Beirdobot, irc.libera.chat

Wednesday, January 26th, 2022, 00:12 UTC

[00:12:09]

paul-h (paul-h!~paul@mythtv/developer/paul-h) has quit (Quit: Konversation terminated!)

[00:16:52]

ulmus-scott (ulmus-scott!~ulmus-sco@user/ulmus-scott) has joined #mythtv

[00:24:49]

peterbennett (peterbennett!~peter@mythtv/developer/peterbennett) has joined #mythtv

[00:24:49]

Mode for #mythtv by ChanServ!ChanServ@services.libera.chat : +v peterbennett

[00:43:28]

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

[00:45:21]

peper03_ (peper03_!~peper03@mythtv/developer/peper03) has quit (Ping timeout: 268 seconds)

[02:04:28]

peterbennett (peterbennett!~peter@mythtv/developer/peterbennett) has quit (Quit: Leaving.)

[02:25:21]

ulmus-scott (ulmus-scott!~ulmus-sco@user/ulmus-scott) has quit (Ping timeout: 256 seconds)

[07:05:58]

Steve-Goodey (Steve-Goodey!~steve@2a00:23c5:7d83:6501:6ea1:20d:5872:d872) has joined #mythtv

[07:30:05]

gigem (gigem!~david@47.189.229.40) has quit (Quit: WeeChat 3.4)

[07:31:13]

gigem (gigem!~david@47.189.229.40) has joined #mythtv

[08:48:51]

stuarta:

morning all

[09:57:56]

Steve-Goodey (Steve-Goodey!~steve@2a00:23c5:7d83:6501:6ea1:20d:5872:d872) has quit (Quit: Konversation terminated!)

[11:18:05]

amessina (amessina!~amessina@user/amessina) has quit (Remote host closed the connection)

[11:58:54]

Steve-Goodey (Steve-Goodey!~steve@2a00:23c5:7d83:6501:6ea1:20d:5872:d872) has joined #mythtv

[13:18:37]

SteveGoodey (SteveGoodey!~steve@2a00:23c5:7d83:6501:fa68:6631:5ed6:d871) has joined #mythtv

[13:18:52]

hampton:

Morning

[13:37:39]

Steve-Goodey:

Afternoon.

[13:53:57]

peterbennett (peterbennett!~peter@mythtv/developer/peterbennett) has joined #mythtv

[13:53:57]

Mode for #mythtv by ChanServ!ChanServ@services.libera.chat : +v peterbennett

[15:05:04]

peterbennett (peterbennett!~peter@mythtv/developer/peterbennett) has quit (Quit: Leaving.)

[15:23:08]

peterbennett (peterbennett!~peter@mythtv/developer/peterbennett) has joined #mythtv

[15:23:08]

Mode for #mythtv by ChanServ!ChanServ@services.libera.chat : +v peterbennett

[16:44:40]

Steve-Goodey (Steve-Goodey!~steve@2a00:23c5:7d83:6501:6ea1:20d:5872:d872) has quit (Quit: Konversation terminated!)

[17:01:17]

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

[17:15:16]

peterbennett (peterbennett!~peter@mythtv/developer/peterbennett) has joined #mythtv

[17:15:16]

Mode for #mythtv by ChanServ!ChanServ@services.libera.chat : +v peterbennett

[17:35:27]

Steve-Goodey (Steve-Goodey!~steve@2a00:23c5:7d83:6501:6ea1:20d:5872:d872) has joined #mythtv

[17:59:32]

peterbennett (peterbennett!~peter@mythtv/developer/peterbennett) has quit (Quit: Leaving.)

[18:30:59]

peterbennett (peterbennett!~peter@mythtv/developer/peterbennett) has joined #mythtv

[18:30:59]

Mode for #mythtv by ChanServ!ChanServ@services.libera.chat : +v peterbennett

[18:32:31]

peterbennett (peterbennett!~peter@mythtv/developer/peterbennett) has quit (Client Quit)

[19:20:58]

amessina (amessina!~amessina@user/amessina) has joined #mythtv

[19:49:44]

DevMythNotifyBot (DevMythNotifyBot!~nodebot@cust32-dsl91-135-4.idnet.net) has quit (Remote host closed the connection)

[19:49:52]

DevMythNotifyBot (DevMythNotifyBot!~nodebot@cust32-dsl91-135-4.idnet.net) has joined #mythtv

[19:56:41]

peterbennett (peterbennett!~peter@mythtv/developer/peterbennett) has joined #mythtv

[19:56:42]

Mode for #mythtv by ChanServ!ChanServ@services.libera.chat : +v peterbennett

[20:22:10]

peterbennett:

The new http server is giving me "HTTPParser: Unusually long header – quitting" on scheduling, which works with the old server

[20:31:28]

stuarta:

what very long header are you sending it?

[20:36:30]

peterbennett:

http://192.168.0.115:6744/Dvr/AddRecordSchedu . . . 022-01-26+21

[20:36:30]

peterbennett:

%3A00%3A00&SeriesId=EP00517851&ProgramId=EP005178513508&ChanId=10704 &Station=WBZDT&FindDay=4&FindTime=15%3A00%3A00.000&ParentId=0&am p;Inactive=false&Season=20&Episode=88&Inetref=&Type=Single+Recor d&SearchType=None&RecPriority=0&PreferredInput=0&StartOffset=0&a mp;EndOffset=0&DupMethod=Subtitle+and+Description&DupIn=All+Recordings&a mp;NewEpisOnly=false&Filter=0&RecProfile=Default&RecGroup=Default&am p;StorageGroup=Default&PlayGroup=Default&AutoExpire=false

[20:36:30]

peterbennett:

&MaxEpisodes=0&MaxNewest=false&AutoCommflag=false&AutoTranscode= false&AutoMetaLookup=true&AutoUserJob1=false&AutoUserJob2=false& AutoUserJob3=false&AutoUserJob4=false&Transcoder=0

[20:36:40]

peterbennett:

Scheduling a recording

[20:36:51]

stuarta:

so send it in the body, rather than in the url

[20:36:58]

peterbennett:

Its not really a long header it is a l;ong url

[20:37:08]

stuarta:

true

[20:37:19]

peterbennett:

Easy for you to say

[20:37:35]

stuarta:

heh yeah, be nice to know why a long url translates to a long header

[20:37:35]

peterbennett:

This works on the old server

[20:37:55]

stuarta:

the old server has no protection against anything

[20:37:58]

stuarta:

hence it works

[20:38:23]

peterbennett:

Can we increase the size allowed for the URL ?

[20:38:45]

peterbennett:

Otherwise I have to start generating xml

[20:39:14]

stuarta:

it's worth investigating why the url is considered too long a header

[20:39:15]

peterbennett:

I can take a look see what is onvolved

[20:40:14]

peterbennett:

Maybe I don't need all those fields in the request for a schedule

[20:47:11]

peterbennett:

where are we with the freeze ? I suppose no changes should be made until the new fixes branch is created.

[20:47:49]

stuarta:

hmpf. no idea. realistically, branch needs cutting no later than tue

[20:51:28]

stuarta:

peterbennett: can you capture the conversation with the backend? i'd like to understand why the backend thinks the url is a long header

[20:52:18]

peterbennett:

Umm whats the best way to do that ?

[20:52:19]

hampton:

I have one fix to the recording extender code that needs to go in. Its done, but I haven't committed it yet. I was adding to the "tv listing name" to "espn name" translation table to handle college basketball.

[20:53:12]

stuarta:

we just have to get the branch cut, and the packaging done, so that can drop into ubuntu

[20:53:25]

stuarta:

we can still update it right up until final freeze

[20:53:39]

stuarta:

provided it's actually "in" the repo

[20:57:14]

stuarta:

peterbennett: sorry missed your question. `tcpdump -w long-header.pcap -i <interface_used_to_talk_to_client_sending_request> port 6744`

[20:57:36]

stuarta:

which should only capture the traffic sent to the backend port

[20:58:14]

stuarta:

btw. you'll need to ctrl-c tcpdump once you have sent your request, and it should tell you it's captured a non zero number of packets

[20:58:40]

stuarta:

you can open the pcap in wireshark if you wish to check what's in there, otherwise feel free to email me the pcap file

[20:58:59]

peterbennett:

So I start that up and direct it to a file then try the request ?

[20:59:06]

stuarta:

correct

[20:59:16]

stuarta:

(on the backend)

[20:59:25]

peterbennett:

Maybe I shoud just try it with a curl and see what happens first

[20:59:39]

stuarta:

do curl -v then

[20:59:46]

stuarta:

maybe even -vv

[21:00:20]

stuarta:

fyi, there's also httpie package which provides the command http, much like curl, but does pretty colours and shows you all the headers

[21:01:26]

stuarta:

ah, even from doing a quick test. i suspect i see the problem.

[21:02:42]

stuarta:

the code is considering the headers, and also the 'GET / HTTP/1.1' (in your case with a much longer url than '/') all as headers

[21:02:44]

peterbennett:

I don't get the error with curl

[21:03:00]

stuarta:

the limit is 1000 chars

[21:03:53]

peterbennett:

This gut is 1060 characters

[21:04:05]

peterbennett:

s/gut/guy/

[21:04:39]

stuarta:

i think we just have to add a count of the "headers" received in that code

[21:04:41]

stuarta:

https://github.com/MythTV/mythtv/blob/master/ . . . .cpp#L55-L68

[21:05:04]

stuarta:

this verifies that what's been sent is actually a valid HTTP starting verb

[21:05:17]

stuarta:

but doesn't pull it down, and then treats it as a header

[21:08:15]

stuarta:

peterbennett: to line 77 you could make the conditional (m_linesRead > 1 && line.size() > 1000)

[21:08:42]

peterbennett:

So will 1060 characters be OK? Can it be increased?

[21:08:57]

stuarta:

try the conditional change i suggested above

[21:10:14]

peterbennett:

Oh so it treats the url as if it were a header line

[21:10:29]

stuarta:

just checking what mdn says about the length of a header

[21:11:06]

peterbennett:

I see, so we only need to check the second line onwards

[21:11:14]

stuarta:

hold on for a moment

[21:11:34]

peterbennett:

OK I will try that and see if it fixes the problem

[21:11:54]

stuarta:

peterbennett: i'm just checking the RFC to see if there is a specified max chars

[21:12:02]

peterbennett:

ok

[21:12:08]

stuarta:

if there is then we change 1000 -> rfc value

[21:12:43]

gigem:

peterbennett: Can you refresh my memory on how to run gdb on the shield? When I run gdb.sh, I get a popup on the screen that mythtv is waiting for the debugger to connect but it never doesn.

[21:13:44]

peterbennett:

gdb.sh is supposed to first copy the executable to the shield. I suspect the latest shield os upgrade has mucked it up.

[21:14:04]

peterbennett:

I find it impossible to copy anything to the shield

[21:14:12]

gigem:

I haven't updated, at least not to the latest one.

[21:14:40]

gigem:

adb install -r works fine for me if that's what you mean by copy.

[21:15:00]

peterbennett:

Oh the other thing is I always launch mythtv on the shield first. and then connect the debugger

[21:15:44]

gigem:

That's what I remembered but it doesn't work either. I get a gdb prompt but when I press c to continue, it says the program isn't running.

[21:17:12]

stuarta:

peterbennett: i would suggest 2048 is a better value. light reading -> https://stackoverflow.com/questions/417142/wh . . . ent-browsers

[21:19:05]

peterbennett:

So maybe allow 2048 without checking line number?

[21:19:22]

stuarta:

yep, simpler change, and still enforces some kind of limit on the url

[21:19:47]

stuarta:

unlimited would be a potential DOS vector

[21:24:06]

peterbennett:

OK I will try that and also check if I really need the description when scheduling a recording

[21:25:29]

stuarta:

unlikely

[21:26:12]

stuarta:

you prob just need chanid, starttime and endtime

[21:41:47]

peterbennett:

OK that worked, just changing 1000 to 2048

[21:42:32]

stuarta:

great

[21:44:34]

DevMythNotifyBot:

COMMIT: bennettpeter pushed 1 commit(s) to branch master: https://github.com/MythTV/mythtv/compare/848b . . . 126c64eb7091

[21:45:46]

peterbennett:

stuarta: Thank you.

[21:46:09]

stuarta:

np

[21:47:39]

peterbennett:

gigem: I have not done gdb on the shield for some time now so not sure what the issue may be. It was always flaky.

[22:07:17]

peterbennett (peterbennett!~peter@mythtv/developer/peterbennett) has quit (Quit: Leaving.)

[22:09:30]

Steve-Goodey (Steve-Goodey!~steve@2a00:23c5:7d83:6501:6ea1:20d:5872:d872) has quit (Quit: Konversation terminated!)

[22:22:04]

gigem:

peterbennett: I usually use logs to help debug. Yes, I'm old and old fashioned. However, mythfrontend is crashing before I get useful logs. Hopefully, Mark Spieth will see this and comment when he wakes up. I'll try later to run the gdb.sh commands one at a time and watch more closely for errors and anomalies.

[22:43:58]

SteveGoodey (SteveGoodey!~steve@2a00:23c5:7d83:6501:fa68:6631:5ed6:d871) has quit (Quit: Konversation terminated!)

[23:07:34]

stuarta:

right, that's the basics of the program guide done. i'm calling it a night

[23:36:21]

peper03_ (peper03_!~peper03@mythtv/developer/peper03) has joined #mythtv

[23:36:37]

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

MythLogBot@irc.libera.chat :: #mythtv

Current users (28):