:: #mythtv

Daily chat history

Current users (68):

alan`, aloril, amessina, Anssi, Beirdo, brfransen, caelor, cesman, ChanServ, Chutt, clever, coling, cybrNaut, dekarl, ElmerFudd, esperegu, fetzerch_, gary_buhrmaster, ghoti, gregL, GreyFoxx, Hydr0p0nX, J-e-f-f-A, jab416171, jams_, jheizer, jpabq_, jpharvey_, jst, jwhite, jya, Kevin`, kormoc, kurre2, len, moparisthebest, MythBuild, MythLogBot, nephyrin, peper03, poptix, pppingme, purserj_, rhpot1991, rich0, robink, rOOb, seld, sheedy-away, skd5aner, sl1ce_1g, sphery, sraue, stuarta, stuartm, suffice, superm1, taylorr, tgm4883, toeb, tonsofpcs, tris, unforgiven512, wagnerrp, XDS2010_, xris, zentec, _charly_
Thursday, February 12th, 2015, 00:00 UTC
[00:00:32] joki (joki! has quit (Ping timeout: 265 seconds)
[00:05:12] joki (joki! has joined #mythtv
[01:01:47] arescorpio (arescorpio! has joined #mythtv
[01:10:42] Hydr0p0nX (Hydr0p0nX!~hydr@ has quit ()
[01:17:39] Hydr0p0nX (Hydr0p0nX!~hydr@ has joined #mythtv
[02:25:44] SteveGoodey (SteveGoodey! has joined #mythtv
[02:50:24] peper03_ (peper03_!~peper03@mythtv/developer/peper03) has joined #mythtv
[02:52:17] moparsthbest (moparsthbest!~mitb@unaffiliated/moparisthebest) has joined #mythtv
[02:52:18] peper03 (peper03!~peper03@mythtv/developer/peper03) has quit (*.net *.split)
[02:52:18] moparisthebest (moparisthebest!~mitb@unaffiliated/moparisthebest) has quit (*.net *.split)
[02:52:18] peper03_ is now known as peper03
[02:54:20] SteveGoodey (SteveGoodey! has quit (Quit: Konversation terminated!)
[03:06:37] peper03 (peper03!~peper03@mythtv/developer/peper03) has quit (Ping timeout: 265 seconds)
[03:06:54] peper03_ (peper03_!~peper03@mythtv/developer/peper03) has joined #mythtv
[03:07:01] peper03_ is now known as peper03
[03:34:10] sl1ce_1g (sl1ce_1g! has quit (Remote host closed the connection)
[03:35:20] sl1ce (sl1ce! has joined #mythtv
[04:03:18] gigem (gigem!~david@mythtv/developer/gigem) has quit (Ping timeout: 245 seconds)
[04:16:21] gigem (gigem!~david@mythtv/developer/gigem) has joined #mythtv
[04:51:40] fetzerch_ (fetzerch_!~quassel@unaffiliated/fetzerch) has joined #mythtv
[04:56:03] fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has quit (Ping timeout: 264 seconds)
[05:18:12] arescorpio (arescorpio! has quit (Excess Flood)
[05:55:41] jackdahi (jackdahi! has joined #mythtv
[05:56:15] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has quit (Quit: Poof)
[05:57:45] jackdahi (jackdahi! has left #mythtv ()
[05:58:22] kormoc (kormoc!~kormoc@mythtv/developer/kormoc) has joined #mythtv
[07:34:52] FabriceMG (FabriceMG! has joined #mythtv
[07:41:18] len (len! has quit (Read error: Connection reset by peer)
[07:43:22] FabriceMG (FabriceMG! has quit (Ping timeout: 240 seconds)
[07:46:22] Tobbe5178 (Tobbe5178!~asdf@2001:2002:d9d4:ce7e:cc72:86d1:fa4c:4a58) has joined #mythtv
[07:48:43] Roklobsta (Roklobsta! has joined #mythtv
[08:13:01] Roklobsta (Roklobsta! has quit (Ping timeout: 252 seconds)
[08:13:10] Roklobsta (Roklobsta! has joined #mythtv
[08:17:43] FabriceMG (FabriceMG!~Thunderbi@ has joined #mythtv
[08:20:41] SteveGoodey (SteveGoodey! has joined #mythtv
[08:53:20] Merlin83b (Merlin83b! has joined #mythtv
[09:31:07] SteveGoodey (SteveGoodey! has quit (Quit: Konversation terminated!)
[09:48:00] buu (buu! has left #mythtv ()
[09:49:38] FabriceMG (FabriceMG!~Thunderbi@ has quit (Ping timeout: 246 seconds)
[09:50:53] FabriceMG (FabriceMG! has joined #mythtv
[10:44:27] clever (clever! has quit (Ping timeout: 276 seconds)
[10:46:09] clever (clever! has joined #mythtv
[10:52:54] SteveGoodey (SteveGoodey! has joined #mythtv
[11:12:36] SteveGoodey (SteveGoodey! has quit (Quit: Konversation terminated!)
[11:25:53] dekarl1 (dekarl1!~dekarl@mythtv/developer/dekarl) has joined #mythtv
[11:28:17] dekarl (dekarl!~dekarl@mythtv/developer/dekarl) has quit (Ping timeout: 246 seconds)
[12:31:21] moparsthbest (moparsthbest!~mitb@unaffiliated/moparisthebest) has quit (Ping timeout: 246 seconds)
[12:46:46] moparisthebest (moparisthebest!~mitb@unaffiliated/moparisthebest) has joined #mythtv
[13:13:16] stuartm:
[13:14:42] stuarta: yeah, i found that interesting
[13:21:47] warpme (warpme! has joined #mythtv
[13:22:11] stuartm: xbmc has the same coding standards as us? Looking at the changes I could have sworn for a minute that I was actually looking at MythTV code
[13:23:14] stuartm: well same variable naming standard, they seem to use a 2 char ident
[13:24:38] warpme: greetings all!
[13:27:05] stuarta: have enabled rss feeds on the forum. time to promote a link
[13:27:55] stuartm: stuarta: weren't they always enabled? I'm sure I remember posting links to the -users list when we first launched the forum
[13:28:04] warpme: stuartm: in respect of #12369: yesterday i had another segfault. Interesting is that probably I start to see some regularity with these segfaults: it looks like they happening after backend restart while browser has still opened thesame page with webfrontend. It pretend like kind of missmatch between be and browser internal state. it is pure hipothessis – but I'm wonder Your opinin here
[13:28:32] stuarta: stuartm: we didn't have any sections defined which apparently disables them
[13:29:24] stuartm: the websocket code tries to reconnect after a backend restart, I suppose it's possible that a connection attempt made while the websocket server is still starting up is triggering something
[13:30:00] stuartm: warpme: ^^
[13:31:02] ** stuarta tries something **
[13:31:16] stuartm: huh, strange
[13:31:36] stuartm: #12369
[13:31:36] ** MythLogBot **
[13:31:49] stuarta: hmm, okay, makes no difference, the feed is still embedded
[13:32:49] warpme: yesterday sequence was following: 1.have working be & webfrontend. all ok; 2.upgrade be, restart be, frontend page still opened, not touched durring be upgrade; started; 4.go to webfrontend, list rec.rules, ask for rec.rule detais; 5.after fes sec of webfrontend silence details popups; 6.1..2sec later be segfaults
[13:33:06] stuartm: warpme: doesn't seem from the log though that it occurs before or during server startup, so there must be something else going on
[13:34:08] stuartm: warpme: how long between starting the backend and making the request in the webfrontend?
[13:35:11] warpme: for sure segfault is trigered by user actions on webfrontend. Common thing seems to be BE restart while browser has constantly opened rec.details page.
[13:35:46] warpme: sory: rec.rules page not rec.details
[13:37:09] warpme: i restarted BE and watchig BE logs for first succesfull reschedule. I'm assuming BE is fully functiuonal after first successfull schedulle.
[13:38:11] warpme: stuartm: If You want – I can attach log from yesterday segfault to #12369
[13:38:11] ** MythLogBot **
[13:38:47] stuartm: warpme: it might reveal something new, so please
[13:39:33] warpme: should I attach only BE trace or also BE log or BE+system logs?
[13:40:05] stuartm: bt and be log
[13:41:06] stuartm: the qscript engine is a black box, hard to tell what's happening from a back trace, the qt debugger might help but that only works on windows at the moment
[13:41:35] warpme: interesting is that kernel generated core dump for prcess named 'PT23' not 'mythbackend'
[13:45:29] stuarta: is that a thread dump?
[13:52:35] stuarta: hmmm, time to kill off some old builds
[13:53:11] warpme: stuarta: right. Kernel's core dump is following "core.!usr!bin!mythbackend.PT23.9223". My script parses this and auto prepares last 5min of process/system logs + process trace. It looks like I have to update parsing to include case when las part of codedump filename has TID+PID....
[13:53:55] stuarta: does anyone give a crap about freebsd9 builders?
[13:54:30] warpme: stuartm: traces already attached.
[13:55:11] ** stuarta decides to kill it **
[13:58:42] stuarta: anyone attached to the iptv branch?
[13:59:10] stuartm: stuarta: seems like it's dead atm
[13:59:35] stuartm: I'm not sure whether there are outstanding changes in there that could or should be merged back into master
[14:00:03] warpme: btw: hans anybody know how to compie qt4.8.6 with debug symbols?. I added -debug to configure and disable strip in package generator – but it look like copiled qt packages still don't have symbols....
[14:00:12] warpme: s/hans/has/
[14:01:43] Roklobsta (Roklobsta! has quit (Ping timeout: 252 seconds)
[14:02:25] stuartm: why on earth is it sending google analytics cookies in the request to the WebFrontend?
[14:04:09] stuartm: which browser is this?
[14:05:33] stuartm: Firefox 35 ...
[14:06:01] stuartm: hmm, didn't think any browsers still supported global cookies, creepy
[14:07:41] stuartm: Feb 11 22:34:24 mythtv mythbackend[9223]: ASSERTION FAILED: m_heap.operationInProgress == NoOperation
[14:07:42] stuartm: Feb 11 22:34:24 mythtv mythbackend[9223]: (../3rdparty/javascriptcore/JavaScriptCore/runtime/Collector.cpp:366 void* QTJSC::Heap::allocate(size_t))
[14:08:21] stuartm: warpme: that looks like the issue, an assert in javascriptcore
[14:08:48] stuartm: that's pulled in by QT
[14:09:01] MythBuild (MythBuild! has quit (Remote host closed the connection)
[14:09:38] MythBuild (MythBuild! has joined #mythtv
[14:10:10] stuartm: now I believe we should be catching that – dblain any thoughts?
[14:10:58] MythBuild (MythBuild! has quit (Remote host closed the connection)
[14:11:12] MythBuild (MythBuild! has joined #mythtv
[14:13:18] stuartm: warpme: any updates for JavaScriptCore for your distro?
[14:15:06] stuartm: although it should be part of QT ...
[14:15:28] MythBuild (MythBuild! has quit (Remote host closed the connection)
[14:15:42] MythBuild (MythBuild! has joined #mythtv
[14:16:09] stuartm: warpme: looks like a QT/JSC bug and not something we can fix from our end, but I'll continue looking at it to see if we can somehow avoid triggering it
[14:16:10] MythBuild (MythBuild! has quit (Remote host closed the connection)
[14:16:30] MythBuild (MythBuild! has joined #mythtv
[14:17:29] stuartm: warpme: it shouldn't be crashing unless QT/JSC was built in debug mode, so that's something to look at
[14:19:02] stuarta: right, cleanup done
[14:19:05] stuartm: although maybe it still would, if the error condition wasn't properly handled
[14:21:15] warpme: stuartm: I'm not sure but I believe QT/JSC comes with Qt in my distro (ArchLinux). How can I verify this?
[14:23:57] warpme: stuartm: You mention "it shouldn't be crashing unless QT/JSC was built in debug mode". I'm building QT with -debug switch in configure. maybe this is problem here?
[14:25:36] stuartm: warpme: it may NOT crash if QT wasn't built in debug mode, enabling debug mode enables ASSERT, breakpoints showing the developer where the internal state is not as they expect it to be
[14:26:21] ** stuarta recycles a few old build vm's **
[14:26:24] Jordack (Jordack! has joined #mythtv
[14:26:40] stuartm: assert deliberately stops execution of the program, but a good developer will also add error handling that in production (not debug) will handle the error and continue execution
[14:28:40] stuartm: assert is the equivalent of the Director yelling "Stop!, Stop!, it's all wrong" then walking onto stage during the dress rehearsal of the play – something he wouldn't do during the live production, the mistake would just be ignored :)
[14:31:22] warpme: ah OK. So definitelly I'll rebuild 4.6.8 without -debug. Oh BTW: first trace in #12369 was with Archlinux stock 4.8.1 Qt package. Looking on Arch sources it was build without -debug switch (so with defaults). I believe defaults are -release (not sure as there is no info in configure help)
[14:31:22] ** MythLogBot **
[14:34:32] warpme: stuartm: Is it possible that QT/JSC mentioned by You is provided by "js17" package described "JavaScript interpreter and libraries (legacy)"?
[14:40:51] stuartm: warpme: maybe, if you already have that installed
[14:41:28] stuartm: warpme: is there anything special about the rule you try to edit – is it a search rule, or custom rule?
[14:42:01] warpme: OTOH is see alot of references to javascript during QT compile...
[14:43:34] stuartm: warpme: it's probably built-in to qt then
[14:45:01] warpme: stuartm: absolutelly on. I simply was walking durring various parts of mythfrontend to see whats new here. BTW: webfrontend becomes REALLY NICE thing. It is altready impressive! But when we will have on-demand-transcode and support for music – it will be killer part of MythTV!
[14:45:09] warpme: s/on/no/
[14:46:19] stuartm: warpme: ok, looking for reasons our code might be triggering a fault in the javascript engine, not finding anything yet
[14:47:50] warpme: BTW: I'm thinking how can I setup my system to have access to webfrontend across our corporate web proxy. Will world port 80 redirected to 6544 (+some security) be enough?
[14:48:52] clever (clever! has quit (Ping timeout: 240 seconds)
[14:50:02] stuartm: warpme: I'm building up the security and resilience of the internal webserver, so I'd hope to reach that point, but right now I'd have to say no it's not yet safe enough to expose the webfrontend to the internet
[14:50:21] stuartm: the main problem is actually the services API, there's no authentication on that at all
[14:50:38] stuartm: and it's exposed via the same server
[14:51:28] stuartm: warpme: if you put the WebFrontend behind a proxy with some http authentication (digest) in front, then yes it's perfectly safe
[14:51:49] warpme: I was not following internal architecture but is BE side of webfrontend calling services API or rather webbrowser side is doing this?
[14:52:11] stuartm: better still if you define certificates and use TLS (SSL) – https://{backend_IP}:6554 (note different port)
[14:53:13] stuartm: warpme: for all the stuff I've written it's calling the services API on the server side, but the older setup stuff accesses a lot of stuff directly from the client
[14:54:54] warpme: oh – for sure I have to go with SSL. I'm wonder is it possible to organize my fronting HTTP server to give access to webfrontend by putting prefix in URL. This will require on-line URL rewrites. It is possible to do this with webfrontend?
[14:55:42] warpme: I mean will webfrontend deal OK with HTTP proxy which rewrites URLs?
[14:56:40] stuartm: depends on exactly how it does it, not really familiar with the options for proxying
[14:59:47] warpme: right. I believe as long as webfrontend is not using URLS in 'data' packets – all should be OK. This is IMHO exact analogy to NAT and ALG gateways for protocols like SIP or active FTP. I'll investigate this :-). BTW: do we have wiki somwhere about gen.SSL certs for mythtv?
[15:00:50] stuarta: the proxying web server should do the translation in place
[15:01:03] stuarta: what i'm unsure about is how the websockets will work with that
[15:01:24] stuartm: warpme: no, I've not added anything about generating self-signed certificates yet, there are plenty of howtos online, although the quality of those varies greatly, it's actually pretty simple but many guides make it seem more difficult
[15:02:06] stuartm: stuarta: websocket should be transparent to proxies, that's why it's started as a standard HTTP query
[15:02:38] stuartm: it's also why all communication from the client to the server is masked (obscured) to prevent proxy cache poisoning
[15:04:14] warpme: stuartm: indeed – finding quide how to gen.certs is easy. May You hint me where i should put certs for getting them by BE webserver?
[15:04:33] stuartm: at the weekend I plan to switch the WebFrontend and Service API to require authentication by default on external IP addresses, maybe internal too, haven't decided about that yet
[15:07:29] stuarta: cool
[15:07:42] stuartm: warpme: you can put them anywhere you like, paths are specified via the following settings – hostSSLKey (path to host key file), hostSSLCertificate ( path to host certificate), caSSLCertificate (path to ca certificate, if applicable)
[15:08:42] warpme: oh – nice. we have DB setting for this. perfect
[15:09:23] warpme: stuartm: are You considering to have BE setup webserver running on different port than webfrontend?
[15:11:42] stuartm: no, we already have a lot of ports in use, users may revolt if we expand that number too much more
[15:12:28] stuartm: besides, it potentially means having more servers running which means more resources required
[15:14:23] stuartm: a simple password based authenication in front of the whole lot, maybe with certificate based authentication added later will be enough
[15:15:53] stuartm: jheizer: what do you think to having to do digest authentication in order to use the services API ? From the user's perspective it would just involve entering their username/password into the client which were then used to grant access to the backend
[15:17:00] stuartm: the client could remember the details for the user, so it would be a one-time setup
[15:18:55] jheizer: No complaints here. Since I provide public proxy access to the backend services my current security is just a white list of what URLs are able to be called from the client.
[15:19:21] jheizer: And if they want MM itself secured they can htaccess it on their own
[15:21:01] jheizer: And while I had always planned on auto detecting the backend and everything, I never got around to finishing it so they already have to hand key in stuff.
[15:21:47] jheizer: This is all assuming attaching a digest auth to .net web services calls is a built in thing already. If not be a pain in the butt.
[15:22:27] stuartm: I imagine it should be built in, it forms part of the original HTTP 1.1 specification
[15:22:55] stuartm: although web standards and Microsoft ...
[15:24:55] jheizer: It looks like you have to jimmy it in there. Since I point VS at the wsdl and it just auto generates everything the only time I even care about the underlying services calls is for the image urls that I proxy for the "FE" directly.
[15:29:06] stuartm: only way to know how it will cope is to try it and see
[15:29:44] stuartm: guess we'll find out when I've done the work
[15:30:18] jheizer: Yeah, when you do it and check it in let me know and I'll update my master VM and give it a try.
[15:31:15] jheizer: One other thing is I wonder how tools that do the generation for .net/java/whatever would handle the wsdl being behind digest.
[15:31:34] jheizer: Or if they'll all be fine with the simple http://user:pass@
[15:33:13] stuartm: I can probably exclude the wsdl from the authentication
[15:33:38] stuartm: that's not really sensitive in any way that I can think of
[15:34:32] jheizer: K, wasn't sure if it was possible based on what you were planning. Just a thought.
[15:35:38] gigem: stuartm, stuarta: I mentioned the Haswell deinterlacing back in December. I got my parents an ASUS Chromebox for Christmas and it's a very nice box. About the only thing it can't do is timestretch H.264 higher than 1.5x or so.
[15:37:01] warpme: gigem: do You have running 2x DI on Haswell?
[15:38:17] gigem: warpme: Not with MythTV. I only tested it with Kodi. I did put it on my TODO list, but it's behind the recorder/scheduler changes I'm slowly working on.
[15:38:47] stuartm: gigem: it should be implemented, but I don't have the hardware or time to tackle it, I don't even know whether I have the relevant knowledge to implement it – it's one thing if it's simply a matter of sending the right arguments to libvaapi, but I suspect it may require more work on our end
[15:39:23] warpme: ah now i understand. Indeed – Kodi can get almost full potential of VPP – that's why for Kodi Intel's HW is nice....
[15:41:08] stuarta: gigem: i'm on a 4th gen intel nuc with i3 processor. verynice
[15:41:56] gigem: stuartm: It's the same for me. When I finish my current changes, I intend to get a small(er), probably Intel based frontend — either a Broadwell ChromeBox or build my own. If nothings benn done on the VAAPI front by then, I'll probably take a stab at it myself.
[15:42:47] gigem: stuarta: The NUCs are nice. The ChromeBoxen are noticably cheaper, though, and not that much bigger.
[15:44:23] stuartm: can you replace the OS on the chromebox? I like the format, but I don't really like the idea of Chrome OS
[15:44:27] warpme: gigem: I'm wonder how new recorder work You are working on will deal with multi-rec. Will there be virtual tuner concept at all?
[15:44:37] gigem: FWIW, the ChromeBox I got my parents was on sale for US$135. All I neede to add was an IR receiver since it came with 2GB of RAM and a 16GB SDD. AIUI with the NUCs, you still have to add memory and storage.
[15:44:51] stuarta: yeah you have to add
[15:45:40] stuartm: that's the bit I don't like about the NUCs, they seem a little expensive to start with and then you have to supply your own harware on top?
[15:45:45] gigem: warpme: Yes, virtual tuners won't go since there will still be cases where they are the only solution.
[15:47:38] stuarta: celeron tho? gigem what cpu is in the one you have?
[15:47:59] ** stuarta has an aversion to the name celeron from about 10yrs ago **
[15:51:13] stuartm: all depends what you're trying to get from it, as a frontend I'm sure it's more than capable
[15:52:11] stuarta: i really don't understand why intel recycled the celeron name when people have such an aversion to it
[15:52:13] gigem: stuarta: In the ChromeBox? It's a 2955U. Yes, it's technically a Celeron, but for this particular model, it's a Haswell chip and is really not that different than an underclocked i3. For example, even though this chip only runs at 1.4GHz, it has no problem decoding HD MPEG2 at 2x and HD H.264 at ~1.4x in software.
[15:52:14] warpme: OMG: I have this Intel naming for their GFX. On 2955U GFX is "HD (Haswell)". How in hell this compares to i.e. HD4000 in terms of VPP caps?
[15:52:17] stuartm: although the lack of IR seems like an odd decision, iirc Google want you to control it via your phone or tablet – which is fine as an option, but neither is a proper replacement for a remote control
[15:52:21] warpme: s/have/hate/
[15:53:06] stuarta: stuartm: getting the ir receiver to work on the NUC is a pain because of bios bugs, i do however have a new bios to flash
[15:53:29] stuartm: warpme: all graphics manufacturers have terrible naming systems, it's a complete minefield
[15:53:38] gigem: stuartm: For the Chromebox, Google expects you to hook up a keyboard and mouse not an IR remote. It's only supposed to run Chrome and little more after all.
[15:53:44] stuarta: does anyone else see the screensaver kicking in while watching programs? it's really beginning to piss me off such that i might have to fix it
[15:54:28] gary_buhrmaster: stuarta: Intel branding is interesting. But I have never understood marketing anyway (that is not entirely true, I understand marketing, I just despise how misdirection has been moved to necessary art form for most companies).
[15:54:34] stuarta: my workaround has been to set the timeout to 1hr or disabled
[15:55:19] stuartm: I hate the way Nvidia will take older hardware and renumber it, so a 610 is worse than a 480? How does make any sense? Then they have the GT vs GTX confusion, and even once you figure out the model numbers they are still largely meaningless until you factor in the other stats like memory bandwidth
[15:56:17] stuartm: I want a simple system, where no matter what hardware is actually on a card the higher the model number the better it simply is
[15:57:07] gary_buhrmaster: stuartm: "Get used to disappointment".
[15:57:10] jheizer: That sounds too easy and straight forward...
[15:57:33] stuartm: stuarta: which distro? We have code to disable the screensaver but every distro does their own thing with screensavers and they don't honour the X settings
[15:58:15] stuarta: gnome3 is especially bad
[15:58:30] stuarta: lemme see, i'm seeing it on mint, ubuntu & fedora
[15:58:40] stuarta: ubuntu only started after a recent update
[15:58:53] stuartm: at one point we had three different methods, for X, kde and gnome, and it still didn't work on all platforms
[16:00:20] stuartm: stuarta: the code you are looking for is in libs/libmythui/screensaver-x11.cpp
[16:00:44] stuartm: sphery is the expert on it
[16:01:19] warpme: yeah: all this is dirty marketing tricks riding on observation that higher number is perceived as "better". For produces it can give 2 potential benefits: 1\bigger margin by possibility to sell worse(=hepaer) product with higher price (possible due created miss-perception in buyer) or 2\increase reve. by creating later additional demand because miss-informed buyer will have replace initially bought product by better one. Both are dirty enough
[16:02:49] stuartm: I've seen lots of complaints from people who bought a 'new' card with a higher model number, only to find it performed worse than the one it replaced, that crap shouldn't happen
[16:05:13] stuartm: warpme: let me know how the ssl stuff works for you, as far as I know, I'm the only one who actually uses it
[16:05:44] stuartm: I plan on exposing the settings for the key/certificate via the webfrontend itself, and possibly in mythtv-setup
[16:08:04] warpme: right. this is example how vendors are exchanging long-term loyality to short term reve/margins increase. I hate todays times with all this dirty marketing tricks assuming user becomes more dumb and more pocket than brain.
[16:08:59] dekarl-work (dekarl-work!51c8c678@gateway/web/freenode/ip. has joined #mythtv
[16:09:08] warpme: stuartm: sure. I'll try this today evening :-)
[16:09:14] gary_buhrmaster: With nVidia, once it is explained that the first number is the marketing generation and the second number is the marketing target capability, it starts to make sense that a 610 is less than a 480, but expecting the average consumer to need a secret decoder ring (or competent sales people (an oxymoron)) is not an ideal world.
[16:10:00] warpme: gary_buhrmaster: nice :-)))))))))))
[16:14:34] warpme: stuarta: regarding spourious xscreensaver problems: I have such thing rare: simply during playback screensaver kicks-on. it pretends like xscreensaves stops to honour disable call. I have it only on small subset of movies and only on movies with 23.976fps. Usually Xserver restart is needed as restarting xcreensaver not helps.
[16:15:32] stuarta: no idea what it is. happening all the time on my prod frontend now, and it's bloody annoying
[16:16:59] knightr (knightr!~Nicolas@mythtv/developer/knightr) has quit (Read error: Connection reset by peer)
[16:17:16] dekarl-work: Am I the only one using software volume control? The frontend resetting the volume on commercial cuts is not so nice.
[16:17:24] knightr (knightr! has joined #mythtv
[16:17:24] knightr (knightr!~Nicolas@mythtv/developer/knightr) has joined #mythtv
[16:17:24] knightr (knightr! has quit (Changing host)
[16:17:36] dekarl-work: workaround is exit the recording (persists the volume change) then watch
[16:17:51] warpme: stuarta: what ver. of Xorg-server/xscreensaver are You running?
[16:19:16] stuarta: no idea off the top of my head
[16:24:09] stuartm: dekarl-work: maybe, I avoid software volume control since it can distort the audio
[16:24:17] Tobbe5178 (Tobbe5178!~asdf@2001:2002:d9d4:ce7e:cc72:86d1:fa4c:4a58) has quit (Read error: Connection reset by peer)
[16:25:06] stuartm: I have mythfrontend controlling the pcm volume mixer
[16:26:59] dekarl-work: hmm, I need to look at my terms, I picked one of the ALSA mixers, too
[16:27:37] dekarl-work: But your frontend is not "resetting" the volume on stream changes or similar?
[16:29:30] sl1ce_1g (sl1ce_1g! has joined #mythtv
[16:30:25] sl1ce (sl1ce! has quit (Ping timeout: 265 seconds)
[16:30:25] kc (kc!~Casper@unaffiliated/kc) has quit (Ping timeout: 265 seconds)
[16:30:47] kc (kc! has joined #mythtv
[16:30:47] kc (kc!~Casper@unaffiliated/kc) has joined #mythtv
[16:30:47] kc (kc! has quit (Changing host)
[16:38:41] SteveGoodey (SteveGoodey! has joined #mythtv
[16:39:59] warpme (warpme! has quit (Remote host closed the connection)
[16:57:40] clever (clever! has joined #mythtv
[17:05:53] stuartm: not that I've noticed
[17:08:33] stuartm: dekarl-work: I thought you meant the software volume mixer which adjusts the compression of the audio to give the impression of altering the actual volume, vs the ALSA mixer which actually changes the volume at the card, adjust the current to the speakers instead
[17:09:21] stuartm: software volume control lets you alter the volume of the application without changing the volume for everything else on the same system
[17:09:24] dekarl-work: I have to check what I currently use. was switching between HDMI and analogue connection
[17:10:12] stuartm: but it's nasty IMHO, often results in distortion at higher volumes
[17:10:34] FabriceMG (FabriceMG! has quit (Quit: FabriceMG)
[17:10:53] dekarl-work: Now that I noticed the volume reset simply when running into a cut its WAF relevant, when I just noticed it when manually switching between alternate audio streams it wasn't so high on my list. (just noticed it last night)
[17:11:11] stuartm: strange
[17:11:47] dekarl-work: It started to appear when caching of the volume was added (to avoid hitting the database once per remote key press)
[17:12:11] dekarl-work: looks like there is a place where we reset to the stored volume without storing the changed volume first
[17:12:47] dekarl-work: I browsed the relevant changes and looked around the related code but didn't see any obvious issue
[17:13:47] dekarl-work: but as always ;)
[17:30:16] MythBuild (MythBuild! has quit (Remote host closed the connection)
[17:30:31] MythBuild (MythBuild! has joined #mythtv
[17:30:32] clever (clever! has quit (Remote host closed the connection)
[17:32:05] clever (clever! has joined #mythtv
[17:32:24] dekarl-work (dekarl-work!51c8c678@gateway/web/freenode/ip. has quit ()
[17:52:43] len (len! has joined #mythtv
[17:54:17] Merlin83b (Merlin83b! has quit (Quit: Leaving)
[18:05:07] MythBuild: build #1 of master-debian-wheezy-qt5–32bit is complete: Success [3build successful] Build details are at . . . bit/builds/1
[18:15:03] stuartm: so, any reason not to push the change to the file extension for mpeg-ts recordings?
[18:41:14] natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has joined #mythtv
[18:58:21] Roklobsta (Roklobsta! has joined #mythtv
[19:06:52] skd5aner (skd5aner! has quit (Ping timeout: 250 seconds)
[19:15:26] skd5aner (skd5aner! has joined #mythtv
[19:16:46] MythBuild: build #3380 of cppcheck-master is complete: Failure [4failed shell_2] Build details are at . . . /builds/3380 blamelist: Stuart Auchterlonie < >
[19:33:57] MythBuild: build #45 of 0.27-centos7–64bit is complete: Exception [6exception setproperty] Build details are at . . . it/builds/45 blamelist: Stuart Auchterlonie < >
[19:37:53] tonsofpcs (tonsofpcs! has quit (Changing host)
[19:37:53] tonsofpcs (tonsofpcs!~mythbuntu@rivendell/member/tonsofpcs) has joined #mythtv
[19:44:17] stuarta: clearly the build host had a wibble
[19:51:37] MythBuild: build #321 of 0.27-ubuntu-12_04-lts-64bit is complete: Success [3build successful] Build details are at . . . t/builds/321
[19:57:38] MythBuild: build #317 of 0.27-debian-wheezy-64bit is complete: Success [3build successful] Build details are at . . . t/builds/317
[20:48:27] MythBuild: build #3381 of cppcheck-master is complete: Success [3build successful] Build details are at . . . /builds/3381
[21:17:37] stuartm: gigem: it may be nothing, but I noticed something odd in my logs that I don't remember seeing before – the EIT scanner tried to tune to a channel, but failed because that tuner was already in use – I'm wondering if it's somehow related to the input changes you've already committed, even though they shouldn't have altered the behaviour any
[21:18:59] stuartm: the active scanner should be stopped for an input (and all it's virtual inputs) before a recording starts on it
[21:20:12] stuartm: it would seem that this no longer happens
[21:36:59] Roklobsta (Roklobsta! has quit (Quit: KVIrc 4.2.0 Equilibrium
[21:50:23] natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has quit (Ping timeout: 245 seconds)
[21:53:57] jya (jya!~jya@mythtv/developer/jya) has quit (Ping timeout: 244 seconds)
[21:55:01] jya (jya!~jya@mythtv/developer/jya) has joined #mythtv
[21:56:50] stuartm: . . . l__graph.png
[22:04:20] stuarta: no wonder it's hard to understand
[22:04:29] stuarta: what drew that btw?
[22:22:04] SteveGoodey (SteveGoodey! has quit (Quit: Konversation terminated!)
[22:25:11] stuartm: doxygen
[22:25:37] stuartm: appears here – . . . derBase.html
[22:25:52] stuartm: generates one for each class
[22:48:36] dekarl1: stuarta,
[22:48:42] dekarl1 is now known as dekarl
[22:49:21] gigem: stuartm: I suppose anything is possible. Can you check your capturecard and inputgroup tables to make sure your virtual tuners are in the appropriate groups?
[23:02:51] Jordack (Jordack! has quit ()
[23:54:12] Sharky112065 (Sharky112065! has quit (Quit: “The only way to have a friend is to be one.” ― Ralph Waldo Emerson)
[23:59:22] joki (joki! has quit (Ping timeout: 240 seconds)

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