:: #mythtv

Daily chat history

Current users (71):

aloril_, amessina, Anssi, blahdodo, brfransen, Captain_Murdoch, ChanServ, Chutt, clever, coling, dblain_, dekarl, eharris, ElmerFudd, enyc, gary_buhrmaster, ghoti, gregbert, GreyFoxx, Guest65046, Guest65292, Hydr0p0nX, ikevin, J-e-f-f-A, jab416171, jams, jheizer, jmcentee, jnylen, joki, jpabq, jpharvey, jst_, jya, kc, knightr, letifosiferrari, libsci, mad_enz, markspieth, Merlin83b, MythBuild, MythLogBot, nephyrin, peper03, poptix, pppingme, rich0, Roklobotomy, saaki, Seeker, seld, ShapeShifter499, sphery, SpootDev, sraue, stuartm, suffice, superm1, taylorr, tgm4883, tonsofpcs, tris, unforgiven512, vincent42, wagnerrp, Warped, XDS2010, xris, _charly_, _iwc
Wednesday, February 24th, 2016, 01:00 UTC
[01:00:40] amessina (amessina!~amessina@unaffiliated/amessina) has quit (Quit: Konversation terminated!)
[01:16:56] superm1: Natanojl that is wrong 3.0.1 is what's complied against and running
[01:20:26] superm1: tgm4883: i think natanojl's assessment is wrong, there is only 3.0.1 in the archive
[01:20:40] tgm4883: ok
[01:25:24] superm1: i'm hoping our 16.04 ISO will be working soon and can debug more easily in a VM what's up
[01:28:14] ShapeShifter499 (ShapeShifter499!~ShapeShif@unaffiliated/shapeshifter499) has joined #mythtv
[01:28:53] Roklobotomy (Roklobotomy! has quit (Quit: Leaving)
[01:30:34] andreaz (andreaz! has quit (Read error: Connection reset by peer)
[03:46:11] peper03 (peper03!~peper03@mythtv/developer/peper03) has quit (Ping timeout: 248 seconds)
[03:47:11] peper03 (peper03!~peper03@mythtv/developer/peper03) has joined #mythtv
[04:23:07] GhostOfRaven (GhostOfRaven! has quit (Quit: Leaving)
[04:29:21] GhostOfRaven (GhostOfRaven! has joined #mythtv
[04:29:34] dmfrey (dmfrey! has quit (Ping timeout: 244 seconds)
[05:11:13] GreyFoxx (GreyFoxx! has quit (Ping timeout: 240 seconds)
[05:17:03] GreyFoxx (GreyFoxx! has joined #mythtv
[05:37:34] poopshoot (poopshoot!446fb755@gateway/web/freenode/ip. has joined #mythtv
[05:52:48] poopshoot: thanks for all your work!
[06:11:30] poopshoot (poopshoot!446fb755@gateway/web/freenode/ip. has left #mythtv ()
[07:00:21] SteveGoodey (SteveGoodey! has joined #mythtv
[07:13:47] SteveGoodey (SteveGoodey! has quit (Quit: Konversation terminated!)
[07:14:33] SteveGoodey (SteveGoodey! has joined #mythtv
[07:28:57] Tobbe5178 (Tobbe5178! has joined #mythtv
[07:45:10] knightr (knightr! has quit (Read error: Connection reset by peer)
[08:12:31] lautriv__ (lautriv__!~lautriv@funtoo/user/lautriv) has joined #mythtv
[08:16:10] lautriv_ (lautriv_!~lautriv@funtoo/user/lautriv) has quit (Ping timeout: 255 seconds)
[08:19:46] natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has joined #mythtv
[08:21:56] natanojl: superm1: tgm4883: Then why is it showing in the memory map?
[08:37:54] Roklobotomy (Roklobotomy! has joined #mythtv
[09:04:04] joki (joki! has quit (Ping timeout: 248 seconds)
[09:09:08] joki (joki! has joined #mythtv
[09:11:05] stuarta: i think somebody with CEC needs to run valgrind or similar
[09:12:57] stuarta: either that or we have to write a unit test and valgrind that. might actually be faster
[09:28:33] jya: stuartm: managed to reproduce the crash in mythmusic?
[09:30:18] GhostOfRaven (GhostOfRaven! has quit (Read error: Connection reset by peer)
[09:33:51] stuartm: not yet no
[09:34:34] stuartm: which is a good thing, means the chance of it happening is low enough
[09:35:17] markspieth (markspieth! has joined #mythtv
[10:08:30] stuarta: Roklobotomy: if we wrote a unit test for the cec code would you be able to dig your adapter out and try it?
[10:14:02] Roklobotomy: i would but my tv isn't cec (facepalm)
[10:14:12] Roklobotomy: so pulse8 gathers dust
[10:15:43] lomion0815 (lomion0815!5bd9374e@gateway/web/freenode/ip. has joined #mythtv
[10:16:13] stuarta: Roklobotomy: tbh, i don't think it matters if the device is connected to a tv
[10:16:24] Roklobotomy: ok
[10:16:32] stuarta: from the stacktrace, the error seems to occur when opening the device
[10:21:20] lomion0815: I'm trying to find a bug: When using "episode sort order" season/episode is set it is done for every year ...
[10:22:15] lomion0815: So i see in one recordings list the ordered episodes recorded in 2016 followed by the ordered list of 2015 ... no matter what the season/episode says
[10:22:44] lomion0815: but I don't find anything checking the year in playbackbox-code ...
[10:22:57] lomion0815: driving me nuts!
[10:33:27] dblain_ (dblain_! has joined #mythtv
[10:33:38] dblain_ (dblain_! has quit (Changing host)
[10:33:38] dblain_ (dblain_!~dblain@mythtv/developer/dblain) has joined #mythtv
[10:33:49] dblain (dblain!~dblain@mythtv/developer/dblain) has quit (Read error: Connection reset by peer)
[10:44:20] Roklobotomy (Roklobotomy! has quit (Quit: Leaving)
[11:49:13] knightr (knightr! has joined #mythtv
[11:49:14] knightr (knightr! has quit (Changing host)
[11:49:14] knightr (knightr!~knightr@mythtv/developer/knightr) has joined #mythtv
[12:44:32] stuartm: does someone want to help this guy with his storage group configuration, I'm assuming that has to be the issue he's having with 'browse library'
[12:45:58] stuartm: I actually see MORE folders with Browse Filesystem not less, since it operates from the root for all video storage groups
[12:46:28] stuartm: there's no logical reason he'd seen no parent folders in filesystem mode
[13:05:23] mad_enz (mad_enz!~mad_enz@2607:f090:488d:8d:828e:f92a:cb4:c56b) has joined #mythtv
[13:39:25] dmfrey (dmfrey!~dmfrey@ has joined #mythtv
[13:49:35] mad_enz (mad_enz!~mad_enz@2607:f090:488d:8d:828e:f92a:cb4:c56b) has quit (Ping timeout: 250 seconds)
[14:36:23] jheizer: stuarta, will the same cec error happen on RPis? If so the 2 builder can run the test for you also.
[14:46:01] stuarta: jheizer: possibly
[14:46:43] stuarta: and the idea of the unit test, was so that that small piece of code could be analyzed with valgrind, without using the whole frontend
[14:47:01] stuarta: tho that will probably struggle on the rpi
[14:47:54] jheizer: ah still needing valgrind. nevermind.
[14:48:03] stuarta: yeah, that requires memory
[14:48:10] jheizer: I was more on the then you don't have to wait for anyone thought
[15:00:12] mad_enz (mad_enz! has joined #mythtv
[15:07:48] lomion0815 (lomion0815!5bd9374e@gateway/web/freenode/ip. has quit (Quit: Page closed)
[15:27:22] dekarl1 (dekarl1!~dekarl@mythtv/developer/dekarl) has joined #mythtv
[15:29:33] dekarl (dekarl!~dekarl@mythtv/developer/dekarl) has quit (Ping timeout: 240 seconds)
[15:33:30] dekarl1 is now known as dekarl
[16:01:43] brfransen (brfransen!~brfransen@ has quit (Quit: Byebye)
[16:25:41] brfransen (brfransen!~brfransen@ has joined #mythtv
[16:30:07] gregL (gregL! has quit (Remote host closed the connection)
[17:18:58] gary_buhrmaster (gary_buhrmaster!~gtb@2601:647:4801:98e3:222:4dff:fe51:6728) has joined #mythtv
[17:21:04] gary_buhrmaster: stuarta: Re: CEC unit test. I can (I think) use USB passthrough (and plug my pulse eight adapter in) to one of my buildbots to run whatever you need (or run it native on a more capable system if needed).
[17:22:11] gary_buhrmaster: stuarta: (since I have not been following the issue, I just caught the trailing irc discussion. I can likely do other tests, too, I just do not know the particular issue you are looking for.)
[17:25:48] stuarta: gary_buhrmaster: lemme find the link for you, there's a stack smash reported on ubuntu 16.04
[17:26:27] stuarta: at time 21:06:51
[17:27:38] stuarta: if you look at the stacktraces, you'll see that the Open call in cecadapterprivate triggers the stack protection
[17:27:47] stuarta: in theory it should be simple to track down
[17:39:15] letifosiferrari (letifosiferrari!~letifosif@ has joined #mythtv
[17:55:28] Steve-Goodey (Steve-Goodey! has joined #mythtv
[18:01:27] gary_buhrmaster: In theory, yes. But I do not see the abort (in default fedora), and valgrind points out various issues, but not (currently) relevant to this issue.
[18:02:25] Steve-Goodey (Steve-Goodey! has quit (Quit: Konversation terminated!)
[18:05:31] gary_buhrmaster: Possibly different library version issues. I will need to compare and contrast those, or build a ubuntu vm and test there (which may be easier)
[18:28:00] gary_buhrmaster: (first I will try to recompile with toolchain hardened, I note I had not done so (so no wonder no abort, even if there was overflow) *sigh*).
[18:35:59] blahdodo (blahdodo!~blahdodo@ has quit (Ping timeout: 240 seconds)
[18:39:04] blahdodo (blahdodo!~blahdodo@ has joined #mythtv
[18:45:00] blahdodo (blahdodo!~blahdodo@ has quit (Ping timeout: 276 seconds)
[18:47:05] blahdodo (blahdodo!~blahdodo@ has joined #mythtv
[19:34:57] dmfrey (dmfrey!~dmfrey@ has quit (Quit: Ex-Chat)
[19:58:54] andreaz (andreaz! has joined #mythtv
[20:23:26] dmfrey (dmfrey!~dmfrey@ has joined #mythtv
[20:34:53] dmfrey (dmfrey!~dmfrey@ has quit (Quit: Ex-Chat)
[20:47:12] Roklobotomy (Roklobotomy! has joined #mythtv
[20:52:27] gigem: What are the downsides to using dvb_on_demand setting? Do some DVB cards take a particularly long time to open? I used to use on combined ATSC/analog cards with no ill effects. I think that forcing dvb_on_demand on is going to be the easy solution for the remaining diseqc showstopper.
[21:17:59] gary_buhrmaster: Do some DVB cards (re)load firmware at open? Nominally I would think not, but I had little/no experience with the wide variety of DVB cards out there.
[21:50:02] gigem: Hence my question. Reloading firmware I doubt, but there's some reason Daniel or a predecessor added the option. I wonder if git still has enough bread crumbs to track it donw.
[21:54:44] gary_buhrmaster: I suspect (but do not know) that the issue with "on_demand" is one of proper scheduling for a (potentially) random set of available tuners.
[21:55:04] gary_buhrmaster: Unless/Until someone adds in a lot more smarts for temporary card unavailability/retries/restores the user experience may be "different" with on_demand.
[21:57:05] SteveGoodey (SteveGoodey! has quit (Quit: Konversation terminated!)
[21:58:11] dekarl: would be nice if someone (hint gigem) could point out how failing fast due to tuner unavailability (DVB openend on demand, or HDHR locked by someone else) could work in the scheduler. aka "can't tune on tuner x, skip over to the next tuner and try again"
[21:58:34] dekarl: can we already do that, or do we need a bigger development?
[22:00:17] gary_buhrmaster: Remembering, of course, that BEs that people are putting on low/under powered systems might take quite some time for (re)scheduling...... (so you might need to open a tuner (and fail) minutes before a schedules recording).
[22:03:23] gary_buhrmaster: Seems like the challenge is not the simple case where you have overprovisioned tuners (so one (or another) will always be available), but the results where you end up having to throw the recording off the island to reserve a tuner for a future higher priority case.
[22:06:16] dekarl: backend on underpowered systems better not be handling dozens of tuners in the first place...
[22:06:18] gary_buhrmaster: [Everytime I suggest on the -users list that the only proper way to run MythTV is to overprovision tuners I receive strong disagreement (usually from those that later complain about missing recordings)]
[22:07:00] dekarl: how can that be, when we have one of the few solutions with multirec?
[22:08:26] dekarl: likely the same people that complain that 6 live tv sessions occupy 6 tuners
[22:08:28] gigem: gary_buhrmaster: I don't think the user experience would be all that different, just perhaps a little slower at recording start. FWIW, I setup my development system with two, inputs using the same atsc card with different qam lineups. This is essentially how diseqc is now intended to be configured. Not surprisingly, it failed with the same device is busy errors reported by the user when dvb_on_demand was
[22:08:30] gigem: off. As hoped, and mostly expected, it worked just fine when dvb_on_demand was on.
[22:09:59] gary_buhrmaster: In the US, multirec does not help as much as some other locations. That might change with the results from Auction 1000/1001, but the results (and changes) are some time off.
[22:10:35] gigem: dekarl: We have almost all of the pieces to do that already. Currently, failed recordings can be reactivated by the user. It's not really difficult to do the same programatically. The part we don't have is getting the scheduler to not choose the same (bad?) tuner over and over again.
[22:11:39] dekarl: ahh, not choosing the same bad/locked/pull from the usb socket tuner again appears to be important for the solution
[22:16:28] gigem: IME, what to do after tuning failures is not clear cut. Sometimes, the tuner is fubar until mythbackend is restarted or even the backend system rebooted. Other times, it's a transient error and retrying on the same tuner will work.
[22:16:43] dekarl: everything that I came up with is a more or less ugly hack. But generally an indenpendent issue from our recorder locking the tuner against usage by others. (HDHR locking topic)
[22:18:32] gigem: Regarding dvb_on_demand, from commit 34cad010, "Mian's patch to add an option to close the dvbchannel when not needed." I guess it was kept open because that's the way Chutt originally. Chutt, can you provide any more background?
[22:20:23] stuarta: gigem: i though dvb_on_demand was because of flaky hardware that crapped out if used constantly
[22:20:42] gigem: Locking the tuner would work. What I'd want to go with it is an absolutely impossible to miss notification stating so and a way to unlcok the tuner from the frontend.
[22:22:15] gigem: stuarta: I thought it was so other things could use the device when MythTV wasn't. Also, V4L devices are always on demand. What makes DVB special?
[22:22:31] jams: stuarta..that and it stopped other non-mythtv programs from accessing the tuner. And frankly sometimes dvb cards were not ready when mythbackend started.
[22:37:28] Tobbe5178 (Tobbe5178! has quit (Read error: Connection reset by peer)
[22:40:29] gary_buhrmaster: jams: While cards can take some time to become ready, with systemd as your init system you can wait for a device to indicate it is ready before continuing startup, which mitigates against that particular issue (with all the usual caveats about some tuners are special).
[22:41:04] jams: sure but systemd didn't exist when that open was added
[22:41:56] jams: option was added
[22:48:56] gary_buhrmaster: Understood. Just saying that the past justifications may have a limited application to the current and future environments. Otherwise you might need to keep XV support in the code forever :-)
[23:14:11] dekarl: hmm, thinking about my last fix... @American EIT users, do you care about multiple spaces between words? Just trimming instead of simplifying might be better for EIT/ETT but I prefer actual simplifying for the channel name wrt . . . 936d459ebd9d
[23:27:30] amessina (amessina!~amessina@unaffiliated/amessina) has joined #mythtv
[23:27:41] gary_buhrmaster: dekarl: I suspect there are so few @American EIT users that you are going to get almost no answers. Given the limited US OTA EIT data (around 12 hours), other than testing, you use Schedules Direct (or alternative) and are done with it.
[23:53:02] andreaz (andreaz! has quit (Read error: Connection reset by peer)

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