:: #mythtv

Daily chat history

Current users (77):

aloril, amessina, andreaz, Anssi, bhaak, blahdodo, brfransen, Captain_Murdoch, ChanServ, Chutt, clever, coling_, CyberJacob, dblain_, dekarl, dmfrey, eharris, ElmerFudd, enyc, gary_buhrmaster, ghoti, gigem, gregbert, gregL, GreyFoxx_, Guest41682, Hydr0p0nX, ikevin, J-e-f-f-A, jab416171, jams, jmcentee_, jnylen, joki, jpabq, jpharvey, jst_, jya, kc, knightr, KungFuJesus, lautriv_, libsci, markspieth, MythBuild, MythLogBot, nephyrin`, peper03, poptix, pppingme, raven42, rich0, saaki, Seeker`, seld, ShapeShifter499, sphery, SpootDev, sraue_, stefan__1, stuarta, stuartm, suffice, superm1, taylorr, tgm4883, Tobbe5178, tonsofpcs, tris, unforgiven512, vincent43, wagnerrp, Warped, XDS2010, xris, _charly_, _iwc
Sunday, February 7th, 2016, 00:21 UTC
[00:21:33] GreyFoxx_ (GreyFoxx_! has joined #mythtv
[00:22:01] GreyFoxx (GreyFoxx! has quit (Ping timeout: 240 seconds)
[00:22:56] knightr (knightr!~knightr@mythtv/developer/knightr) has quit (Ping timeout: 240 seconds)
[00:22:56] natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has quit (Ping timeout: 240 seconds)
[00:23:24] natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has joined #mythtv
[00:23:40] knightr (knightr!~knightr@mythtv/developer/knightr) has joined #mythtv
[00:26:57] dmfrey (dmfrey! has joined #mythtv
[00:39:44] natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has quit (Ping timeout: 276 seconds)
[00:57:07] Roklobotomy (Roklobotomy! has quit (Remote host closed the connection)
[01:16:49] gary_buhrmaster: dekarl: re: missing spaces between text snippets. This section of code at line 1545 of diseqcsettings.cpp looks suspicious to me (no space between "...FROM capturecard" and "WHERE...")
[01:25:27] Roklobotomy (Roklobotomy! has joined #mythtv
[01:34:15] blah_ (blah_! has joined #mythtv
[01:36:56] Roklobotomy (Roklobotomy! has quit (Ping timeout: 240 seconds)
[01:37:12] knightr: dekarl, this could be fixed as we are allowed to fix problems and this is clearly one but like you said it's difficult to know what would be the proper fix.
[01:38:24] knightr: I wonder if an tuner card I have could give me access to those menus so I could see what it looks like, I will keep you posted...
[01:38:51] knightr: (I can afford some downtime on my backend so I could test this...)
[01:39:04] knightr: gary_buhrmaster, ouch, that's clearly wrong...
[01:43:34] andreaz (andreaz! has quit (Read error: Connection reset by peer)
[01:49:02] knightr: easy enough to fix but this should definitely cause problems and apparently didn't so fixing it might actually cause problems...
[01:52:47] gary_buhrmaster: I think there has previously been suggestions/requests that there be a "test mule" out there somewhere with a diseqc configuration for testing. I do not recall any offers, which clearly makes a lot of this guesswork.
[01:54:45] knightr: gary_buhrmaster, that code doesn't seem to be used...
[01:55:00] knightr: I found it in convert_diseqc_db()
[01:55:09] knightr: which doesn't seem to be called from anywhere...
[01:55:47] knightr: still we might have this problem elsewhere in code which is actually used...
[01:58:09] gary_buhrmaster: Great, then it (would seem) it should be deleted (I will admit I did not look at the code or where/if it was called, I just saw it and said "Hmmmmm").
[01:58:43] knightr: I won't make the call on this, maybe it's just a work in progress...
[02:00:41] knightr: the comments seem to suggest this is some sort of legacy support...
[02:01:17] knightr: I wonder if we could ask one of the tools we use to report what appears to be dead code...
[02:02:32] knightr: This line is interesting
[02:02:32] knightr: mythtv/libs/libmythtv/dbcheck.cpp:#include "diseqcsettings.h" // for convert_diseqc_db()
[02:03:16] knightr: apparently dbcheck was making use of it, most likely to convert from one format to another...
[02:05:06] knightr: dbcheck was cleaned up years ago IIRC to remove support from upgrading from very old databases, maybe that's when it became dead code...
[02:06:40] gary_buhrmaster: That particular (set of) lines of code in diseqcsettings was added by danielk back in 2006.
[02:07:27] knightr: gary_buhrmaster, how do you track those down? I was wondering about this recently...
[02:07:58] gary_buhrmaster: github, "blame" is usually what I use.
[02:08:19] knightr: thank you! I will take a look at it...
[02:09:38] gary_buhrmaster: (github, select project, navigate to the code in question, and you have the viewing option "blame". It is not always useful (especially when the code has been shuffled), but sometimes it give one what one wants.
[02:09:42] gary_buhrmaster: )
[02:10:05] gary_buhrmaster: (missing right parens bother me from my (very old) LISP days)
[02:10:21] knightr: (-; (-; (-;
[02:17:11] knightr: gary_buhrmaster, the call was removed by 4753c7af83821f352862c8a38e345a5bbe044bdd
[02:17:25] knightr: in 2012...
[02:18:10] knightr: Require a reasonably current DB version on upgrade
[02:18:10] knightr: This removes support for upgrading the database from very old versions
[02:18:11] knightr: of MythTV (prior to 0.22).
[02:19:55] knightr: (so my hunch about it having been removed by that cleanup earlier was right...)
[02:20:25] Warped (Warped!~Warped@unaffiliated/warped) has quit (Quit: ChatZilla 0.9.92 [Firefox 44.0/20160123151951])
[02:23:09] gary_buhrmaster: Might be time to run cppcheck without suppressing the unused function list, and then having no mercy at the start of 0.29 regarding removing the chaff. But it is not my call.
[02:23:36] knightr: if cppcheck can do that
[02:24:01] knightr: than by all mean we should at least have it run with this enabled...
[02:24:09] gary_buhrmaster: It can, but I think (I would have to go back to check) that while the checks are enabled, those reports are suppressed.
[02:25:40] knightr: I don't consider that I have the autority to do this kind of cleanup
[02:27:22] knightr: but if it had been reported at work I would remove that function/method as this now appears to be old stuff which should be cleaned up...
[02:27:51] gary_buhrmaster: Code janitorial work has little positive reinforcement.
[02:29:04] knightr: I don't mind being the one who cleans up but as was seen for so long as someone who can't code and I am sure some people still think it's the case...
[02:29:37] knightr: at work I have the autority to do this kind of cleanup, not here..
[02:33:07] gary_buhrmaster: Seems to be unnecessarily limiting, but whom am I to say?
[02:33:34] gary_buhrmaster: (I am just a lurker here, with sometimes a loud opinion)
[02:48:57] knightr: gary_buhrmaster, if someone wants to complain, they can go ahead, I have just pushed this:
[02:54:07] Warped (Warped!~Warped@unaffiliated/warped) has joined #mythtv
[02:55:24] knightr: Cute, it looks like there's a problem with the buildbot(s)...
[02:55:25] knightr: remote: git_buildbot: ERROR: Could not connect to localhost:9989: Connection was refused by other side: 111: Connection refused.
[02:55:51] knightr: either that or it doesn't like me...
[02:56:13] knightr: (if it somehow tries to use my credentials...)
[02:59:21] arescorpio (arescorpio! has joined #mythtv
[03:14:38] gary_buhrmaster: (the buildbots have been offline for a bit, I think. dekarl poked stuarta about 18 hours ago regarding some parts of the build infrastructure, I think. I presume things will get looked at in good time.)
[03:15:55] gary_buhrmaster: !seen MythBuild
[03:15:55] MythLogBot: MythBuild was last seen 20 hours 26 minutes 24 seconds ago
[03:16:35] gary_buhrmaster: It would appear that things may have been in a "challenged" state for some time now.
[03:22:10] gary_buhrmaster: (perhaps the project needs to have dev/net/sys/secops in multiple disjoint timezones, but that can have its own challenges. Maybe alcor has thrown a fit again, I think (at least some) of the builder infrastructure is still there.)
[03:24:01] knightr: gary_buhrmaster, thank you!
[03:25:33] knightr: for security reasons the number of people with those kind of accesses is very limited to only a handful of people...
[03:25:50] knightr: the principle is sound...
[03:27:05] knightr: it would need to be taken a step further though, since those select few have all the keys if those people get hacked we are in trouble..
[03:29:04] knightr: by the way, I tried blame on a problem Karl told me about and like you said sometimes it can get confused...
[03:29:45] gary_buhrmaster: The principal is sound. I enforced it appropriately in a previous engagement. But the principal (at that engagement) also involved the entire CIA eval, and availability is in there too.
[03:29:46] knightr: I could not track down who had worked on what I wondered about but it looks like it could have been one of the Stuart's...
[03:31:22] gary_buhrmaster: (and I presume the long term plan would be to use TFA for the infrastructure services, and for the github account to reduce the impact of credential losses).
[03:32:41] knightr: what is not sound is that all the accesses are concentrated in those select few...
[03:33:03] gary_buhrmaster: You can (sometimes) checkout to a commit before all the refactor/renames occurred, and blame all over again. But it can get tedious.
[03:34:34] knightr: yikes...
[03:34:49] knightr: it did give me an idea of who worked in those sections
[03:35:05] gary_buhrmaster: Every project has their own ways. I might agree, I might disagree, but it is the project membership that ultimately have to decide.
[03:35:29] gary_buhrmaster: Sometimes you do not want to know the history ?
[05:08:20] joki (joki! has quit (Ping timeout: 248 seconds)
[05:11:25] blah_ is now known as Roklobsta
[05:11:32] Roklobsta is now known as Roklobotomy
[05:14:09] joki (joki! has joined #mythtv
[05:36:54] arescorpio (arescorpio! has quit (Quit: Leaving.)
[06:19:26] knightr (knightr!~knightr@mythtv/developer/knightr) has quit (Ping timeout: 272 seconds)
[07:29:00] SteveGoodey (SteveGoodey! has joined #mythtv
[09:43:41] paul-h (paul-h!~Paul@ has joined #mythtv
[10:03:42] dekarl1 (dekarl1!~dekarl@mythtv/developer/dekarl) has joined #mythtv
[10:07:11] dekarl (dekarl!~dekarl@mythtv/developer/dekarl) has quit (Ping timeout: 276 seconds)
[10:27:06] dekarl1 is now known as dekarl
[10:51:22] dekarl: Anybody mind to write a mythutil --clearqtcache? the crashes due to old QT4 caches killing QT5 programs are nasty
[11:12:42] Roklobotomy (Roklobotomy! has quit (Remote host closed the connection)
[11:50:57] dekarl: knightr, done
[11:51:39] dekarl: the help text is hard to understand anyway as its an ugly hack to avoid a schema change
[12:06:08] lautriv_ (lautriv_!~lautriv@funtoo/user/lautriv) has quit (Ping timeout: 276 seconds)
[12:18:45] lautriv_ (lautriv_!~lautriv@funtoo/user/lautriv) has joined #mythtv
[12:36:37] knightr (knightr!~knightr@mythtv/developer/knightr) has joined #mythtv
[13:10:50] MythBuild (MythBuild! has joined #mythtv
[13:11:01] stuarta: dunno what happened to the build master, restarted it
[13:16:49] stuarta: bunch of unhandled exceptions, but that didn't kill it, logs just stop :-/
[13:16:58] stuarta: ah well it'll be upgraded soon enough
[13:17:02] stuarta: when it moves over
[13:21:40] ShapeShifter499 (ShapeShifter499!~ShapeShif@unaffiliated/shapeshifter499) has quit (Quit: Leaving)
[13:54:43] ShapeShifter499 (ShapeShifter499!~ShapeShif@unaffiliated/shapeshifter499) has joined #mythtv
[14:08:56] ShapeShifter499 (ShapeShifter499!~ShapeShif@unaffiliated/shapeshifter499) has quit (Quit: Leaving)
[14:16:27] ShapeShifter499 (ShapeShifter499!~ShapeShif@unaffiliated/shapeshifter499) has joined #mythtv
[14:18:34] dekarl: no wonder it goes OOM suddenly
[15:09:26] jmcentee (jmcentee! has quit (Ping timeout: 276 seconds)
[15:10:33] jmcentee (jmcentee! has joined #mythtv
[15:23:58] dekarl: jheizer, you may want to update irqbalance... this was fixed almost 2 years ago and identified as a fix to cherry-pick one year ago. looks like no one cares in Debian/Ubuntu land :/
[15:26:20] dekarl: If someone wants to support RPI on Ubuntu <16.04 then look at . . . /comments/12
[15:31:59] jmcentee_ (jmcentee_! has joined #mythtv
[15:34:50] jmcentee (jmcentee! has quit (Ping timeout: 260 seconds)
[15:41:21] natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has joined #mythtv
[15:55:29] enyc: tgm4883: hrrrrrrrm was experimenting with clearer-head trying to reproduce packanging glitches and discovered a different one lol
[15:57:44] enyc: tgm4883: _if_ running debian 8 64bit under sysvinit rather than systemd, when doing dpkg -P libmyth-0.27–0 mythtv-backend mythtv-backend-master mythtv-common mythtv-database mythtv-transcode-utils (i.e. all the myth packages installed) , it bombs out because trying to delete user mythtv still in use by running PID of mythbackend process.... BUT if ran dpkg -P command again it completes fine =). I can
[15:57:50] enyc: only guess the /etc/init.d/mythtv-backend sysvinit-style-script isn't "waiting" for process to close before returning, creating a race-condition =).
[15:58:32] enyc: however, this seems to be unrelated to the other problem that I'm trying to clearly document in a few hours when i get back
[16:07:03] ** enyc glares at ran-sewather-map and changel plan =) **
[16:37:28] tgm4883: enyc: I feel like I should care, but really I just want to say "use a modern init system"
[16:37:53] enyc: tgm4883: indeed, i'm just trying ot recreate usecases/errors ;p . i think i have a ptach for that one anyway...
[16:38:21] tgm4883: good, cause there was no way I was going through the trouble to reproduce that locally and figure out a patch
[16:38:41] enyc: =)
[16:38:46] enyc: i was trying to reproduce the OTHER bug ;p
[16:38:47] enyc: lol
[16:39:01] enyc: but 'when you find a problem fix it anyway' generally seems to be sensible (most cases)
[16:40:24] ** enyc wonders where the upstart init script is **
[16:40:49] enyc: /etc/init/ conf file i guess
[16:44:47] enyc: tgm4883: once mythbackend send normal TERM signal, it seems it can simply take a while to stop
[16:46:28] enyc: tgm4883: simply changing /etc/init.d/mythtv-backend and changing "--stop" for "--stop --retry=0/30/KILL/5" (2 places) fixes that anyway... useful on older/upgraded (but supported) debian servers.... are you happy to just do that? I can TRY to resurrect my github connectivity
[16:46:54] enyc: actually its 4 places but you get idea
[16:49:00] ** enyc mutters about illusive bugs **
[17:08:26] MythBuild: build #41 of master-ubuntu-lts-armhf is complete: Success [3build successful] Build details are at . . . hf/builds/41
[18:23:15] jpabq (jpabq!~quassel@mythtv/developer/jpabq) has quit (Ping timeout: 240 seconds)
[18:26:17] jpabq (jpabq!~quassel@mythtv/developer/jpabq) has joined #mythtv
[18:45:48] enyc: oh well, left wookey (original author) of that a query ... will try to create a commit, rule that out and try to reproduce my first bug (its' now not happening for me either ;p )
[18:48:23] enyc: any more general thoughts on how long mythbackend CAN take to 'stop' ?
[19:06:05] stuarta: dekarl: heh, those ';' struck me as being not quite right too
[19:06:08] ** stuarta runs off again **
[19:23:57] Tobbe5178: enyc: my backend always stops in a few seconds
[19:34:16] dekarl: enyc: the upstart script grants it 30 seconds before killing it . . . kend.upstart
[19:34:46] dekarl: but whats the problem with simply waiting until the process is gone?
[19:45:14] Tobbe5178: not sure but there was one time i had other problems with one of the daemons on my linux box, and it was given infinite time to shutdown and that never happend
[19:45:24] Tobbe5178: so system got stuck late in the shutdown
[19:45:37] Tobbe5178: so had to press the powe button
[19:46:05] Tobbe5178: so maybe it is a safety feature?
[19:53:46] dekarl: yes, its a safety feature, both ways. So the trick is to find a sensible timeout to switch from "safe stopping of all services" to "do actually shutdown automatically eventually"
[19:55:25] andreaz (andreaz! has joined #mythtv
[20:27:33] Roklobotomy (Roklobotomy! has joined #mythtv
[20:40:16] Hydr0p0nX (Hydr0p0nX! has quit (Ping timeout: 250 seconds)
[20:51:25] enyc: dekarl: right, so sysvinit script grants 30 seconds before killing, makes sense too =)
[20:59:51] stuarta: wagnerrp: you see anything wrong with #12260? looks a trivial patch for 0.28
[21:00:19] stuarta: hmm, where's that bot
[21:02:53] stuarta: wagnerrp: and #12642
[21:02:53] ** MythLogBot **
[21:13:25] enyc: Am I right in saying MythTV still usies a custom/internal ffmpeg because upstream ffmpeg still aren't interested in the custom patches needed within mythtv?
[21:14:12] stuarta: we have patches that aren't upstreamed yes. at some point we should probably try to get them accepted again.
[21:20:23] enyc: okies =)
[21:22:38] stuarta: hmm, there's another error with the mythmusic conversion of mythweb to mysqli
[21:26:52] enyc: tgm4883: i'm suspecting your patches to fixes/0.27 r.e. NEWMYSQLCONFDIR etc. may have worked around the first issue i was having that I can't seem to reproduce in a hurry ;p
[21:31:28] Hydr0p0nX (Hydr0p0nX! has joined #mythtv
[21:52:14] Roklobotomy (Roklobotomy! has quit (Quit: Leaving)
[21:52:34] Roklobotomy (Roklobotomy! has joined #mythtv
[21:52:34] Roklobotomy (Roklobotomy! has quit (Client Quit)
[21:55:04] gigem: jpabq: I've been looking at #12624. It seems TVRec::StartRecording() bypasses the TVRec::StartedRecording() call there and calls it elsewhere when live TV needs to be stopped. That causes the scheduler to not get the actual recorded.starttime and recorded.recordedid returned to it and leads to the problem reported. Can you take a look and see if there's an easy way to get that information back to the
[21:55:04] ** MythLogBot **
[21:55:05] gigem: scheduler? I'm starting to think we might need to pass the starttime and recordingid back asynchronously.
[22:11:59] dekarl: stuarta, re#12642, #12529 I had something like that running localy and wondered why we are not using chanid+starttime instead of throwing the full serialized text on the wire.
[22:11:59] ** MythLogBot **
[22:12:20] dekarl: goes on with #12260
[22:12:20] ** MythLogBot **
[22:13:01] SteveGoodey (SteveGoodey! has quit (Quit: Konversation terminated!)
[22:30:47] stuarta: :shipit: ??
[22:37:55] ** stuarta cries **
[22:38:19] stuarta: icon lookup, missing 253 icons, of which none of the guesses will be anything close, and we never implemented "skip all"
[22:41:00] dekarl: its not in there? . . . ;view=markup
[22:41:54] stuarta: i have no idea, i don't use xmltv
[22:42:24] dekarl: i thought our icon service ingested nick's uk icon map
[22:42:32] stuarta: nope
[22:42:40] stuarta: it clearly needs to
[22:43:18] stuarta: we ingest lyngsat, then stir the icon name around in a pot, read some tea leaves, and guess the callsign from that
[22:43:42] dekarl: hint, nick has dvb triple to lyngsat icon mapd and other nice things . . . ;view=markup
[22:44:05] stuarta: do the other xmltv grabbers have the same thing?
[22:44:40] dekarl: no, thats where our service could step in
[22:44:44] stuarta: he's even got proper BAT stuff there
[22:48:05] stuarta: seriously, our icon search is on drugs. Q: BEN? A: syfy benelux.... really????
[23:04:48] dekarl: try "TV 5" ;)
[23:05:15] stuarta: plenty of work to be doing that's for sure
[23:07:43] stuarta: who watches some of this crap that's on these satelites?
[23:09:18] dekarl: just remember, its all for profit services. so either satellite bandwidth is dirt cheap. or there are lots of consumers watching it
[23:11:14] stuarta: and the lag, x-forwarding over ssh to the backend in another room is painful
[23:11:51] stuarta: channel icon choosing is a natural fit for the backend webserver
[23:11:53] dekarl: no crashes?
[23:12:07] stuarta: no, just bloody slow
[23:12:22] stuarta: not helped by the inconsistent movement around screens with the arrow keys
[23:12:36] dekarl: i get nonstop crashes when the backend isn't found, or with the opengl1 painter in the theme chooser etc
[23:12:54] stuarta: so you actually have to *wait* for it to catch up before you can select it
[23:13:35] stuarta: sigh, so many itv channels
[23:21:15] paul-h (paul-h!~Paul@ has quit (Quit: Konversation terminated!)
[23:22:56] natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has quit (Ping timeout: 240 seconds)
[23:37:05] enyc: created pull req
[23:37:22] enyc: is it possible to request something pulled into fixes/0.27 and master at same time ??
[23:37:43] enyc: [see packaging]
[23:37:59] stuarta: if you create a ticket in trac, with the pull reference, and state it's good for 0.27 and master
[23:38:12] enyc: not tried trac at all iirc
[23:38:23] enyc: is there some way i can just create second pull req easily and be done with it?
[23:39:23] dekarl: "just" checkout, branch, cherry-pick, push, send pull request
[23:39:45] dekarl: might be easier for the packager just to directly cherry-pick it
[23:40:01] stuarta: will be imho
[23:40:41] enyc: put explicit comemnt in the pull req...
[23:40:49] enyc: tgm4883: see packaging repo pull req?
[23:44:04] Hydr0p0nX (Hydr0p0nX! has quit (Ping timeout: 248 seconds)
[23:50:11] enyc: starting to get an idea what i'm doing =)).
[23:57:08] Hydr0p0nX (Hydr0p0nX! has joined #mythtv

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