| 00:00 | <poptix> | IMO, the net has been made too user friendly |
| 00:00 | <poptix> | it's like handing out a drivers license in a box of cereal |
| 00:02 | <bigguy> | I've been using linux for a long time now I just don't like the old pine type email clients |
| 00:02 | <bigguy> | pine mutt elm |
| 00:02 | <bigguy> | etc |
| 00:05 | <thor_> | Check out http://www.sigvaldason.com/teaser.png |
| 00:06 | -!- | paperclip_ is now known as paperclip |
| 00:06 | <bigguy> | thor_: looks nice |
| 00:06 | <thor_> | getting there |
| 00:06 | <thor_> | and that's code |
| 00:06 | <bigguy> | but the fonts could use some work ;) |
| 00:06 | <thor_> | (not a mockup) |
| 00:06 | <thor_> | my dev machine has crappy fonts (slackware) |
| 00:07 | <bigguy> | not hard to fix |
| 00:08 | <thor_> | Chutt, I figured out the post-flicker updating problem |
| 00:08 | <thor_> | (the problem was that I'm an idiot) |
| 00:09 | -!- | Timon [~dan@157-11-237-24.gci.net] has joined #MythTV |
| 00:09 | <Chutt> | heh |
| 00:09 | <thor_> | forgot to set draw order |
| 00:09 | <Chutt> | ah |
| 00:09 | <bline> | wow, that looks nice |
| 00:09 | <thor_> | so sometimes by chance, they'd end up with the right int |
| 00:09 | <Chutt> | thor, one request, if it's not too hard |
| 00:09 | <thor_> | duh! |
| 00:09 | <thor_> | yup |
| 00:09 | <Chutt> | i'd like a way to just display the active playlist |
| 00:10 | <Chutt> | and not the tree |
| 00:10 | <thor_> | ok |
| 00:10 | <Chutt> | configurable, or whatnot |
| 00:10 | <thor_> | if you don't have accelerators on, you can't navigate the tree anyway |
| 00:10 | <Chutt> | ah |
| 00:10 | <thor_> | arrows do buttons |
| 00:10 | <Chutt> | well, i just want more room =) |
| 00:10 | <thor_> | ok |
| 00:10 | <thor_> | user setting |
| 00:11 | <Chutt> | if it's not too much trouble |
| 00:11 | <Chutt> | but, that looks nice |
| 00:11 | <thor_> | no prob ... and it 3 months time someone on the -user list will say "could someone add a feature to let me change lists from the playing screen" |
| 00:11 | <thor_> | heh |
| 00:13 | <Chutt> | big improvement over what you sent me last night =) |
| 00:13 | <thor_> | oh yeah (that's why mu butt fell asleep) |
| 00:13 | <thor_> | been glued here all day |
| 00:13 | <Chutt> | heh |
| 00:13 | <thor_> | time for a walk around the block I think |
| 00:14 | <poptix> | Chutt: try hitting escape while fast forwarding through a recording |
| 00:15 | <Chutt> | works fine |
| 00:15 | <poptix> | hmm |
| 00:15 | <Chutt> | hang for you? |
| 00:15 | <poptix> | mine froze at the select recording screen, without updating it |
| 00:15 | <poptix> | the overlay window disappeared though |
| 00:15 | <Chutt> | think you could make it do that in gdb? |
| 00:16 | <poptix> | i can try |
| 00:16 | <poptix> | hold. |
| 00:16 | <Chutt> | then hit ctl-c and get a bt of it all |
| 00:16 | <Chutt> | needs to be compiled for debugging, of course |
| 00:16 | <poptix> | looks like it's reproducable |
| 00:17 | <poptix> | but only if i fast forward to the end-ish of the recording |
| 00:17 | * poptix | recompile w/ debug |
| 00:17 | -!- | hfb [~hfb@pool0304.cvx38-bradley.dialup.earthlink.net] has joined #mythtv |
| 00:19 | <Timon> | inman: You here? |
| 00:25 | <thor_> | Chutt, you want just a settings value or a live button option? |
| 00:25 | <Chutt> | a setting is fine |
| 00:25 | <Chutt> | but however you want to do it |
| 00:25 | <thor_> | one nice thing would be right arrow press on track area makes it full width |
| 00:26 | <thor_> | hmmm |
| 00:26 | -!- | rcaskey [] has quit [Read error: 110 (Connection timed out)] |
| 00:27 | -!- | justin [~justin@ool-18b81fb6.dyn.optonline.net] has joined #mythtv |
| 00:38 | <Captain_Murdoch> | Chutt: not sure if that usepre commit I just did fixed the segfault that rkulagow was seeing with the commDetect destructor, but I've recorded 7 shows in a row so far and haven't been able to get it to show up yet since I applied that change to my source. |
| 00:38 | <Chutt> | odd |
| 00:38 | <Chutt> | but, sure =) |
| 00:39 | <Captain_Murdoch> | haven't looked into the buffer stuff enough to know but valgrind was complaining about uninitialized variable before and if usepre was negative??? I'm going to keep recording stuff all night long probably to see if I can make it popup. |
| 00:40 | <Chutt> | ah |
| 00:42 | <Captain_Murdoch> | probably unrelated. I don't see how the prebuffering var staying true or false could cause that if usepre was negative or really high. |
| 00:47 | -!- | Timon [] has quit ["Client Exiting"] |
| 00:57 | <Chutt> | oops |
| 00:58 | <thor_> | ? |
| 00:58 | <Chutt> | bug in the commits script ate my commit message =) |
| 00:58 | <thor_> | seen that before |
| 00:58 | <Chutt> | you and i committed at pretty much the same time |
| 00:58 | <Chutt> | think it's fixed |
| 00:59 | <thor_> | I'm only in mythlib |
| 00:59 | <Chutt> | i was just doing a libavcodec update |
| 00:59 | <thor_> | always worried on stomping on something moegreen's working on |
| 01:00 | <thor_> | s/on/am/ |
| 01:01 | -!- | Sembiance [sembiance@68.168.72.164] has joined #mythtv |
| 01:01 | <Sembiance> | I'm trying to find a place to buy a WinTV PVR-250 card, and found this one: http://www.buy.com/retail/product.asp?sku=10322340&loc=101 |
| 01:01 | <moegreen> | thor_: heh - i'm working on mythvideo stuff - the ui code i have is pretty much done for now |
| 01:01 | <Sembiance> | but it says it's PAL, I'm in the USA don't I need NTSC? |
| 01:02 | <thor_> | moegreen, good . You see http://www.sigvaldason.com/teaser.png ? |
| 01:02 | <moegreen> | thor_: yeah, looks nice |
| 01:02 | <thor_> | getin there, but not as fast as you would have =) |
| 01:04 | <thor_> | Sembiance, you want NTSC if you're in NA |
| 01:05 | <Sembiance> | thor: okay, thought so. thanks :) |
| 01:05 | <Sembiance> | thor: If it doesn't say, do I assume NTSC? Amazon.com has the PVR 250 for $128 and it doesn't say whether it's PAL or NTSC |
| 01:05 | <Captain_Murdoch> | moegreen: I think if I replace that cursordown(false) I mentioned with "if (listCount && curShowing >= listCount) curShowing = listCount - 1;" that will fix it. the curShowing needs to be reset in the case where we delete the bottom item but in all other cases it should just remain the same (which causes us to highlight the next item in the list) |
| 01:06 | <thor_> | Sembiance, no guarantees, but it would be very odd for Amazon to ship you a non-NA device in NA without you specifically asking for it |
| 01:08 | <Captain_Murdoch> | moegreen: that appears to do the trick. if I delete the last item the new last item is highlighted. if I delete in the middle or the first then the highlight stays in the same position and highlights the item that was next but just got moved up. |
| 01:09 | <moegreen> | Captain_Murdoch: sounds good - want to go ahead and commit that then? |
| 01:09 | <moegreen> | http://www.untzuntz.com/mythtv/testing/video1.png |
| 01:10 | <thor_> | moegreen, is that mythvideo? |
| 01:10 | <moegreen> | thor_: yeah |
| 01:10 | <thor_> | imbd grabs off the filename? |
| 01:10 | <moegreen> | data is grabbed from imdb |
| 01:10 | <Captain_Murdoch> | yeah, will do it in the morning after one more round of delete testing. :) I'm trying to track down another bug right now and have 9 shows setup to record. in the morning I'll test delete all of them (if my bug doesn't show up and kill the backend) and then commit if the delete works ok for those tests. |
| 01:11 | <moegreen> | thor_: well, I had to do some trickery, but if the filename has a decent name it should work fine |
| 01:11 | <thor_> | righteous |
| 01:11 | <moegreen> | I've got it cutting out certain things to get a rather general name then do a fuzzy search on imdb to get a list - if there is just one (as in the the case shown) it just grabs that page |
| 01:12 | <thor_> | this sit in the database, or does it pull as you scroll |
| 01:12 | <moegreen> | thor_: database |
| 01:13 | <Captain_Murdoch> | moegreen: nice screenshot. |
| 01:13 | <moegreen> | thor_: this is slightly different than the typical mythvideo, I've got several windows options now (like in mythmusic) this part is the 'Video Manager' section to manage the information and what not |
| 01:13 | <moegreen> | if there are multiple results returned from imdb the user will have to select from a list |
| 01:14 | <thor_> | sounds like a major improvement ... |
| 01:14 | <moegreen> | Once I get this all working I'll add the cover art/movie poster and have it download that :) |
| 01:14 | <thor_> | my better half is going to be even more confused about the difference between Video-> and TV->Watch Recordings |
| 01:15 | <moegreen> | I guess either could be integrated into the other at some point |
| 01:15 | <moegreen> | well, at least one way :) |
| 01:16 | <thor_> | and then -user list people will be even more confused about mplayer =) |
| 01:17 | <moegreen> | heh, yeah - aparently MythTV uses mplayer for EVERYTHING :) (even the menus) |
| 01:17 | <Chutt> | the ideas of having a the directory be traversable by something like the tv playback box was pretty good |
| 01:17 | <Chutt> | idea |
| 01:18 | <thor_> | would simplify my life |
| 01:19 | <Chutt> | directories on the left, files on the right |
| 01:19 | <thor_> | nah, whole navigable tree (automagically!) |
| 01:19 | <Chutt> | naw |
| 01:19 | <Chutt> | heh |
| 01:20 | <Chutt> | why can't people spell my name right |
| 01:20 | <Chutt> | it's not that difficult. |
| 01:20 | <thor_> | try a week in my shoes |
| 01:20 | <Chutt> | heh |
| 01:25 | <moegreen> | Chutt: QT 3.0's QHttp object eats any HTTP errors (like 302 - location moved), it was driving me crazy until I figured that out |
| 01:25 | <Chutt> | heh |
| 01:25 | <Chutt> | fixed in 3.1, or? |
| 01:26 | <moegreen> | Chutt: they added a QHttpResponseHeader or something like that |
| 01:26 | <Chutt> | ah |
| 01:30 | <Chutt> | moegreen, any estimates on when you think that stuff'll be ready to go into cvs? |
| 01:30 | <Chutt> | just curious =) |
| 01:33 | -!- | hfb [] has quit ["Client exiting"] |
| 01:33 | <moegreen> | Chutt: It's moving along now, but I've still got the setup to do (the video browser and video listing I want to add should be easy) - a few days maybe |
| 01:34 | <Chutt> | cool |
| 01:34 | <moegreen> | Chutt: Are you trying to get this in a .9 release? :) |
| 01:34 | <Sembiance> | What's a good 'video-out' card to use in a MythTV box? |
| 01:34 | <Chutt> | potentially |
| 01:34 | <Sembiance> | something that's nice and cheap and gives me standard RCA output |
| 01:35 | <Chutt> | i've been really busy lately, so i keep pushing stuff back =) |
| 01:35 | <moegreen> | Chutt: have you seen nziarek around lately? |
| 01:35 | <Chutt> | nope |
| 01:35 | <Chutt> | hasn't been on here |
| 01:37 | <Sembiance> | or should I use a VGA-to-TV adapter? |
| 01:37 | <moegreen> | Sembiance: Anything that works with linux should be fine - though overscan/resizing control is helpful |
| 01:38 | <Sembiance> | moegreen: What do you use? |
| 01:38 | <moegreen> | Sembiance: Matrox G400 |
| 01:39 | <Sembiance> | ooh expensive :) |
| 01:39 | <moegreen> | Sembiance: Dual-Headed $30 on ebay :) |
| 01:40 | <Sembiance> | woah |
| 01:40 | <Sembiance> | buy.com has it listed at $170 :) |
| 01:40 | <Sembiance> | http://www.buy.com/retail/product.asp?sku=10286878&loc=101&queryType=comp |
| 01:41 | <Sembiance> | moegreen: that card just has VGA out.... |
| 01:42 | <moegreen> | Sembiance: the card I have has two heads, the first is for a monitor, the second can be another monitor, digial for lcd, or TV out (with a $10 dongle adapter) |
| 01:43 | <Sembiance> | hrm... |
| 01:46 | <Sembiance> | moegreen: Where would one get a VGA to RCA/Composite dongle adapter for $10? |
| 01:48 | <moegreen> | I think I just got mine at matrox's used/refurb online store (I think it was 10 or maybe 20 there), I'd guess eBay has some cheaper ones |
| 01:48 | <moegreen> | It's got RCA and S-Video |
| 01:48 | <Sembiance> | hrm..... |
| 01:51 | <Sembiance> | I think I'll go with a GeForce4 MX |
| 01:51 | <Sembiance> | It's only $40 and has the RCA and SVideo ports on the card instead of needing a dongle |
| 01:55 | -!- | Chutt [] has quit [Read error: 110 (Connection timed out)] |
| 01:58 | -!- | jrh [asdf@ip68-5-230-73.oc.oc.cox.net] has joined #mythtv |
| 02:38 | -!- | Chutt [~ijr@dsl093-011-148.cle1.dsl.speakeasy.net] has joined #mythtv |
| 02:42 | -!- | kiwi_uk [~kiwiuk@pc3-nfds1-6-cust201.nott.cable.ntl.com] has joined #mythtv |
| 02:44 | -!- | Sembiance [] has quit ["thanks"] |
| 02:51 | -!- | bline [] has quit ["Client exiting"] |
| 03:29 | <moegreen> | While it's not quite as polished as it will be, http://untzuntz.com/mythtv/testing/video3.png - shows the selection from imdb based on the title 'Airplane', video2.png in the same directory shows a list with all info grabbed from imdb |
| 03:30 | <FryGuy> | nice moegreen |
| 03:31 | <moegreen> | thanks, I already like it better than the old way :) |
| 03:32 | <moegreen> | Though that's not saying much |
| 03:32 | <FryGuy> | i need to get the latest version from cvs again.. I don't know how to update instead of checking out though |
| 03:33 | <moegreen> | cvs update [module] |
| 03:33 | <FryGuy> | when I'm at my dad's place I'll be able to spend a bit more time on mythtv I think |
| 03:33 | <FryGuy> | and learn how to develop in linux too |
| 03:34 | <moegreen> | sounds like a plan ... well, I'm out - movie poster/coverart tomorrow (and setup) |
| 03:34 | <FryGuy> | the most I've ever used linux to code was a coding competetion and I used vi and gcc :p |
| 03:36 | -!- | bline [~sbeck@24.84.93.233] has joined #mythtv |
| 03:36 | <bline> | that new commercial skipping is working very well for me |
| 04:11 | -!- | FryGuy [] has quit [Read error: 104 (Connection reset by peer)] |
| 04:11 | -!- | froggy19 [~fryguy@12-246-43-18.client.attbi.com] has joined #mythtv |
| 04:11 | -!- | froggy19 is now known as FryGuy |
| 04:46 | -!- | bigguy [] has quit [Read error: 110 (Connection timed out)] |
| 05:34 | -!- | Viddy [] has quit [Remote closed the connection] |
| 06:25 | -!- | Viddy [david@visp194-179.visp.co.nz] has joined #mythtv |
| 07:06 | -!- | choenig [~choenig@p50858FED.dip.t-dialin.net] has joined #mythtv |
| 08:45 | -!- | bline [] has quit ["Client exiting"] |
| 09:19 | <rkulagow_> | it seems that some people have a lot a lot of trouble getting the pvr250 to work with redhat... |
| 09:23 | -!- | poptix [] has quit [Read error: 110 (Connection timed out)] |
| 09:38 | -!- | poptix [poptix@precious.net] has joined #mythtv |
| 09:39 | -!- | mem16421 [~mark@pcp01115546pcs.flint01.mi.comcast.net] has joined #mythtv |
| 09:40 | <mem16421> | I got a newbie question, but I can't find a answer in english anywhere........ |
| 09:41 | <mem16421> | ....I get "_IceTransOpen: Unable to Parse address none" when i run setup. |
| 09:42 | <mem16421> | .....it also follows with QMYSQL3 driver not loaded errors. |
| 09:43 | <rkulagow_> | did you read the HOWTO? |
| 09:43 | <mem16421> | I did, but it doesn't reference this error. |
| 09:44 | <mem16421> | I have searched usenet, but the only people talking about it seems it be non-english posts. |
| 09:45 | <rkulagow_> | well, the fact that you're getting QMYSQL3 driver errors leads me to believe that you've skipped some steps, since that's covered in the HOWTO. you can start by fixing that, or at least confirming that you've done _everything_ correctly. what distribution are you using? |
| 09:46 | <mem16421> | RH9 |
| 09:47 | -!- | lofgren [~andre@h29n1fls33o954.telia.com] has joined #mythtv |
| 09:47 | <rkulagow_> | did you perform the steps detailed in section 4 of the HOWTO? |
| 09:48 | <mem16421> | setting up QTDIR, and PATH to point to /usr/lib/qt3, and /usr/lib/qt3/bin? yep |
| 09:50 | <rkulagow_> | and when you check them using the set commands, are they there? do a ls -l /usr/lib/qt3/plugins/sqldrivers/ |
| 09:51 | -!- | Captain_Murdoch_ [~cpinkham@linoleum.engrs.infi.net] has joined #mythtv |
| 09:53 | <mem16421> | The QTDIR and PATH are set, but there is no sqldrivers in /usr/lib/qt3/plugins |
| 09:55 | <rkulagow_> | well, that's part of your problem right there. load the qt3-mysql drivers. |
| 10:03 | -!- | hfb [~hfb@lsanca1-ar2-4-60-015-093.lsanca1.dsl-verizon.net] has joined #mythtv |
| 10:04 | <mem16421> | So I will need to compile qt from scratch to get that function? |
| 10:06 | <rkulagow_> | no. i think you should search the archives at gossamer-threads for clues on how to get your RH system operational. you're obviously not installing the correct RPM or something like that. |
| 10:06 | <rkulagow_> | or check rpmfind.net |
| 10:17 | <mem16421> | There is no qt3 package for mysql on rpmfind, just the whole package. According to trolltech you have to compile the mysql plugins. |
| 10:18 | <Chutt> | you do not have to compile it yourself. |
| 10:18 | <Chutt> | there is a rpm for the mysql plugin. |
| 10:22 | <mdz> | there may not be a section in the howto on it, but if you search any search engine for the QMYSQL3 error you got, you should find hundreds of answers |
| 10:29 | -!- | schultmc [~schultmc@zealot.progeny.com] has joined #mythtv |
| 10:36 | <mem16421> | Thanks.... I found the qt-MySQL package... just resolving the dependancies |
| 11:01 | -!- | lofgren [] has quit [Read error: 110 (Connection timed out)] |
| 11:12 | -!- | bigguy [bigman@h97.44.102.166.ip.alltel.net] has joined #mythtv |
| 11:13 | -!- | choenig [] has quit [Remote closed the connection] |
| 11:34 | -!- | Drikus_ [~Drikus@cc45940-a.deven1.ov.home.nl] has joined #mythtv |
| 11:45 | -!- | justin [] has quit [Read error: 110 (Connection timed out)] |
| 11:48 | <-- mem16421 | (~mark@pcp01115546pcs.flint01.mi.comcast.net) has left #mythtv |
| 12:08 | -!- | syah [punt@cs242711-25.austin.rr.com] has joined #mythtv |
| 12:43 | -!- | choenig [~choenig@p50858FED.dip.t-dialin.net] has joined #mythtv |
| 13:29 | -!- | paperclip [] has quit [brunner.freenode.net irc.freenode.net] |
| 13:32 | -!- | paperclip [~joe@ip68-11-30-158.no.no.cox.net] has joined #mythtv |
| 13:46 | -!- | L17dr [~mark@pcp01115546pcs.flint01.mi.comcast.net] has joined #mythtv |
| 13:47 | <L17dr> | Anyone got a ATI VE working with Myth? |
| 13:48 | <moegreen> | L17dr: the capture card? |
| 13:48 | <L17dr> | right |
| 13:48 | <moegreen> | yeah |
| 13:49 | <L17dr> | I got my line-up, and I am getting channels, but they are off by 6 and the video is crappy. |
| 13:49 | <moegreen> | L17dr: You've probably got the wrong tuner type defined when you load the bttv module |
| 13:50 | <moegreen> | L17dr: Are you North America? If so, try insmod bttv tuner=2 |
| 13:53 | <L17dr> | That did it!, thanks. |
| 13:57 | <L17dr> | I am running RH9, with Myth but when I shutdown mythfrontend, and mythbackend the audio wont quit |
| 14:00 | <moegreen> | You need to mute your line volume, see http://www.mythtv.org/docs/mythtv-HOWTO-18.html#ss18.11 |
| 14:01 | <L17dr> | thanks again.... I will leave everyone alone now :-) |
| 14:05 | -!- | L17dr [] has quit [Remote closed the connection] |
| 15:16 | -!- | syah [] has quit ["freebsd"] |
| 15:29 | -!- | kimmel [~mark@pcp01115546pcs.flint01.mi.comcast.net] has joined #mythtv |
| 15:29 | <kimmel> | Anyone getting "IceTransOpen: Unable to Parse address none" when starting mythbackend? |
| 15:54 | <kimmel> | Anyone getting "IceTransOpen: Unable to Parse address none" when starting mythbackend? |
| 15:55 | <moegreen> | no |
| 15:57 | -!- | choenig [] has quit [Remote closed the connection] |
| 15:59 | <kimmel> | alsamixer |
| 16:22 | <moegreen> | ? |
| 16:30 | <rkulagow_> | captain_murdoch: for the double/triple, etc reporting of NFS space by the backends, howabout if the master writes a magic value to the NFS store, and if the slave sees it it reports the free disk space as "0"? |
| 16:58 | -!- | kiwi_uk [] has quit ["Peace and Protection 4.22"] |
| 17:36 | <Captain_Murdoch_> | rkulagow: that would work but is kinda kludgey. could possibly check the f_type field from the statfs the backend does and see if it's nfs. if nfs and the storage dir is the same as the master then don't count. |
| 17:37 | <Captain_Murdoch_> | or could even check the total size to make sure the size of the master's dir is the size of the slave's dir as another check. |
| 17:38 | <Captain_Murdoch_> | or could just say don't count nfs storage unless you're the master backend. |
| 17:38 | <rkulagow_> | all excellent ideas. |
| 17:39 | <inman> | i commented on this before. you can't verify filesystem identity using nfs root inode numbers. |
| 17:40 | <Captain_Murdoch_> | inman, I said f_type. do a "man statfs" since that's what mythbackend uses to get free space. |
| 17:41 | <inman> | ftype doesn't tell you anything. |
| 17:41 | <Captain_Murdoch_> | is that what you're referring to? |
| 17:41 | <Captain_Murdoch_> | man says it tells me it's a nfs mounted volume. |
| 17:41 | <inman> | you can't depend on identical type/size/usage. usage could be zero for several filesystems. |
| 17:41 | <inman> | i'm sorry, maybe i'm misunderstanding the problem. :-) |
| 17:41 | <inman> | if you just want to see if it's nfs, then you've got the right technique. |
| 17:42 | <Captain_Murdoch_> | right now if you mount the same filesystem on 3 backends, your free space in the delete screen is 3x normal. |
| 17:42 | <inman> | i think the solution is to see if the filesystems are the same. |
| 17:42 | <Captain_Murdoch_> | so we're just trying to figure out a fairly accurate way to exclude counting that same fs 3 times. |
| 17:42 | <inman> | not to check and see if only one master is mounting it via nfs. |
| 17:42 | <inman> | all backends could be using nfs storage. |
| 17:43 | <Captain_Murdoch_> | yeah, that's why I said if anything matches the master then exclude it. |
| 17:44 | <inman> | the backends may be using different nfs filesystems. |
| 17:45 | <Captain_Murdoch_> | again I repeat. that's why I said if it matches the master. I said check the total to see if it matches and possibly the free space as well. |
| 17:45 | <inman> | i don't understand what "it matches the master" means. |
| 17:46 | <Captain_Murdoch_> | if the total space of a slave's storage dir matches the total of the master's dir. also can double-check the free space. if they match and the slave is nfs then it's reasonably safe to assume they're all the same directory. |
| 17:47 | <inman> | that sounds like a race-condition to me. |
| 17:47 | <inman> | why don't you just create a marker and then ask the other host to check for it? |
| 17:48 | <Captain_Murdoch_> | that was rkulagow's original idea |
| 17:49 | <inman> | sounds good to me, as long as the marker's filename is generated with mktemp() or something. |
| 17:50 | <inman> | note that some filesystems share disk space but not namespace. |
| 17:50 | <inman> | you can't really depend on the total-filesystem-size value. |
| 17:52 | <Captain_Murdoch_> | if the 2 values match and it's nfs then you can be probably 99% sure they're the same dir. the odds of someone having 2 or 3 backends with the exact same size drive with the exact same amount free on each is kinda miniscule |
| 17:53 | <inman> | it happens all the time when the drives are full or empty. |
| 17:53 | <Captain_Murdoch_> | and if I have one nfs server serving one filesystem but with 3 different directories mounted, then the problem still exists. |
| 17:53 | <Captain_Murdoch_> | and just when is a myth drive empty? :) |
| 17:54 | <inman> | i'm just pointing out that things like disk virtualization and real NAS servers will break the more naive approach. |
| 17:54 | <inman> | i'd do it right the first time so you don't have to redo it later. |
| 17:56 | <Captain_Murdoch_> | I think the odds of having multiple backends with the same exact total and free space as the master is kinda small in this scenario, but the whole point is moot anyway cause neither of us is coding the fix. :) |
| 18:11 | <poptix> | sigh |
| 18:11 | <poptix> | i'm having the same problem that one guy was |
| 18:12 | <poptix> | where, if you start watching a recording before it's done, the commercial cutlist isn't written |
| 18:12 | <poptix> | neither is the seektable |
| 18:12 | <Chutt> | track it down =) |
| 18:12 | <poptix> | rebuffering (135169 256000) |
| 18:12 | <poptix> | rebuffering (211969 256000) |
| 18:13 | <poptix> | rebuffering (157697 256000) |
| 18:13 | <poptix> | .. |
| 18:13 | <poptix> | no idea =) |
| 18:13 | <Chutt> | should be pretty easy to figure out why it's not writing the seek table |
| 18:13 | <poptix> | does it go back over the stream afterward? |
| 18:13 | <poptix> | or buffer it then write it at the end |
| 18:14 | <rkulagow_> | chutt: did you see the dialog above about slave not reporting their disk free if they're on NFS? do you have any insight to technique that you'd accept? |
| 18:14 | <Chutt> | which type of files? |
| 18:14 | <poptix> | Chutt: this is on my PVR-250 system |
| 18:14 | <Chutt> | rkulagow, i don't really care how it's done, i don't have that issue =) |
| 18:14 | <Chutt> | poptix, the seektable is written to the database at the end of the recording, then |
| 18:14 | <poptix> | hmm |
| 18:15 | <Chutt> | should be _really_ easy to figure out why/why not |
| 18:16 | -!- | Drikus_ [] has quit ["leaving"] |
| 18:17 | <poptix> | interesting |
| 18:17 | <poptix> | the data is there |
| 18:17 | <Chutt> | the commercial skip table doesn't get generated until the generation thread finished |
| 18:18 | <Chutt> | runs after the recording's complete, of course |
| 18:18 | <Captain_Murdoch_> | poptix: what data? |
| 18:18 | <Chutt> | it'll print out 'Position map found' for mpg recordings |
| 18:18 | <Chutt> | if it finds a seektable |
| 18:18 | <poptix> | i assume the 'recordedmarkup' table is the seektable |
| 18:19 | <Captain_Murdoch_> | type 3 = blank frame, 4 = comm start, 5 = comm end, 6 = gop |
| 18:19 | <poptix> | yes, or no? |
| 18:19 | <Captain_Murdoch_> | so after a pvr250 recording you'll have a bunch of 6's then after commercial flagging finishes you'll have a bunch of 3's and a few 4's & 5's |
| 18:20 | <poptix> | select * from recordedmarkup where chanid=1043 and starttime=20030522140000 and type='6'; |
| 18:20 | <poptix> | Empty set (0.00 sec) |
| 18:21 | <poptix> | 320 3's |
| 18:21 | <Captain_Murdoch_> | did the backend die? |
| 18:21 | <poptix> | nope |
| 18:21 | <poptix> | 5 4's and 5's |
| 18:21 | <Captain_Murdoch_> | so comm flagging ran evidently. |
| 18:22 | <Captain_Murdoch_> | just no seektable to benefit from the comm flagging? |
| 18:22 | <poptix> | guess not |
| 18:24 | <poptix> | tv shows that recorded while i wasn't around work fine, and have ~7100 6's per hour =p |
| 18:24 | <Captain_Murdoch_> | Chutt: you said it's ok to modify the recorders to write their seektable to the database every so often during record right? I've got that on my todo list unless someone beats me to it. |
| 18:25 | <Chutt> | yeah, that's fine |
| 18:25 | <Chutt> | as long as the nuppelvideo stuff still gets written to the file |
| 18:25 | <Captain_Murdoch_> | nuppel will still apend to file, but if it's not ther in the file it can use the table. |
| 18:25 | <Captain_Murdoch_> | yeah, won't touch the file part |
| 18:26 | <Chutt> | there's some stuff that assumes that the last entry in the table is the end of the file |
| 18:26 | <Chutt> | might have to fix that assumption |
| 18:26 | <Chutt> | might not |
| 18:26 | <Captain_Murdoch_> | ok, made a note of that. |
| 18:29 | <Captain_Murdoch_> | maybe this weekend, it's supposed to be rainy here. |
| 18:31 | <Captain_Murdoch_> | I'll probably make it so the comm flagging function checks whether it found a seektable or not when playing through the file. if no seektable then write out the positionmap to the db. |
| 18:32 | <Captain_Murdoch_> | that way old files w/o seektables can be fixed just by re-flagging with mythcommflag |
| 18:33 | <Chutt> | yup |
| 18:33 | <Captain_Murdoch_> | and the flagging always insures there's something to seek with. |
| 18:33 | <Chutt> | sounds good to me |
| 18:33 | <Captain_Murdoch_> | anyway dinner's calling. |
| 18:33 | <-- Captain_Murdoch_ | has quit () |
| 18:36 | -!- | schultmc [] has quit ["Client exiting"] |
| 19:43 | -!- | tstm [~tstm@dsl-hkigw4i16.dial.inet.fi] has joined #mythtv |
| 19:43 | <tstm> | Hi. =) |
| 19:45 | <tstm> | Has anyone here toyed with the MythTV DVB? |
| 19:56 | -!- | hfb [] has quit ["Client exiting"] |
| 20:14 | -!- | kimmel [] has quit ["using sirc version 2.211+KSIRC/1.2.4"] |
| 21:09 | <rkulagow> | chutt: mailing you a segfault. happened on the masterbackend; only two cards defined in the database. two programs were scheduled to start at the same time; it appears the segfault happened pretty close to the start time. synced with CVS. |
| 21:14 | -!- | bline [office-1@office.gossamer-threads.com] has joined #mythtv |
| 21:18 | -!- | choenig [~choenig@p50858FED.dip.t-dialin.net] has joined #mythtv |
| 21:42 | <Chutt> | rkulagow, still have the backend in gdb? |
| 21:44 | <rkulagow> | chutt: yes |
| 21:45 | <Chutt> | excellent |
| 21:45 | <Chutt> | can you switch to thread 15 |
| 21:46 | <rkulagow> | ok |
| 21:46 | <Chutt> | and move up to frame #2? |
| 21:46 | <rkulagow> | ok |
| 21:46 | <Chutt> | allright |
| 21:46 | <Chutt> | print s->f_code |
| 21:47 | <rkulagow> | 1 |
| 21:47 | <Chutt> | print s->me.mv_penalty |
| 21:47 | <rkulagow> | $2 = (uint8_t (*)[4097]) 0x0 |
| 21:48 | <Chutt> | interesting |
| 21:48 | <rkulagow> | interesting good, or interesting bad? |
| 21:48 | <rkulagow> | or interesting, "shouldn't happen" |
| 21:48 | <Chutt> | that's the problem |
| 21:49 | <Chutt> | can you go up to frame #3? |
| 21:50 | <Chutt> | and print default_mv_denalty |
| 21:50 | <rkulagow> | ok |
| 21:50 | <Chutt> | err |
| 21:50 | <Chutt> | default_mv_penalty |
| 21:50 | <rkulagow> | $3 = (uint8_t (*)[4097]) 0x85a1180 |
| 21:51 | <Chutt> | print mv_penalty |
| 21:51 | <rkulagow> | $4 = (uint8_t (*)[4097]) 0x0 |
| 21:52 | <Chutt> | hmmm |
| 21:53 | <Chutt> | pretty much died right away |
| 21:53 | <Chutt> | and this was the first recording? |
| 21:53 | <Chutt> | right? |
| 21:54 | <rkulagow> | there were two recordings; friends and csi. this was the first recording in this session after a fresh compile / make install. |
| 21:54 | <Chutt> | right |
| 21:54 | <Chutt> | hmm |
| 21:56 | <rkulagow> | brb |
| 22:00 | <Chutt> | fixed |
| 22:00 | <Chutt> | checked in to CVS |
| 22:00 | <Chutt> | thanks for helping track that down =) |
| 22:01 | <rkulagow> | NP. now all you need to do is write MythP2P so my wife and M-I-L don't kill me because they missed their shows. =) |
| 22:02 | <bigguy> | M-I-L? |
| 22:03 | -!- | choenig [] has quit [Remote closed the connection] |
| 22:03 | <rkulagow> | mother in law. |
| 22:04 | <bigguy> | ah |
| 22:04 | <bigguy> | she lives with you |
| 22:04 | <rkulagow> | no; visiting. |
| 22:04 | <bigguy> | ah |
| 22:05 | <bline> | mornin |
| 22:05 | <bigguy> | lo bline |
| 22:05 | <bigguy> | bline: you just getting up? |
| 22:06 | <rkulagow> | chutt: can i dump the gdb session? |
| 22:06 | <bline> | yeah, got my schedule all messed |
| 22:06 | <Chutt> | yup |
| 22:06 | <rkulagow> | ok |
| 22:06 | <Chutt> | shouldn't happen again |
| 22:07 | <rkulagow> | anything i can do to try to retrigger other than having two programs scheduled for the same time? |
| 22:08 | <Chutt> | that's the only way it can happen |
| 22:08 | <bigguy> | well I'll bbl gotta do some work |
| 22:08 | <Chutt> | and it would have been pretty difficult to reproduce anyway |
| 22:09 | <Chutt> | first time anything was recorded, two at the same time, and the scheduler had to do just the right (wrong) thing |
| 22:14 | <rkulagow> | moegreen: unknown condition for zipcode 28371, "116" for Friday. |
| 22:27 | <rkulagow_> | "Heavy T-storms" |
| 22:47 | <-- rkulagow_ | (~rkulagow@12-207-131-29.client.attbi.com) has left #mythtv |
| 23:14 | <Ripp> | anyone know if windows media player likes the nup .avi format? (any version of wmp?) A certain non-linux user in the house wants to know :p |
| 23:14 | -!- | rkulagow [] has quit [Read error: 104 (Connection reset by peer)] |
| 23:14 | <Ripp> | mpeg4 not rtjpeg |
| 23:19 | <Ripp> | nm just searched the list archives :p |
| 23:35 | <FryGuy> | i think you need a codec for them |
| 23:35 | <FryGuy> | although you can transcode them |
| 23:35 | <FryGuy> | mythmkmovie or something |
| 23:49 | -!- | rkulagow_ [~rkulagow@12-207-131-29.client.attbi.com] has joined #mythtv |
| 23:50 | <-- rkulagow_ | (~rkulagow@12-207-131-29.client.attbi.com) has left #mythtv |
| 23:52 | -!- | rkulagow_ [~rkulagow@12-207-131-29.client.attbi.com] has joined #mythtv |
| 23:53 | -!- | rkulagow [~mythtv@12-207-131-29.client.attbi.com] has joined #mythtv |
