Beirdo: OK, doxygen output should be all there
[23:51:03] Beirdo: I manually fixed perms and finished the rsync
[23:51:13] Beirdo: it should be good to go
[23:52:18] Beirdo: I'll be back in a little over an hour or so, it should have run an resynced by about then
[23:55:26] natanojl (natanojl!~jonatan@c83-252-237-63.bredband.comhem.se) has quit (Ping timeout: 276 seconds)
[23:56:54] stuarta: all i want is a bloody pastebin that works

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