Beirdobot, irc.freenode.net

Sunday, June 1st, 2014, 00:59 UTC

[00:59:03]

gigem (gigem!~david@mythtv/developer/gigem) has quit (Quit: WeeChat 0.4.3)

[01:00:28]

gigem (gigem!~david@pool-71-170-165-247.dllstx.fios.verizon.net) has joined #mythtv

[01:00:30]

gigem (gigem!~david@pool-71-170-165-247.dllstx.fios.verizon.net) has quit (Changing host)

[01:00:30]

gigem (gigem!~david@mythtv/developer/gigem) has joined #mythtv

[01:12:13]

arescorpio (arescorpio!~arescorpi@105-24-245-190.fibertel.com.ar) has joined #mythtv

[01:38:47]

bill6502 (bill6502!~bill@205.178.26.43) has left #mythtv ()

[02:30:12]

peper03 (peper03!~peper03@mythtv/developer/peper03) has quit (Ping timeout: 240 seconds)

[02:33:38]

jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv

[02:37:23]

SteveGoodey (SteveGoodey!~steve@host86-157-184-176.range86-157.btcentralplus.com) has joined #mythtv

[02:56:32]

peper03 (peper03!~peper03@mythtv/developer/peper03) has joined #mythtv

[03:03:57]

SteveGoodey (SteveGoodey!~steve@host86-157-184-176.range86-157.btcentralplus.com) has quit (Quit: Konversation terminated!)

[03:04:29]

rsiebert_ (rsiebert_!~quassel@g225059238.adsl.alicedsl.de) has quit (Ping timeout: 252 seconds)

[03:41:45]

jya_:

wagnerrp: looking at why my frontend canâ€™t connect to my backend if IPv6 is set.

[03:42:01]

jya_:

in serverpool.cpp, line 103: I see bool v6IsSet = config_v6.isNull() ? true : false;

[03:42:11]

jya_:

are you sure thatâ€™s rght?

[03:42:27]

jya_:

why would ipv6 considered as set if thereâ€™s no IP address defined

[03:44:32]

fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has quit (Ping timeout: 252 seconds)

[03:45:43]

fetzerch (fetzerch!~quassel@unaffiliated/fetzerch) has joined #mythtv

[03:54:59]

wagnerrp:

jya_: the IsSet variables track whether the IP address defined in the database are configured on the system

[03:55:12]

jya_:

ok.. yeah I figured that.

[03:55:25]

wagnerrp:

if there is no address defined in the system, then there is no reason to perform the check, so it sets it to true

[03:55:38]

jya_:

donâ€™t get why it doesnâ€™t connect in my caseâ€¦ it seems to always seek the IP address from a different hostname than the name of my backend

[03:55:56]

jya_:

my backend is cinema.avenard.org, yet the frontend request to get settings from htpc.avenard.org

[03:56:07]

jya_:

which doensâ€™t exist (not for several months)

[03:56:23]

jya_:

was working yesterday, come back this morning, and nothing works anymore

[03:57:01]

jya_:

when the frontend connect to my cinema.avenard.org backend, in the log on cinema.avenard.rog it reads: No address defined for host: htpc.avenard.org

[03:57:18]

wagnerrp:

might be something here... https://code.mythtv.org/cgit/mythtv/tree/myth . . . ext.cpp#n874

[03:59:14]

wagnerrp:

get settings from the htpc.... host? or get settings from the htpc... profile?

[03:59:37]

wagnerrp:

i.e. is it using htpc.avenard.org as the index to look up the setting in the database?

[04:00:08]

wagnerrp:

is it possible you have some old recordings mapped to that machine name, so it is trying to figure out how to connect to it to stream from it?

[04:00:30]

jya_:

no idea

[04:00:44]

jya_:

thatâ€™s just starting the frontend

[04:00:54]

jya_:

i get stuck on the first screen asking for sql settings

[04:00:59]

jya_:

and it loops whenever i validate

[04:01:22]

arescorpio (arescorpio!~arescorpi@105-24-245-190.fibertel.com.ar) has quit (Excess Flood)

[04:02:54]

jya_:

sigh

[04:03:14]

jya_:

so annoying this network configuration businessâ€¦ when it has decided to not work, it really doesnâ€™t work

[04:03:22]

wagnerrp:

if it's asking for sql settings, that code hasnt even engaged yet

[04:03:37]

wagnerrp:

that's all internal to the mysql library

[04:04:26]

jya_:

i can see a connection being made to the backend

[04:04:40]

jya_:

the backend shows a new connection being made

[04:10:23]

jya:

dblain: do you manually compile the ffmpeg libs in your buildbot?

[04:14:41]

jya:

wagnerrp: the reason I started looking is because when the backend is configured to listen on a link-local IPV6 address, the frontend will fail to connect.

[04:14:48]

jya:

I remember seeing a bug on this matter.

[04:15:14]

jya:

as it is , it will only ever work if the backend uses the same ethernet card name as the frontend

[04:15:50]

jya:

as the scopeID is saved with the address. the frontend attempts then to connect using the IPv6 address including the scope.

[04:16:06]

jya:

thatâ€™s just wrong.. scope is useful to listen on an address, not to connect to

[04:16:46]

wagnerrp:

it won't work if you just specify the IP address sans scope in the database?

[04:17:20]

jya:

well, you can listen to an link-local IPv6 address without the scope

[04:17:35]

jya:

as if you have multiple network card, you have multiple link-local address, and all in the same subnet

[04:17:54]

jya:

so it needs to the scope to know which card to use

[04:18:11]

jya:

so if the backend is set to listen IPv6 on link-local

[04:18:28]

jya:

BackendServerIP6 will contain FE80:xxxxx:%eth0 (typically)

[04:18:31]

jya:

brb

[04:20:28]

jya:

so when the frontend attempts to connect to the backend, it retrieves the BackendServerIP6 setting.

[04:21:25]

jya:

the scope must be removed when it attempts to use it, otherwise it will only work with a machine that has the same ethernet card adapter (e.g. will work with your typical all linux bo

[04:21:35]

jya:

but on mac, freebsd, windows, that won't

[04:21:42]

jya:

eth0 not being a valid adapter

[04:22:17]

jya:

to fix it we can either look at all code calling GetBackendServerIP() and remove the scope when required

[04:22:28]

jya:

or have by default GetBackendServerIP() be de-scoped

[04:22:46]

jya:

and have the backend itself use a different function call to get its own listening address

[04:24:09]

jya:

ah that #10921

[04:24:09]

** MythLogBot http://code.mythtv.org/trac/ticket/10921 **

[04:25:01]

jya:

the logs of that ticket are obsolete, as the original bug report didnâ€™t set the scope id at all

[04:25:07]

jya:

so listening would have failed

[04:29:42]

jya:

but to fix that, I first need my frontend to at least connect

[04:31:02]

Warped (Warped!~Warped@108-85-161-113.lightspeed.cicril.sbcglobal.net) has quit (Quit: ChatZilla 0.9.90.1 [Firefox 29.0.1/20140506152807])

[04:36:38]

knightr_ (knightr_!~Nicolas@69-165-170-178.dsl.teksavvy.com) has quit (Excess Flood)

[04:36:49]

J-e-f-f-A (J-e-f-f-A!~J-e-f-f-A@unaffiliated/j-e-f-f-a) has quit (Excess Flood)

[04:39:28]

J-e-f-f-A_ (J-e-f-f-A_!~J-e-f-f-A@24-177-48-184.dhcp.oxfr.ma.charter.com) has joined #mythtv

[04:39:30]

J-e-f-f-A_ is now known as J-e-f-f-A

[04:40:16]

64MAAWR3F (64MAAWR3F!~Warped@108-85-161-113.lightspeed.cicril.sbcglobal.net) has joined #mythtv

[04:41:28]

J-e-f-f-A (J-e-f-f-A!~J-e-f-f-A@24-177-48-184.dhcp.oxfr.ma.charter.com) has quit (Changing host)

[04:41:28]

J-e-f-f-A (J-e-f-f-A!~J-e-f-f-A@unaffiliated/j-e-f-f-a) has joined #mythtv

[04:43:29]

64MAAWR3F is now known as Warped

[04:45:40]

knightr (knightr!~Nicolas@69-165-170-178.dsl.teksavvy.com) has joined #mythtv

[04:45:40]

knightr (knightr!~Nicolas@69-165-170-178.dsl.teksavvy.com) has quit (Changing host)

[04:45:40]

knightr (knightr!~Nicolas@mythtv/developer/knightr) has joined #mythtv

[04:53:28]

dblain:

jya: in order to speed up builds, the powershell script just checks to see if the ffmpeg dll/libs are present, if so, it skips calling into minwg. This hasn't been a problem up till now, since you are actively resyncing. A simple change to the script will allow ffmpeg to decide for itself what needs to be compiled.

[04:53:42]

dblain:

Let me know if you need me to make that change.

[04:54:18]

dblain:

(I fixed the last build by simplly doing a clean)

[05:00:19]

jya_:

dblain: nah thatâ€™s fineâ€¦ glad it compiles okay on windows

[05:10:42]

jya_ (jya_!~jyavenard@mythtv/developer/jya) has quit (Quit: jya_)

[05:11:22]

jya_ (jya_!~jyavenard@mythtv/developer/jya) has joined #mythtv

[05:14:46]

jya_:

wagnerrp: for some reason MasterServerIP is read as null on that particular frontend.

[05:14:51]

jya_:

another frontend can connect just fine

[05:14:56]

jya_:

how bizarre....

[05:16:50]

jya_:

must be something to do with my mysql pluginâ€¦

[05:16:51]

sl1ce (sl1ce!~johnathan@pool-100-0-124-62.bstnma.fios.verizon.net) has quit (Remote host closed the connection)

[05:18:06]

sl1ce (sl1ce!~johnathan@pool-100-0-124-62.bstnma.fios.verizon.net) has joined #mythtv

[05:55:29]

jya_:

itâ€™s the same issue as yesterday.. itâ€™s myth_system that somehow fails to run

[05:57:17]

jya_:

the ping is seen as failing, and then it never attempts to connect

[06:21:39]

jya_:

oh well, restored from a backup, and itâ€™s all working againâ€¦

[07:11:47]

jheizer (jheizer!~jheizer@73.51.93.177) has quit (Ping timeout: 260 seconds)

[07:26:30]

jheizer (jheizer!~jheizer@73.51.93.177) has joined #mythtv

[07:33:10]

jpharvey__ (jpharvey__!~jpharvey@host109-148-113-119.range109-148.btcentralplus.com) has joined #mythtv

[07:36:33]

jpharvey_ (jpharvey_!~jpharvey@host109-148-113-119.range109-148.btcentralplus.com) has quit (Ping timeout: 252 seconds)

[08:25:39]

Weaselweb (Weaselweb!~quassel@77-64-221-251.dynamic.primacom.net) has joined #mythtv

[08:30:51]

SteveGoodey (SteveGoodey!~steve@host86-157-184-176.range86-157.btcentralplus.com) has joined #mythtv

[08:40:04]

rsiebert (rsiebert!~quassel@g225156112.adsl.alicedsl.de) has joined #mythtv

[08:46:08]

SteveGoodey (SteveGoodey!~steve@host86-157-184-176.range86-157.btcentralplus.com) has quit (Quit: Konversation terminated!)

[09:10:47]

wagnerrp (wagnerrp!~wagnerrp_@mythtv/developer/wagnerrp) has quit (Ping timeout: 252 seconds)

[09:31:35]

natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has joined #mythtv

[09:36:05]

Seeker (Seeker!~cjo20@unaffiliated/seeker) has quit (Ping timeout: 255 seconds)

[09:43:24]

Seeker (Seeker!~cjo20@unaffiliated/seeker) has joined #mythtv

[10:19:54]

wagnerrp (wagnerrp!~wagnerrp_@mythtv/developer/wagnerrp) has joined #mythtv

[10:52:56]

jpharvey__ (jpharvey__!~jpharvey@host109-148-113-119.range109-148.btcentralplus.com) has quit (Ping timeout: 276 seconds)

[11:04:33]

jpharvey__ (jpharvey__!~jpharvey@host109-148-113-67.range109-148.btcentralplus.com) has joined #mythtv

[11:19:12]

SteveGoodey (SteveGoodey!~steve@host86-157-184-176.range86-157.btcentralplus.com) has joined #mythtv

[11:55:03]

natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has quit (Ping timeout: 265 seconds)

[12:05:19]

Seeker (Seeker!~cjo20@unaffiliated/seeker) has quit (Ping timeout: 252 seconds)

[12:12:56]

Seeker (Seeker!~cjo20@unaffiliated/seeker) has joined #mythtv

[12:59:43]

jya_:

IPv6 config was so broken reallyâ€¦

[13:08:01]

Warped (Warped!~Warped@108-85-161-113.lightspeed.cicril.sbcglobal.net) has quit (Quit: ChatZilla 0.9.90.1 [Firefox 29.0.1/20140506152807])

[13:22:27]

stuartm:

Spot the problem here: http://mythtv.co.uk/imagebin/spot%20the%20problem.png

[13:27:23]

jpharvey__ (jpharvey__!~jpharvey@host109-148-113-67.range109-148.btcentralplus.com) has quit (Ping timeout: 245 seconds)

[13:33:23]

jya_:

next 5 showings only have one ? :) font is weird, letters arenâ€™t of the same height. It says starring Billy Campbell, yet itâ€™s not in the list of actors ..

[13:33:55]

jya_:

duration 2h et itâ€™s from 9 to 10

[13:34:00]

jya_:

yet

[13:34:25]

jya_:

and the next showing as the same time, on the same channel, but a different title

[13:34:48]

jya_:

1 more problem to make it 7 :)

[13:35:46]

jya_:

why was the master server port 6543, yet status is 6544â€¦ why wasnâ€™t that put on the same port?

[13:35:52]

jya_:

just wondering

[13:36:23]

joki (joki!~joki@p54860CE9.dip0.t-ipconnect.de) has quit (Ping timeout: 255 seconds)

[13:40:11]

jpharvey__ (jpharvey__!~jpharvey@host109-148-113-67.range109-148.btcentralplus.com) has joined #mythtv

[13:41:35]

joki (joki!~joki@p54863F37.dip0.t-ipconnect.de) has joined #mythtv

[13:49:59]

stuartm:

yeah, duration is wrong, cast list is wrong, category is wrong – in short it's showing the show details for that programme :/

[13:50:12]

stuartm:

jya_: no idea, long before my time

[13:50:29]

stuartm:

s/show/wrong/

[13:50:53]

jya_:

also, whatâ€™s the difference between GetSetting() and GetSettingOnHost(), especially to retrieve BackendStatusPort

[13:51:44]

jya_:

like in mythbackend::httpstatus(), it uses GetSetting("MasterServerIPâ€) for retrieving the IP address, but GetSettingOnHost("BackendStatusPort" to retrieve the port number

[13:51:47]

jya_:

doesnâ€™t make sense to me

[13:53:09]

jya_:

and sometimes itâ€™s the opposite

[13:53:35]

jya_:

in all cases, the aim being to query the master backend (judging by the name of the function, and the comments (if any))

[13:53:55]

stuartm:

jya_: one is for retrieving a global setting, the other a per-host setting, but it's an archaic distinction and shouldn't apply to those settings

[13:54:32]

jya_:

so itâ€™s fine to just use GetSetting() to retrieve the backend status port?

[13:56:05]

jya_:

Iâ€™m replacing the occurrences with helpers from gCoreContext

[13:56:15]

jya_:

(cleaner IMHO)

[13:59:03]

stuartm:

commit emails working for everyone?

[13:59:14]

stuartm:

just noticed the last one I received was last Thursday

[14:00:43]

jya_:

never subscribed to thoseâ€¦ so canâ€™t say

[14:02:20]

stuartm:

jya_: not sure, we may allow the port to be changed for slaves so GetSetting() may end up returning the wrong value for the master on slaves

[14:02:40]

jya_:

ok.. wonâ€™t touch those then. just in case

[14:02:42]

stuartm:

IMHO it would be cleaner to have a MasterStatusPort

[14:02:52]

stuartm:

no confusion then

[14:03:09]

jya_:

what Iâ€™m doing I believe needs to be backported to fixes/0.27, as it fixes IPv6 setup

[14:04:04]

stuartm:

last email from -dev was 20th May

[14:05:19]

jya_:

somehow that reminds me that we still donâ€™t have an official announcement of 0.27.1 on our web site

[14:06:53]

stuartm:

bah, how has it taken me three days to notice that I've not received _any_ email? Fetchmail refusing to connect to my server ...

[14:07:58]

jya_:

itâ€™s the weekend :)

[14:08:05]

jya_:

stuartm: look at this one for example: https://github.com/MythTV/mythtv/blob/master/ . . . erer.cpp#L63

[14:08:33]

jya_:

how can it make any sense to attempt to connect to the global master IP address, but at the hostâ€™s port number

[14:08:52]

jya_:

itâ€™s entirely possible, on the frontend that mythtv-setup was run and a different port was set

[14:09:05]

jya_:

so it uses a global IP, and a local-config for the port

[14:11:02]

stuartm:

it's specifying the port of the master host

[14:11:41]

stuartm:

MasterServerIP is a global setting, but then it's looking up the port for the server associated with the master hostname

[14:12:16]

stuartm:

it's a kludge, as I noted before it should we should just use MasterPort or similar

[14:12:33]

jya_:

if itâ€™s the port of the master host, then it should be GetNumSetting("BackendStatusPortâ€, 6544)

[14:13:29]

jya_:

the global setting

[14:14:23]

jya_:

you can tell wherever you have the same issue, itâ€™s because someone just copy/pasted the code from an original place

[14:14:31]

jya_:

itâ€™s all identical code everywhere

[14:14:42]

jya_:

too much of a coincidence

[14:14:43]

stuartm:

GetSetting() will check the current host first

[14:15:11]

stuartm:

think I explained it poorly before

[14:16:09]

stuartm:

GetSetting() checks the local host, then the global value (hostname of NULL), GetSettingOnHost() checks the specified host only

[14:16:49]

stuartm:

jya_: did you see my comment about mythmusic segfaulting following the ffmpeg resync?

[14:16:57]

jya_:

stuartm: no I haven't

[14:17:07]

** jya_ oops, forgot to check the plugins **

[14:17:09]

stuartm:

want a backtrace?

[14:17:14]

jya_:

sure

[14:17:28]

stuartm:

happens instantly on playback start

[14:17:39]

jya_:

wherever Iâ€™ve had a crash following the resync, was because we didnâ€™t call the FFmpeg API properly

[14:18:07]

jya_:

in particular, in ffmpeg 1.2 if you called it with invalid argument, it would initialise the structure.

[14:18:19]

jya_:

now it doesnâ€™tâ€¦ must initialise it first

[14:19:10]

jya_:

can the frontend use and connect to UPnP server?

[14:19:15]

stuartm:

hmm, doesn't want to run under gdb, causes a be/fe connection timeout

[14:19:22]

jya_:

looking at programs/mythfrontend/upnpscanner.cpp

[14:19:40]

stuartm:

jya_: yes, in a primitive sort of way, and it deliberately ignores the internal server

[14:19:41]

jya_:

stuartm: Iâ€™ll set it mythmusic and try it

[14:20:02]

jya_:

what does it do with the upnp server?

[14:20:11]

jya_:

ir is that just to find the master backend ?

[14:21:01]

stuartm:

in mythvideo, when you 'Scan' if it finds a uPnP server with videos they will be displayed in the video library – but I only learnt about this a couple of days ago myself, never tried it

[14:21:16]

jya_:

oh whaoâ€¦ didnâ€™t know that

[14:21:21]

stuartm:

I've never had a upnp server running at the time I scanned :)

[14:21:34]

stuartm:

overlooked the original commit by Mark

[14:21:52]

jya_:

thatâ€™s pretty cool

[14:21:57]

jya_:

if that works that is :)

[14:22:09]

jya_:

if we donâ€™t know.. I wonder who does.

[14:22:14]

stuartm:

from Mark's commit message it was an initial unpolished implementation

[14:22:36]

stuartm:

don't know if he got it into a finished state

[14:24:02]

jya_:

i did read something about having a UPnP client, but that it was disabled for the time being

[14:24:41]

jya_:

stuartm: something to do in the server javascript code (unless we change how the backend IP address is stored)

[14:25:28]

jya_:

the current setting, store in MasterServerIP the scope Id as itâ€™s a list of addresses retrieved by Qt QNetworkInterface::allAddresses()

[14:25:51]

jya_:

which adds the scope Id to all IPv6 address

[14:26:00]

jya_:

e.g. 2001::%eth0

[14:26:47]

jya_:

a client retrieves that MasterServerIP, and attempts to connect to it. that will be fine if the client is running a computer with exactly the same network name and with that card connected

[14:26:53]

jya_:

otherwise it will fail

[14:27:10]

jya_:

so when we retrieve MasterServerIP, the scopeId must be reset and removed.

[14:27:40]

jya_:

so whenever your javascript reads MasterServerIP, it must make sure the scopeId is cleared

[14:28:19]

jya_:

obviously, if the MasterServerIP setting didnâ€™t include the scope Id there would be no issue, but it does; and if we want to be backward compatible without a schema change that would remove the scope id from the stored setting

[14:28:30]

jya_:

same deal with the php , python and mysql bindings

[14:28:35]

jya_:

thatâ€™s for wagnerrp

[14:29:13]

jya_:

Iâ€™ve updated the C code , so whenever that setting is read, it is cleaned up (except where the scope Id is actually required: like the address where to listen to)

[14:52:26]

jya_:

stuartm: see here: https://github.com/MythTV/mythtv/blob/master/ . . . o.cpp/#L2513

[14:52:39]

jya_:

now that makes sense, you retrieve the backendIP and port for the host.

[14:53:02]

jya_:

not the global hostname, with the local port

[15:42:53]

lomion0815 (lomion0815!~markus@212-88-2-177.adsl.highway.telekom.at) has joined #mythtv

[15:45:40]

lomion0815 (lomion0815!~markus@212-88-2-177.adsl.highway.telekom.at) has quit (Remote host closed the connection)

[16:01:54]

jya_:

stuartm: is it still possible to set mythmusic without storage group?

[16:05:14]

jya_:

I see that you can set Directory to hold music, but that value doesnâ€™t appear to be saved

[16:15:20]

paul-h (paul-h!~Paul@2.121.114.15) has joined #mythtv

[16:15:38]

superm1 (superm1!uid4318@ubuntu/member/superm1) has joined #mythtv

[16:15:50]

stuartm:

jya_: I really don't know, paul-h would though

[16:16:14]

paul-h:

jya_: what theme is that? That setting should have been removed by the theme

[16:16:22]

jya_:

Steppes

[16:16:34]

paul-h:

You need to use a storage group for music

[16:16:41]

jya_:

itâ€™s weird, I definitely set a storage for music

[16:16:52]

jya_:

select * from storagegroup WHERE groupname='Music';

[16:16:57]

jya_:

[16:17:17]

jya_:

but whenever I try to do anything, it tells me I need to add a Music storage group

[16:17:56]

jya_:

ah.. I must have a MusicArt storage group

[16:18:18]

jya_:

why canâ€™t it default to the default SG like all other SG ? a bit inconsistent

[16:19:42]

paul-h:

Most people already have separate directories for there music and extracted album art

[16:19:46]

jya_:

paul-h: do you have to access directly the DB to check if a storage group is present or not ?

[16:20:11]

wagnerrp:

the video library only falls back to the base video library SG, not all the way to 'Default'

[16:20:29]

wagnerrp:

'Default' is only the fallback for recordings

[16:20:37]

jya_:

paul-h: having the ability to separate the directory is great.. doesnâ€™t mean you should have toâ€¦ every other SG default to the â€œdefaultâ€ SG.. I think it would be better to keep that behaviour and be consistent

[16:21:09]

jya_:

wagnerrp: coverart are definitely defaulting to default

[16:21:24]

jya_:

thatâ€™s all I have set on my dev system

[16:21:40]

paul-h:

Wouldn't that cause all sorts of problems with available space etc

[16:22:02]

wagnerrp:

coverart for recordings?

[16:22:03]

jya_:

well, thatâ€™s the user problem isnâ€™t it ? :)

[16:22:13]

jya_:

Iâ€™d be surprised if people used a different disk anyway

[16:23:02]

wagnerrp:

yeah. artwork, and even music, should never be a concern for space on systems where users store broadcast video

[16:23:14]

paul-h:

jya_: it does query the DB to check the music storage groups

[16:23:41]

jya_:

paul-h: yes, I see thatâ€¦ I think thatâ€™s a bit dirty

[16:24:29]

jya_:

scan music was incredibly fast.. I doubt that it actually worked :( (I pointed to a 40GB music directory)

[16:25:06]

jya_:

do you need to install a plugin on the backend too?

[16:25:50]

paul-h:

it runs in the background on the BE now

[16:26:20]

jya_:

donâ€™t see anything in the log of the backend, suggesting it did a scan

[16:26:42]

jya_:

ah, it just appeared: MainServer: HandleScanMusic: running filescanner on master BE 'cinema.avenard.org

[16:26:53]

jya_:

that took a while, over 3 minutes

[16:27:41]

stuartm:

jya_: fwiw, I think the only place the server side script explicitly uses the master IP is for the link back to the master webserver from the frontend webserver

[16:28:22]

stuartm:

we should strip the scope id in the services API (C++) before it's ever seen by the javascript, if not earlier, e.g. in GetMasterServerIP

[16:28:41]

jya_:

stuartm: ultimately, having the backend IP setting only containing a clean IP (with no scope ID, proper case etc) would be the best, and least messy

[16:28:58]

jya_:

GetMasterServerIP() does so now

[16:29:07]

jya_:

(strip the scope)

[16:29:31]

stuartm:

ok, then once the services API is updated to use that instead everything should be fine

[16:30:30]

stuartm:

in fact, there's no GetMasterIP in the Frontend services, so we're not using it anywhere

[16:30:46]

jya_:

i think we should store the address without the scopeId, and when we attempt to use the IP, only then do we attempt to find the scope. If the IP is local, itâ€™s trivial and extremely fast

[16:31:13]

jya_:

stuartm: the javascript queries directlry MasterServerIP setting

[16:32:04]

stuartm:

oh really? that's nasty

[16:32:27]

jya_:

or the qsp

[16:32:45]

jya_:

hmmm.. I donâ€™t get it, Iâ€™ve missed plenty of â€œMasterServerIPâ€

[16:33:05]

stuartm:

ah, in the networking setup page ... hmm, it actually makes sense there

[16:33:55]

stuartm:

QSP is javascript + markup, only interpreted by the server, rather than the client

[16:35:15]

jya_:

actually, itâ€™s in mythnetvision that MasterServerIP that Iâ€™ve missed it.. gosh that code is ugly..

[16:35:19]

stuartm:

oh jeez, no, that is ugly – someone has missed the point of qsp and is querying the services API via client side javascript

[16:35:20]

jya_:

does that even work?

[16:36:48]

stuartm:

not well

[16:37:50]

jya_:

IMHO, thereâ€™s the same bug in regards to global MasterServerIP vs host BackendStatusPort, except itâ€™s one via MySQL queries, and that thing is certainly not going to work

[16:39:55]

stuartm:

code reading settings directly from the database doesn't inspire a lot of confidence

[16:40:00]

jya_:

paul-h: what verbosity should I add so I can see if itâ€™s scanning for music properly?

[16:45:09]

jya_:

ahh.. I think I know why

[16:45:16]

jya_:

it runs mythutil --scanmusic

[16:45:46]

jya_:

but on my system mythutil --scanmusic is fixes/0.27, the path to mythutil master isnâ€™t there.. Prefix should be added to it

[16:46:47]

jya_:

paul-h: can I fix this or I let ou do (and I go to bed while at it)

[16:47:55]

jya_:

paul-h: wherever mythutil is called, GetAppBinDir() + should be added before.

[16:48:25]

neufeld (neufeld!~user@69-165-173-139.dsl.teksavvy.com) has quit (Quit: ERC Version 5.3 (IRC client for Emacs))

[16:48:40]

jya_:

e.g. MythSystem::Create(GetAppBinDir() + â€œmythutilâ€, ...

[16:52:13]

jya_:

thatâ€™s betterâ€¦

[16:53:25]

jya_:

stuartm: no crash playing music in mythmusic here

[16:54:07]

jya_:

paul-h: I think it would be nicer if the full path wasnâ€™t displayed when showing the songs.

[16:54:46]

jya_:

ah, after a 2nd scan it doesnâ€™t, only the song name

[16:56:08]

jya_:

stuartm: youâ€™ll have to give me that backtraceâ€¦

[16:56:15]

** jya_ off to bed **

[17:02:41]

stuartm:

jya_, paul-h: http://pastebin.com/bpQ7C1V7

[17:04:54]

paul-h:

jya_: sorry been away grabbing a bit to eat – I can add the path

[17:05:30]

paul-h:

stuartm: I've been playing music all day without any problems

[17:06:16]

stuartm:

paul-h: huh, maybe I'll try a another distclean/rebuild of the plugins

[17:09:10]

stuartm:

since it appears there was no binary version bump it's entirely possible that I built but forgot to install the plugins, I've done that before ...

[17:10:04]

paul-h:

yeah I think we've all done that at some time :)

[17:13:44]

wagnerrp:

alcor's getting hammered by something

[17:15:41]

paul-h:

Yeah I noticed it was taking a long time to respond

[17:16:15]

paul-h:

I think it may have been like that for an hour or two

[17:17:44]

stuartm:

I can never remember where the access logs for alcor are hidden

[17:18:05]

wagnerrp:

/opt/www/[site]/logs

[17:18:43]

wagnerrp:

lot of hits from 'git'

[17:20:11]

Steve-Goodey (Steve-Goodey!~steve@host86-157-184-176.range86-157.btcentralplus.com) has joined #mythtv

[17:21:49]

stuartm:

yeah found them, lot of traffic but not much of note for www

[17:22:41]

wagnerrp:

we shouldn't be running all those instances of git should we?

[17:23:01]

stuartm:

dunno what's normal

[17:24:19]

stuartm:

heavy git traffic would be unusual though since we're not hosting any public repos

[17:24:41]

wagnerrp:

i was thinking it might have to do with cgit, but there's no traffic on code.mythtv.org

[17:25:20]

stuartm:

not seeing any active connections to git directly

[17:26:25]

natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has joined #mythtv

[17:26:45]

wagnerrp:

whatever they're doing, they're long running and have been there since boot

[17:28:18]

wagnerrp:

oh, here we go. check out ssl_access on code.mythtv.org

[17:29:00]

wagnerrp:

someone is really interested in ticket 9223

[17:29:20]

gigem:

stuartm: Where do you think recordedid should be returned in the services API? I was expecting to find a monolithic Recording type, but that doesn't appear to be the case. I have a change ready that, except for the services part, populates recordedid and makes it available in ProgramInfo. Future changes will transition internal interfaces to use recordedid instead of chanid/recstartts. I expect we'll need

[17:29:22]

gigem:

to keep public chanid/recstartts interfaces around for another release or so.

[17:32:59]

stuartm:

gigem: datacontracts/recording.h which is populated in programs/mythbackend/services/dvr.cpp

[17:33:46]

shattingduck (shattingduck!~quassel@cable-94-139-22-2.cust.telecolumbus.net) has quit (Remote host closed the connection)

[17:33:52]

stuartm:

gigem: will there be a similar programid?

[17:34:38]

stuartm:

that's libs/libmythservicecontracts/datacontracts/recording.h

[17:39:54]

gigem:

stuartm: I looked in recording.h, but for some reason I thought is was a lot more sparse than it should have been and looked elsewhere. I'll look again.

[17:42:01]

gigem:

stuartm: Regarding programid, do you mean splitting the current program table so we don't duplicate a lot of data? That should be done, but it's not something I intend to take on right now. Or do you mean something else?

[17:42:26]

gregL (gregL!~greg@cpe-74-76-121-109.nycap.res.rr.com) has quit (Ping timeout: 276 seconds)

[17:43:02]

stuartm:

gigem: I just meant referencing each program by an ID instead of chanid/starttime

[17:45:18]

wagnerrp:

can we easily do UA filtering?

[17:45:19]

stuartm:

gigem: recording.h only contains the information specific to a recording, the rest is in programandchannel.h, the program class, the services API strives not to duplicate data objects, so when loading information for a recording for example we create a Program object which references a RecordingInfo object

[17:45:29]

gigem:

Do we have (or create) programids for all sources now? I know manual programs don't, but that could be fixed easily. If so, yes, programid would probably be appropriate in some cases.

[17:46:01]

gigem:

That makes sense.

[17:46:55]

stuartm:

wagnerrp: yes we can filter by UA easily enough, though it's not currently in the form of pre-existing list we can just append to

[17:47:44]

stuartm:

wagnerrp: sorry, did you mean alcor or for the backend webserver?

[17:47:50]

wagnerrp:

should probably drop "FlipboardProxy/1.1" for now

[17:47:51]

wagnerrp:

alcor

[17:48:10]

wagnerrp:

they just keep hammering /trac/ticket/9223

[17:48:21]

stuartm:

dropping IPs is a lot easier

[17:48:32]

wagnerrp:

they're running on an EC2 cluster

[17:48:39]

wagnerrp:

dropping IPs would be painful

[17:48:44]

stuartm:

heh yeah

[17:49:54]

stuartm:

honestly I don't know how to filter by useragent with apache – stuarta?

[17:51:18]

wagnerrp:

i sent an email to their listed support address, but i don't expect a response any time soon

[17:51:40]

wagnerrp:

i'm about to head out of town for a few days, so I cc'd you on it

[17:53:34]

stuartm:

yeah, IP bans are getting me nowhere

[17:53:44]

stuartm:

I'm off to eat

[17:58:33]

paul-h:

stuartm: just noticed a problem with the webfrontend if you open the setup wizard then go to the Program Guide the wizard is still present under the guide

[18:02:18]

wagnerrp:

on a side note, i'll be in Salem in a couple weeks, about half an hour north of OSUOSL, if anyone wants something done to alcor

[18:02:31]

wagnerrp:

that is, assuming they actually allow non-employees into the DC

[18:14:26]

stuartm:

paul-h: thanks, I'll take a look

[18:14:56]

moparisthebest (moparisthebest!~quassel@cpe-74-140-228-202.swo.res.rr.com) has quit (Ping timeout: 276 seconds)

[18:15:07]

MythBuild:

build #1703 of master-ubuntu-current-64bit is complete: Failure [4failed unit test core] Build details are at http://code.mythtv.org/buildbot/builders/mast . . . /builds/1703 blamelist: Paul Harrison <pharrison@mythtv.org >, Stuart Morgan <smorgan@mythtv.org >

[18:21:16]

gregL (gregL!~greg@cpe-74-76-121-109.nycap.res.rr.com) has joined #mythtv

[18:27:59]

Warped (Warped!~Warped@99-116-29-44.lightspeed.cicril.sbcglobal.net) has joined #mythtv

[18:34:13]

andreaz (andreaz!~andre_000@p5DCA3B0D.dip0.t-ipconnect.de) has joined #mythtv

[18:44:02]

stuartm:

wagnerrp: trying some htaccess rules to block by useragent

[19:00:18]

superm1 (superm1!uid4318@ubuntu/member/superm1) has quit (Quit: Connection closed for inactivity)

[19:06:25]

stuartm:

I think, once I'd moved it to the right directory, that actually worked

[19:09:29]

moparisthebest (moparisthebest!~quassel@cpe-74-140-228-202.swo.res.rr.com) has joined #mythtv

[19:22:31]

stuartm:

the bot traffic in general is pretty heavy, if it didn't mean us dropping from search engines I'd ban the lot of them, it's diminishing the QoS for the human beings

[19:26:02]

stuartm:

#10631

[19:26:02]

** MythLogBot http://code.mythtv.org/trac/ticket/10631 **

[19:45:13]

gigem:

stuartm: I added recordedid to Recording and it gets returned when I call GetRecordedList. However, in the web interface, if I choose Recordings, Upcoming Recordings or Recording Rules, the content comes up but the 'Loading ...' never goes away. That doesn't happen when I choose Program Guide. Is that something I could have caused or is it a new, known issue. I've never seen it before.

[19:46:15]

stuartm:

anything in the backend log?

[19:46:57]

stuartm:

gigem: actually, I don't think it's anything you've done, I've noticed it myself a couple of times today (but not before), possibly something I've done

[19:47:52]

gigem:

Nope, nothing in the backend log. Okay, I didn't think my trivial little change could have done that, but wanted to be sure.

[19:52:56]

stuartm:

paul-h: backend crashed while I was playing music in the frontend (unrelated) but then the frontend crashed as well

[19:54:00]

stuartm:

from the bt it crashed in ffmpeg though ... so not our bug

[19:55:40]

stuartm:

jya_: btw, that crash I mentioned, user error ...

[19:56:01]

stuartm:

the first one anyway, the second crash is a possibly new bug in ffmpeg

[19:56:53]

gigem:

Is anyone around who can easily test mythweb with master? I'd like to make sure my recordedid change to it works.

[20:13:57]

MythBuild:

build #1704 of master-ubuntu-current-64bit is complete: Success [3build successful] Build details are at http://code.mythtv.org/buildbot/builders/mast . . . /builds/1704

[20:44:27]

arescorpio (arescorpio!~arescorpi@105-24-245-190.fibertel.com.ar) has joined #mythtv

[21:05:07]

Weaselweb (Weaselweb!~quassel@77-64-221-251.dynamic.primacom.net) has quit (Remote host closed the connection)

[21:17:59]

Steve-Goodey (Steve-Goodey!~steve@host86-157-184-176.range86-157.btcentralplus.com) has quit (Quit: Konversation terminated!)

[21:20:03]

SteveGoodey (SteveGoodey!~steve@host86-157-184-176.range86-157.btcentralplus.com) has quit (Quit: Konversation terminated!)

[21:46:56]

andreaz (andreaz!~andre_000@p5DCA3B0D.dip0.t-ipconnect.de) has quit (Read error: Connection reset by peer)

[21:54:57]

stuartm:

gigem: I'm only mentioning this because you might have an idea why it happened and not because I expect you to try and fix; I just noticed that there was a film on, it started almost an hour ago, I selected it and chose "Record one showing" (Find one)

[21:56:03]

stuartm:

to my surprise it started recording, suprise because in 10 minutes it starts on a +1 channel – i.e. the scheduler chose to start recording the one which is already halfway through instead of the one that hasn't started yet

[21:57:01]

stuartm:

the channel it's recording on does have higher priority (HD) but shouldn't the fact that it's going to be missing an hour weigh more heavily? There are no other recordings scheduled for tonight

[22:03:06]

natanojl (natanojl!~jonatan@mythtv/developer/natanojl) has quit (Ping timeout: 252 seconds)

[22:04:58]

Tobbe5178 (Tobbe5178!~asdf@h104n2-sv-a13.ias.bredband.telia.com) has quit (Ping timeout: 240 seconds)

[22:11:51]

paul-h (paul-h!~Paul@2.121.114.15) has quit (Quit: Konversation terminated!)

[22:16:00]

stuartm:

heh, this 'Play on frontend' stuff is cool, will interrupt whatever you're currently watching

[22:16:10]

stuartm:

wasn't sure how that would work

[23:54:35]

jya_:

stuartm: what new crash? do you have a bt ? what were you doing on the frontend at the time ?

[23:54:50]

jya_:

wonder if itâ€™s due to starving of data as the backend crashedâ€¦

MythLogBot@irc.freenode.net :: #mythtv

Current users (70):