Thursday, January 29th, 2015, 00:02 UTC | ||
[00:02:29] | joki (joki!~joki@p5486189D.dip0.t-ipconnect.de) has joined #mythtv | |
[00:41:45] | TheCrasher (TheCrasher!~TheCrashe@p5DCE4E8C.dip0.t-ipconnect.de) has joined #mythtv | |
[01:07:54] | TheCrasher (TheCrasher!~TheCrashe@p5DCE4E8C.dip0.t-ipconnect.de) has quit (Quit: KVIrc 4.3.1 Aria http://www.kvirc.net/) | |
[01:41:40] | knightr_ (knightr_!~Nicolas@69-165-170-178.dsl.teksavvy.com) has quit (Read error: Connection reset by peer) | |
[01:42:05] | knightr (knightr!~Nicolas@69-165-170-178.dsl.teksavvy.com) has joined #mythtv | |
[01:42:05] | knightr (knightr!~Nicolas@69-165-170-178.dsl.teksavvy.com) has quit (Changing host) | |
[01:42:05] | knightr (knightr!~Nicolas@mythtv/developer/knightr) has joined #mythtv | |
[02:28:18] | peper03 (peper03!~peper03@mythtv/developer/peper03) has quit (Ping timeout: 246 seconds) | |
[02:28:33] | peper03 (peper03!~peper03@mythtv/developer/peper03) has joined #mythtv | |
[02:28:46] | sheedy-away is now known as sheedy | |
[02:47:23] | rich0_ (rich0_!~quassel@gentoo/developer/rich0) has joined #mythtv | |
[02:50:52] | rich0 (rich0!~quassel@gentoo/developer/rich0) has quit (Ping timeout: 240 seconds) | |
[03:00:56] | sheedy is now known as sheedy-away | |
[03:19:37] | gregL (gregL!~greg@cpe-74-76-121-109.nycap.res.rr.com) has quit (Ping timeout: 252 seconds) | |
[03:32:44] | gregL (gregL!~greg@cpe-74-76-121-109.nycap.res.rr.com) has joined #mythtv | |
[04:43:13] | fetzerch_ (fetzerch_!~quassel@unaffiliated/fetzerch) has joined #mythtv | |
[04:46:54] | fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has quit (Ping timeout: 246 seconds) | |
[05:25:17] | Sharky112065 (Sharky112065!~Sharky112@24.16.144.222) has quit (Ping timeout: 245 seconds) | |
[05:26:05] | Sharky112065 (Sharky112065!~Sharky112@c-24-16-144-222.hsd1.wa.comcast.net) has joined #mythtv | |
[05:27:38] | mRMans (mRMans!~mRMan@75-166-112-136.hlrn.qwest.net) has quit (Ping timeout: 264 seconds) | |
[06:14:48] | Tobbe5178 (Tobbe5178!~asdf@2001:2002:d9d4:ce7e:f5b2:8ffd:cfd0:1cd5) has joined #mythtv | |
[07:09:02] | FabriceMG (FabriceMG!~Thunderbi@LCaen-656-1-100-212.w80-11.abo.wanadoo.fr) has joined #mythtv | |
[08:13:53] | Roklobsta (Roklobsta!~Roklobsta@ppp118-209-232-217.lns20.mel8.internode.on.net) has joined #mythtv | |
[09:17:09] | stuarta: | stuartm: not noticed services returning 404's. but then i've not been watching |
[09:19:40] | stuarta: | hmmm, yes bunch of 404's, so far only on themes.zip for 0.27.[23] which we don't have |
[09:22:51] | dekarl (dekarl!~dekarl@mythtv/developer/dekarl) has quit (Ping timeout: 264 seconds) | |
[09:25:44] | stuarta: | stuartm: since the themes fallback code went in, you see a given ip address check for 0.27, 0.27.4, 0.27.3, 0.27.2 then 0.27.1 and it'll 404 on .2 & .3 |
[09:36:57] | dekarl (dekarl!~dekarl@mythtv/developer/dekarl) has joined #mythtv | |
[09:55:18] | dekarl: | this http caching thingy isn't for the faint of heart... http://paste.ubuntu.com/9933871/ I wonder what went wrong there (that's one instance of a repeating pattern) |
[09:57:15] | stuarta: | dekarl: what's that trying to illustrate? |
[09:57:22] | ** stuarta is not scared by such things ** | |
[09:58:22] | stuarta: | apart from it hammering the backend |
[09:59:19] | lomion0815 (lomion0815!5bd9374e@gateway/web/freenode/ip.91.217.55.78) has joined #mythtv | |
[10:00:08] | aberrios_ (aberrios_!~aberrios@195.130.201.200) has joined #mythtv | |
[10:02:20] | lomion0815: | Hi, are you aware that while the setting "Enable realtime priority threads" still exists the code in main.cpp was removed in 0.25!? |
[10:03:45] | dekarl: | stuarta, some thing in Kodi is requesting my channel list. The thing is that Kodi and no Kodi plugin handle that format. Yet its being requested over and over, triggered by what appears to be one user action. |
[10:03:57] | stuarta: | dekarl: interesting |
[10:04:17] | stuarta: | lomion0815: haha, good spot. can you raise a ticket in trac so we don't forget ;-) |
[10:04:20] | stuarta: | +again |
[10:04:22] | dekarl: | There appears to be a plugin that takes a xmltv file. But to get that you'd have to run the proper grabber |
[10:06:19] | peper03 (peper03!~peper03@mythtv/developer/peper03) has quit (Ping timeout: 265 seconds) | |
[10:06:57] | peper03 (peper03!~peper03@mythtv/developer/peper03) has joined #mythtv | |
[10:09:04] | dekarl: | Also its not scaring. I wondered if its a endian bug in their http caching implementation and if I could help them identify and fix it :) |
[10:10:35] | dekarl: | lomion0815: nice catch. I was thinking about adding some "enable io niceness for lossless cutting" setting lately :) |
[10:11:11] | stuartm: | stuarta: WebFRontend was configured to use https for the channel icon service, but services wasn't configured for https |
[10:12:01] | stuarta: | hmm, never noticed that |
[10:12:02] | stuartm: | so it was 404ing, enabled https for services last night, with a lazy copy/paste but it appears to be working without breaking anything else |
[10:12:22] | stuarta: | commited to the repo??? |
[10:12:44] | stuartm: | no, this is why I shouldn't do these things late at night |
[10:13:46] | stuarta: | :) |
[10:13:51] | stuartm: | it was somehow pointing at code.mythtv.org for ssl before, so the errors were in the wrong log |
[10:14:09] | stuarta: | iirc, code is the default site |
[10:14:50] | stuarta: | so anything not explicitly captured goes to it |
[10:16:03] | ** stuarta goes to watch for other oddities ** | |
[10:18:19] | stuartm: | ok, should be in repo |
[10:18:35] | stuarta: | ta |
[10:20:10] | stuartm: | I want to increase reuse between the configs, but no doubt that's something best left until they are ported to nginx |
[10:23:52] | Merlin83b (Merlin83b!~Daniel@2a00:1ee0:3:1337:4cab:3f41:f080:6a18) has joined #mythtv | |
[10:25:34] | stuarta: | stuartm: yeah, doing that as I go along |
[10:40:16] | stuarta: | stuff like all the ssl config is in 1 file |
[10:45:18] | stuarta: | that reminds me, alcor needs a reboot. doing it now |
[10:52:35] | lomion0815: | I raised the ticket #12362 |
[10:52:35] | ** MythLogBot http://code.mythtv.org/trac/ticket/12362 ** | |
[10:53:54] | stuarta: | thanks |
[11:17:13] | rich0 (rich0!~quassel@gentoo/developer/rich0) has joined #mythtv | |
[11:19:40] | rich0_ (rich0_!~quassel@gentoo/developer/rich0) has quit (Ping timeout: 244 seconds) | |
[11:49:20] | lomion0815: | dekarl: You mentioned io niceness for lossless cut. Did you ever try to run mythfronend with higher IO niceness? I get spradic video buffer underruns in playback. |
[11:50:45] | lomion0815: | I think its IO related since it always happens when recodings end although my database is on a different disc than my recordings |
[11:52:39] | xris (xris!~xris@mythtv/developer/xris) has quit (Ping timeout: 265 seconds) | |
[11:54:02] | Warped (Warped!~Warped@108.85.160.119) has quit (Ping timeout: 264 seconds) | |
[11:56:17] | xris (xris!~xris@xris.forevermore.net) has joined #mythtv | |
[11:59:26] | Warped (Warped!~Warped@108.85.160.119) has joined #mythtv | |
[12:09:27] | Warped (Warped!~Warped@108.85.160.119) has quit (Ping timeout: 252 seconds) | |
[12:11:14] | Hydroponx (Hydroponx!~hydr@71-12-104-200.dhcp.mtgm.al.charter.com) has quit (Ping timeout: 272 seconds) | |
[12:11:29] | stuarta: | lomion0815: pull up the video playback data and check your buffer fill |
[12:14:39] | Hydr0p0nX (Hydr0p0nX!~hydr@71-12-104-200.dhcp.mtgm.al.charter.com) has joined #mythtv | |
[12:18:24] | Warped (Warped!~Warped@108.85.160.119) has joined #mythtv | |
[12:58:32] | stuartm: | snow started falling 10 minutes ago, already an inch on the ground ... :/ |
[12:58:59] | stuartm: | can't remember the last time I saw it falling so heavily |
[12:59:05] | lomion0815: | How to check the buffer fill. I can see the process buffers in the OSD but can I log it? What do mean with pull up the video playback data |
[12:59:06] | Roklobsta (Roklobsta!~Roklobsta@ppp118-209-232-217.lns20.mel8.internode.on.net) has quit (Read error: Connection reset by peer) | |
[12:59:23] | Sharky112065 (Sharky112065!~Sharky112@c-24-16-144-222.hsd1.wa.comcast.net) has quit (Ping timeout: 244 seconds) | |
[12:59:51] | lomion0815: | s/do mean/do you mean/ |
[13:04:20] | stuartm: | it's logged with -v playback iirc, but you can watch it in realtime in the OSD by pressing DEBUGOSD |
[13:08:25] | lomion0815: | I don't think that -v playback logs the buffer fill. |
[13:09:02] | lomion0815: | or I did not recognize it as buffer fill |
[13:20:20] | brfranse_ is now known as brfransen | |
[13:29:50] | stuarta: | stuartm: cool, we are supposed to get some down here too, unlikely that heavy tho |
[13:32:27] | stuartm: | it's stopped now, and is melting fast, guess it wasn't to be |
[13:33:16] | stuartm: | lomion0815: I might be wrong, or you might need --loglevel=debug |
[13:39:30] | stuarta: | lomion0815: watch it in real time, hit 'm', playback, playback data |
[13:39:55] | stuarta: | lomion0815: it's listed as "available buffer" but it really means "buffer fill" |
[13:40:09] | stuarta: | we really should fix that |
[13:40:23] | lomion0815: | I'll give it try this evening |
[13:42:19] | stuarta: | ug, that's a translatable string, and the incorrect name is used as the translation key, which means *all* the translations are also wrong |
[13:44:16] | lomion0815: | I see all 43 buffers filled (I increased the process buffers already) but I won't watch the figures for hours until the underrun occurs ... I need it logged :-) |
[13:44:48] | stuarta: | the buffer figure i'm talking about should stick at around 99% |
[13:44:57] | stuarta: | 99% of 8Mb, or 16Mb |
[13:48:33] | lomion0815: | Everytime I looked these were at 99% |
[13:50:39] | stuarta: | okay, that's good |
[14:19:32] | lomion0815 (lomion0815!5bd9374e@gateway/web/freenode/ip.91.217.55.78) has left #mythtv () | |
[14:28:48] | sl1ce_1g (sl1ce_1g!~johnathan@pool-72-74-164-209.bstnma.fios.verizon.net) has joined #mythtv | |
[14:29:17] | sl1ce (sl1ce!~johnathan@pool-72-74-164-209.bstnma.fios.verizon.net) has quit (Remote host closed the connection) | |
[14:30:27] | Jordack (Jordack!~Jordack@75-151-31-172-Michigan.hfc.comcastbusiness.net) has joined #mythtv | |
[14:54:44] | stuartm: | wagnerrp: I'm making a change to the UPDATE_FILE_SIZE event, switching it to use recordedid instead of chanid/starttime, it appears the python bindings listen for this event but I wouldn't know where to start with fixes |
[14:58:51] | dekarl-work (dekarl-work!51c8c678@gateway/web/freenode/ip.81.200.198.120) has joined #mythtv | |
[14:59:11] | dekarl-work: | lomion0815: what good is ionice for the mythfrontend? Its not such a heavy IO hog to start with :D |
[15:00:00] | dekarl-work: | ah, looks like he left |
[15:01:49] | stuarta: | stuartm: are you aware of any issues relating to loading the 3rdParty modules via the backend webserver? |
[15:02:58] | stuarta: | jquery is 404'ing for me, i can dig some more later, but just wanted to know if it's something you've seen |
[15:05:55] | stuarta: | hmm, so we are referencing jquery-2.0.3.min.js and we install jquery-2.0.3.js |
[15:15:21] | stuartm: | stuarta: ah, my fault, I switched them but forgot to add the min version to the commit |
[15:15:25] | jheizer (jheizer!~jheizer@2601:d:3a00:1d58:853f:a095:7171:2836) has quit (Read error: Connection reset by peer) | |
[15:16:48] | stuarta: | :) |
[15:17:24] | stuartm: | fixed |
[15:18:32] | stuartm: | it was easier early on to use the version without all the whitespace removed as it made debugging easier |
[15:20:07] | stuartm: | plus it seemed pointless since we're gzip compressing all http traffic anyway, but after some research it seems that compression + white space stripping does result in a further size decrease and therefore minutely faster load times |
[15:27:09] | dekarl-work: | should we run some tools over all the .css/.js sources to convert them into a format that is easier to digest for the browsers (aka faster to load)? Maybe as part of the build? |
[15:27:20] | stuarta: | stuartm: thanks, works now |
[15:27:45] | dekarl-work: | s/digest/digest/ |
[15:27:56] | stuarta: | dekarl-work: i'm not convinced that will save much |
[15:28:38] | dekarl-work: | stuarta, then what is the point in replacing jquery-2.0.3.js with jquery-2.0.3.min.js ? |
[15:29:53] | dekarl-work: | stuarta, fwiw I think that is something further down the road. For now "easy for humans to handle" appears to be far more important :) |
[15:30:03] | dekarl-work: | aka debugability |
[15:30:44] | stuarta: | i refer to you stuartm's earlier comment |
[15:30:48] | stuarta: | biab, school run |
[15:31:28] | jheizer (jheizer!~jheizer@2601:d:3a00:1d58:219c:2752:2b46:78f0) has joined #mythtv | |
[15:32:11] | stuartm: | dekarl-work: jquery is huge by comparison to all the other files, and since it's third party code debugging it isn't often necessary |
[15:32:49] | stuartm: | most of my early debugging in fact was just me learning how certain things were supposed to work by reading the code instead of reading the docs ;) |
[15:41:16] | stuartm: | nope, not going to touch the python bindings, I think I'd just make things worse :| |
[16:10:43] | stuarta: | sleet != snow :( |
[16:10:51] | ** stuarta demands proper snow ** | |
[16:12:01] | dekarl-work: | "due to the wrong kind of snow ..." ;) |
[16:12:44] | stuarta: | ooo, it's trying to snow now |
[16:14:03] | FabriceMG (FabriceMG!~Thunderbi@LCaen-656-1-100-212.w80-11.abo.wanadoo.fr) has quit (Quit: FabriceMG) | |
[16:22:34] | stuarta: | that's the most disappointing snow storm i've ever seen, 2 minutes of tiny wafty flakes and it's stopped :( |
[16:34:38] | stuarta: | hmmm, need to add an channel icon not found icon for the backend webserver. stop a whole load of 404's |
[16:36:07] | stuarta: | maybe one of those circles with a slash through it? |
[16:48:27] | stuartm: | sure we can come up with something prettier |
[16:48:55] | stuarta: | i'm going for the "what can i download for free with an appropriate license" |
[16:49:01] | stuarta: | plenty to choose from |
[16:52:15] | jheizer: | or maybe just a single transparent pixel if you are ok with it staying blank |
[16:54:00] | stuarta: | hmm, a question mark might be more appropriate |
[16:54:17] | ** stuarta is currently perusing http://www.iconspedia.com/ ** | |
[16:59:29] | dekarl-work: | might simply not reference a non-existing image? |
[17:02:22] | stuarta: | dekarl-work: that's the idea. if(not_found) return generic not_found image |
[17:02:38] | stuarta: | rather than issuing http 404 response |
[17:04:26] | dekarl-work: | stuarta, I was thinking in terms of "if(found) print <img tag>" |
[17:04:45] | stuarta: | if it's found, it returns the proper channel icon |
[17:04:50] | stuarta: | that bit works fine |
[17:05:40] | dekarl-work: | yes, its the "speculative reference" that makes me wonder |
[17:05:44] | stuartm: | dekarl-work: this is where the database contains a icon filename, but that file isn't present on the backend |
[17:05:59] | dekarl-work: | stuartm: ty, I was thinking of "the field is empty" |
[17:06:00] | stuarta: | you sure? |
[17:06:06] | stuartm: | at which point GetChannelIcon will always return a 404 |
[17:06:06] | stuarta: | stuartm: you sure? |
[17:06:18] | dekarl-work: | lets convert channel itcons to storage groups |
[17:06:24] | stuartm: | dekarl-work: already did |
[17:06:35] | dekarl-work: | stuartm, only new or also exiting icons? |
[17:06:40] | stuartm: | stuarta: no, not now you've asked that :) |
[17:06:48] | stuartm: | dekarl-work: can't remember |
[17:06:59] | stuartm: | was done too long ago |
[17:07:09] | dekarl-work: | users are sharing guides on doing the migration manually to unbreak kodi and stuff |
[17:07:29] | dekarl-work: | it will get easier once we got rid of mythtv-setup... |
[17:07:52] | stuarta: | my wishlist include background icon downloader |
[17:08:11] | dekarl-work: | running mythtv-setup as a different user then the backend makes for funny storage locations and permissions (hi mythbuntu packaging :) |
[17:08:12] | Hydroponx (Hydroponx!~hydr@71-12-104-200.dhcp.mtgm.al.charter.com) has joined #mythtv | |
[17:08:12] | Hydr0p0nX (Hydr0p0nX!~hydr@71-12-104-200.dhcp.mtgm.al.charter.com) has quit (Ping timeout: 245 seconds) | |
[17:08:33] | ** stuarta facepalms ** | |
[17:08:40] | dekarl-work: | btw, I already looked into it, but did not find an easy way to fix it |
[17:10:23] | dekarl-work: | xmltv setup suffers from it, too |
[17:12:38] | stuartm: | stuarta: only 404s I noticed last night were for channels where there was an icon defined but it wasn't actually present on the system, but since all of the setup stuff is client side using horribly obscure jQuery widgets I'm not sure what it's doing, just can't follow it |
[17:12:59] | stuartm: | why the scripting API wasn't used I'll never know :/ |
[17:13:01] | stuarta: | stuartm: this is in the tv guide |
[17:13:13] | stuartm: | stuarta: oh, well that's entirely different then :[ |
[17:13:36] | stuarta: | yeah, i've no icons for my dev backend, so it looks crap |
[17:14:24] | stuartm: | ok, that's my fault then, and easily fixable |
[17:14:52] | stuarta: | so we are calling '/Guide/GetChannelIcon?ChanId=3066&Height=57' and getting a 404 back |
[17:15:05] | stuartm: | just need an alternative icon, or maybe not show anything at all? |
[17:15:06] | stuarta: | rinse and repeat for all channels without icon |
[17:15:42] | stuarta: | stuartm: jheizer's idea of a 1x1 pixel could work, or something that kinda indicates it's missing |
[17:15:55] | stuarta: | i don't really mind, i just hate 404's ;-) |
[17:16:24] | stuarta: | now the question is do you then 301 to a not found icon? |
[17:17:09] | stuarta: | or just return the fake icon each time |
[17:17:27] | stuartm: | stuarta: no, I'll do it in the script which creates the html, rather than in the services API, otherwise third party clients won't be able to pick the icon they want to use when none is available |
[17:17:56] | stuarta: | nice |
[17:18:44] | stuartm: | i.e. the stated behaviour of the services API for files is that if they aren't available you get a 404 back, not we'll send you an alternative file which might do instead |
[17:19:21] | stuarta: | see this is where we are heading into me not knowing how it hooks together |
[17:20:58] | stuarta: | what you are saying is correct, i agree. |
[17:25:47] | stuartm: | it would be clearer if we'd added a prefix for services API stuff, e.g. /API/Guide |
[17:26:43] | stuarta: | bah too many services api's |
[17:26:52] | stuarta: | one internal, one on services.mythtv.org |
[17:27:17] | stuarta: | and i will be adding a prefix to the website, even if only internally |
[17:29:11] | stuartm: | stuarta: and the internal one has two interfaces, the http one (returning json or xml) and a script one where scripts make direct calls to methods |
[17:29:34] | stuartm: | so yes, I can see how that could get confusing |
[17:29:48] | stuarta: | now my head hurts |
[17:30:52] | stuartm: | the WebFrontend largely uses the script api, the scripts are interpreted by the backend itself – very fast, no network overhead, utilises the abundant cpu cycles of the backend to do the heavy lifting |
[17:31:42] | stuartm: | but for stuff like icons and other images loaded by the browser, it's simpler just to point at the http API |
[17:31:50] | stuarta: | aye |
[17:32:47] | stuarta: | stuartm: can I add a feature request to your work? define the default websocket port as backend_port + offset |
[17:33:03] | stuartm: | stuarta: sure |
[17:33:37] | stuarta: | i may be unique, but the dev backend_port is 7543 so everything is after that, and prod is 6543 |
[17:33:54] | stuarta: | thus co-existence is peaceful |
[17:34:12] | stuartm: | I was originally think it would use the backend port, that's still my ultimate goal, but it's just simpler in the initial implementation to pick a separate port |
[17:34:41] | stuarta: | yeah, well if the former doesn't come off, then my feature request stands |
[17:34:55] | stuarta: | if it does, my request is irrelevant |
[17:34:57] | stuarta: | :) |
[17:35:57] | stuartm: | forgotten what I did for the ssl port |
[17:36:09] | stuartm: | whether that was a fixed offset, or configurable |
[17:36:36] | stuarta: | iirc from the commits, it's configurable, but has a default |
[17:36:56] | sheedy-away is now known as sheedy | |
[17:46:50] | Merlin83b (Merlin83b!~Daniel@2a00:1ee0:3:1337:4cab:3f41:f080:6a18) has quit (Quit: Leaving) | |
[17:48:00] | mRMans (mRMans!~mRMan@75-166-112-136.hlrn.qwest.net) has joined #mythtv | |
[17:51:40] | ** stuarta heads off for a while ** | |
[17:56:08] | cecil_ (cecil_!~cesman@pool-98-119-153-47.lsanca.fios.verizon.net) has quit (Quit: Konversation terminated!) | |
[17:57:42] | dekarl-work (dekarl-work!51c8c678@gateway/web/freenode/ip.81.200.198.120) has quit () | |
[18:19:11] | tgm4883: | dekarl: how are you running it as a different user? |
[18:20:53] | tgm4883: | dekarl: or, what are the "funny storage group directories" |
[18:27:35] | pthomas (pthomas!~chatzilla@207.102.77.253) has joined #mythtv | |
[18:51:57] | pthomas (pthomas!~chatzilla@207.102.77.253) has quit (Quit: ChatZilla 0.9.91.1 [Firefox 34.0.5/20141126041045]) | |
[18:52:08] | caelor_ (caelor_!~quassel@cpc72389-sotn14-2-0-cust277.15-1.cable.virginm.net) has quit (Ping timeout: 246 seconds) | |
[18:52:50] | pthomas (pthomas!~chatzilla@207.102.77.253) has joined #mythtv | |
[18:57:26] | caelor (caelor!~quassel@cpc72389-sotn14-2-0-cust277.15-1.cable.virginm.net) has joined #mythtv | |
[19:00:16] | sheedy is now known as sheedy-away | |
[19:17:06] | markus_ (markus_!~markus@194-166-28-225.adsl.highway.telekom.at) has joined #mythtv | |
[19:17:06] | markus_ (markus_!~markus@194-166-28-225.adsl.highway.telekom.at) has quit (Client Quit) | |
[19:43:56] | dekarl-work (dekarl-work!51c8c678@gateway/web/freenode/ip.81.200.198.120) has joined #mythtv | |
[19:45:19] | dekarl-work: | tgm4883: when I run mythtv-setup, e.g. from the command line, it is being run as the logged in user. But when I start the backend service it is run as the dedicated mythtv user |
[19:45:51] | tgm4883: | dekarl-work: doesn't the SG stuff get pulled from the DB? |
[19:45:58] | dekarl-work: | so the configuration/icons end up in ~frontenduser/.mythtv and ~backenduser/.mythtv depending on how they are created |
[19:46:18] | dekarl-work: | What SG stufF? |
[19:46:26] | tgm4883: | storage groups |
[19:46:30] | dekarl-work: | the SG is defined as $USER/.mythtv :) |
[19:46:36] | aberrios_ (aberrios_!~aberrios@195.130.201.200) has quit (Ping timeout: 272 seconds) | |
[19:47:12] | tgm4883: | dekarl-work: I'm not sure why it would ever be that, unless mythtv defaults to that at some point |
[19:47:36] | dekarl-work: | let me look up the real value (yes its a default / embedded SG) |
[19:49:28] | zentec (zentec!~zentec@68-188-175-3.dhcp.bycy.mi.charter.com) has joined #mythtv | |
[19:50:35] | dekarl-work: | here is the ~/.mythtv part https://github.com/MythTV/mythtv/blob/master/ . . . mythdirs.cpp (could use some .empty()) |
[19:51:28] | Casper82 (Casper82!~Casper@pool-108-52-47-183.phlapa.fios.verizon.net) has joined #mythtv | |
[19:51:55] | dekarl-work: | and here is the ChannelIcons StorageGroup https://github.com/MythTV/mythtv/blob/master/ . . . roup.cpp#L62 |
[19:54:44] | dekarl-work: | and here is the static location we write the icons to which happens to be the same https://github.com/MythTV/mythtv/blob/master/ . . . ons.cpp#L271 |
[19:55:59] | GreyFoxx_ (GreyFoxx_!~greg@out.of.phaze.org) has joined #mythtv | |
[19:56:06] | sphery_ (sphery_!~mdean@mythtv/developer/sphery) has joined #mythtv | |
[19:56:36] | zentec_ (zentec_!~zentec@68-188-175-3.dhcp.bycy.mi.charter.com) has quit (Ping timeout: 264 seconds) | |
[19:56:37] | Casper0082 (Casper0082!~Casper@pool-108-52-47-183.phlapa.fios.verizon.net) has quit (Ping timeout: 264 seconds) | |
[19:56:38] | GreyFoxx (GreyFoxx!~greg@out.of.phaze.org) has quit (Ping timeout: 264 seconds) | |
[19:56:38] | sphery (sphery!~mdean@mythtv/developer/sphery) has quit (Ping timeout: 264 seconds) | |
[19:56:52] | stuartm: | stuarta: did you find something suitable icon-wise? Or should I just go with a question mark? |
[19:58:57] | dekarl-work: | tgm4883: later we store the filename without path into the database https://github.com/MythTV/mythtv/blob/master/ . . . ons.cpp#L563 (ignoring the fact that your icons better be named with unique file names for now) |
[19:59:44] | dekarl-work: | the easy fix is to symling the logged in users' .mythtv directory to the mythtv users' .mythtv directory and grant liberal access permissions |
[19:59:51] | dekarl-work: | s/symling/symlink/ |
[20:01:39] | tgm4883: | dekarl-work: I'm still not 100% sure why you got those directories. Was the user you were running it as in the mythtv group? |
[20:01:55] | dekarl-work: | tgm4883: mythfilldatabase on the other hand stores the absolute filename https://github.com/MythTV/mythtv/blob/master/ . . . ata.cpp#L164 |
[20:03:02] | dekarl-work: | tgm4883: I have not verified that, whatever was the default in mythbuntu 12.04. But I'm not sure if the group membership has anything to do with the MYTHCONFDIR |
[20:04:39] | dekarl-work: | it may help with permissions, but still result in separate directories |
[20:04:57] | tgm4883: | dekarl-work: is this on new packaging, or just a standard ISO install? |
[20:06:26] | Roklobsta (Roklobsta!~Roklobsta@ppp118-209-232-217.lns20.mel8.internode.on.net) has joined #mythtv | |
[20:07:11] | dekarl-work: | tgm4883: I'm, not sure I understand the question. It has been like that as long as I remember. The backend was installed from a MythBuntu DVD |
[20:07:36] | tgm4883: | dekarl-work: ah, so this is on a pre-existing backend |
[20:08:00] | dekarl-work: | tgm4883: has something changed in that area lately? |
[20:08:18] | dekarl-work: | I can dist-upgrade to 14.04 if that helps |
[20:08:31] | tgm4883: | dekarl-work: well we don't ship a DB anymore and just let mythtv make one, but that wouldn't affect a preexisting DB |
[20:08:51] | dekarl-work: | tgm4883: I do not think this has anything to do with the DB |
[20:09:30] | tgm4883: | dekarl-work: I'm still a little confused as to how that was working on your system. Doesn't mythtv-setup pull those storage directories from the database if they are already setup? |
[20:09:37] | dekarl-work: | I think the mythtv-setup wrapper should run mythtv-setup.real as the mythtv user, but I'm not sure how that would work |
[20:10:01] | dekarl-work: | "sudo -u mythtv mythtv-setup" ended in X permissions |
[20:10:43] | dekarl-work: | tgm4883: I do not think that these Storage Groups ever go to the database, unless you manually defined them to overwrite the defaults |
[20:11:05] | dekarl-work: | e.g. to point the http-live-streaming transcoding temp space to a location with space |
[20:11:15] | gregL (gregL!~greg@cpe-74-76-121-109.nycap.res.rr.com) has quit (Read error: Connection reset by peer) | |
[20:11:38] | tgm4883: | dekarl-work: uh, wouldn't they have to go to the database? You can query them from the services API |
[20:11:42] | dekarl-work: | so if MYTHCONFDIR will change when you switch the user id |
[20:11:58] | tgm4883: | dekarl-work: I can't reproduce it here at work |
[20:12:23] | dekarl-work: | tgm4883: me neither |
[20:12:42] | tgm4883: | Was this a one time thing? Is there a ticket somewhere on it? |
[20:12:46] | dekarl-work: | but you sound as if its only me :) |
[20:13:07] | tgm4883: | I've never heard of this issue before, I can't reproduce it, and thinking about it doesn't make sense |
[20:13:28] | dekarl-work: | I don't know how the icons get into the backend when you use schedules direct |
[20:13:34] | tgm4883: | since my understanding is that would get put into the DB. If it's not in the DB, where is it stored? |
[20:14:08] | dekarl-work: | the path is dynamically generated on every start of mythtv-setup or mythbackend |
[20:14:27] | tgm4883: | dynamically generated from what, MYTHCONFDIR? |
[20:15:10] | tgm4883: | and why would it be dynamically generated if I go into mythtv-setup and change the storage directories? |
[20:16:13] | dekarl-work: | the storage directory itself is not persisted in the database |
[20:16:44] | dekarl-work: | unless, as you correctly describe it, the user explicitely defines the ChannelIcons storage group |
[20:17:21] | dekarl-work: | yes, dynamically generated from MYTHCONFDIR, which is dynamically generated, containing $HOME :) |
[20:18:12] | dekarl-work: | So there is a default that is not persisted. You can overwrite the default and nail it to one location. If you don't it will change depending on the current user. |
[20:19:01] | tgm4883: | ok, I'm going to believe you looked at the code for that, but my question is, how does mythtv-setup know that all of my storage directories are in /var/lib/mythtv/<SG NAME> except for my recording directory which I just changed to /bobo/the/clown/ |
[20:19:14] | tgm4883: | if that is dynamically generated, then that is fricking magic |
[20:19:57] | peper03 (peper03!~peper03@mythtv/developer/peper03) has quit (Ping timeout: 276 seconds) | |
[20:20:03] | dekarl-work: | tgm4883: its the mythical convergence ;) |
[20:20:45] | peper03 (peper03!~peper03@mythtv/developer/peper03) has joined #mythtv | |
[20:21:37] | tgm4883: | dekarl-work: what is this then http://pastebin.ubuntu.com/9942631/ |
[20:22:02] | dekarl-work: | https://github.com/MythTV/packaging/blob/fixe . . . .0.sql#L2731 |
[20:23:23] | dekarl-work: | tgm4883: notice the absence of ChannelIcons |
[20:24:11] | tgm4883: | dekarl-work: yes, is it only ChannelIcons that is giving you a funny SG location? |
[20:24:58] | dekarl-work: | yes, because the xmltv configuration is not yet stored in a storage group. So it is also MYTHCONFDIR |
[20:25:15] | tgm4883: | ah ok |
[20:25:19] | tgm4883: | "dekarl-work> running mythtv-setup as a different user then the backend makes for funny storage locations and permissions (hi mythbuntu packaging :)" |
[20:25:27] | tgm4883: | that makes it sound like all of your SG's were that way |
[20:26:14] | ** tgm4883 rereads backlog with that new info ** | |
[20:26:37] | dekarl-work: | the packaging part was ment wrt the mythtv-setup.real wrapper needing some kind of addition |
[20:27:24] | tgm4883: | so then let me see if I can summarize this correctly |
[20:28:05] | dekarl-work: | the funny location is for everything under MYTHCONFDIR (as it moved around) |
[20:29:20] | tgm4883: | When running mythtv-setup as someone else than the mythtv user, channel icons get downloaded to the wrong directory because A) mythtv-setup calls an external program (thus running it as the non-mythtv user) and B) the channel icon downloader doesn't use SG's yet so it downloads to MYTHCONFDIR, which means that C) when the backend runs as mythtv, it can't |
[20:29:20] | tgm4883: | find the icons because it's looking in the wrong place |
[20:29:27] | tgm4883: | does that all sound right? |
[20:30:02] | dekarl-work: | close |
[20:30:46] | tgm4883: | the channel icon path? |
[20:31:06] | dekarl-work: | I do not remember mythtv-setup running as mythtv user on mythbuntu |
[20:31:53] | dekarl-work: | b) mythtv-setup can not use the SG via the backend when that isn't running |
[20:32:30] | tgm4883: | it wouldn't have to query the backend, it could query the DB |
[20:33:23] | tgm4883: | but the end issue is that the icon's don't end up in the mythtv users directory so the backend can't find them. Right? |
[20:39:06] | dekarl-work: | yes |
[20:40:10] | dekarl-work: | for a workaround we could store the absolute file name, so the icons get found |
[20:40:47] | dekarl-work: | but only if frontend and backend share the path outside of mythtv |
[20:41:03] | dekarl-work: | so it would not work for remote frontends |
[20:41:14] | dekarl-work: | unless you setup NFS for the icons |
[20:43:34] | dekarl-work: | tgm4883: If the mythtv-setup wrapper could switch the user that mythtv-setup.real runs under, that would be nice. Or maybe just nail the MYTHCONFDIR to ~mythtv/.mythtv. |
[20:43:46] | tgm4883: | dekarl-work: yea I'm going to email superm1 on it |
[20:44:43] | tgm4883: | I think we've not done so in the past because you would need su privs to do that |
[20:46:23] | dekarl-work: | ty, all will be well when as soon as move configuration to the backend webserver ;) (all but mythfilldatabase, but that can be rewritten to use the SG via a running backend for the configuration file and the icons.) |
[20:46:45] | dekarl-work: | s/soon/soon[tm]/ |
[20:47:02] | ** dekarl-work got to run now, see you around ** | |
[20:47:21] | dekarl-work (dekarl-work!51c8c678@gateway/web/freenode/ip.81.200.198.120) has quit (Quit: running) | |
[21:09:43] | agrajag` (agrajag`!~agrajag@pool-74-111-183-81.pitbpa.fios.verizon.net) has quit (Quit: ZNC - http://znc.in) | |
[22:01:44] | Tobbe5178 (Tobbe5178!~asdf@2001:2002:d9d4:ce7e:f5b2:8ffd:cfd0:1cd5) has quit (Read error: Connection reset by peer) | |
[22:07:58] | Roklobotomy (Roklobotomy!~Roklobsta@ppp118-209-87-48.lns20.mel4.internode.on.net) has joined #mythtv | |
[22:08:48] | Roklobotomy (Roklobotomy!~Roklobsta@ppp118-209-87-48.lns20.mel4.internode.on.net) has quit (Client Quit) | |
[22:09:03] | Roklobsta (Roklobsta!~Roklobsta@ppp118-209-232-217.lns20.mel8.internode.on.net) has quit (Ping timeout: 264 seconds) | |
[22:21:33] | MythBuild: | build #2611 of master-debian-wheezy-64bit is complete: Failure [4failed git] Build details are at http://code.mythtv.org/buildbot/builders/mast . . . /builds/2611 blamelist: Stuart Morgan <smorgan@mythtv.org > |
[22:31:05] | MythBuild: | build #5473 of master-freebsd-64bit is complete: Failure [4failed git] Build details are at http://code.mythtv.org/buildbot/builders/mast . . . /builds/5473 blamelist: Stuart Morgan <smorgan@mythtv.org > |
[22:34:10] | Jordack (Jordack!~Jordack@75-151-31-172-Michigan.hfc.comcastbusiness.net) has quit () | |
[22:36:38] | stuartm: | whaa? |
[22:36:51] | stuartm: | oh, failed git |
[23:27:05] | MythBuild: | build #2299 of master-ubuntu-12_04-lts-64bit is complete: Failure [4failed git] Build details are at http://code.mythtv.org/buildbot/builders/mast . . . /builds/2299 blamelist: Stuart Morgan <smorgan@mythtv.org > |
[23:48:20] | XDS2010_ (XDS2010_!sid1218@gateway/web/irccloud.com/x-obexpkmzrfcpabxa) has quit () | |
[23:55:37] | joki (joki!~joki@p5486189D.dip0.t-ipconnect.de) has quit (Ping timeout: 255 seconds) | |
[23:57:47] | XDS2010_ (XDS2010_!sid1218@gateway/web/irccloud.com/x-pvfucdcrghptgzyc) has joined #mythtv |
IRC Logs collected by
BeirdoBot.
Please use the above link to report any bugs.