[newoski]

Hi Guys,

I've tried setting this up twice now, with no luck. I end up getting the "The requested skill took too long to respond" error.

I believe the problem is with Heroku, as I'm unable to access any of the URLs for my app. I get the following error when testing in a browser:

The xxxxx-yyyy-#####-herokuapp.com page isn’t working

I'm completely stumped. Any help would be greatly appreciated!

[newoski]

Hi Guys,

I've tried setting this up twice now, with no luck. I end up getting the "The requested skill took too long to respond" error.

I believe the problem is with Heroku, as I'm unable to access any of the URLs for my app. I get the following error when testing in a browser:

The xxxxx-yyyy-#####-herokuapp.com page isn’t working

I'm completely stumped. Any help would be greatly appreciated!

My logs clearly show a problem, I'm stumped about how to fix it though...

2016-10-29T17:41:53.267084+00:00 app[web.1]: [2016-10-29 17:41:53 +0000] [3] [CRITICAL] WORKER TIMEOUT (pid:12)
2016-10-29T17:41:53.283393+00:00 app[web.1]: [2016-10-29 17:41:53 +0000] [12] [INFO] Worker exiting (pid: 12)
2016-10-29T17:41:53.341412+00:00 app[web.1]: [2016-10-29 17:41:53 +0000] [15] [INFO] Booting worker with pid: 15
2016-10-29T18:14:16.357689+00:00 heroku[web.1]: Idling
2016-10-29T18:14:16.358298+00:00 heroku[web.1]: State changed from up to down
2016-10-29T18:14:20.010432+00:00 heroku[web.1]: Stopping all processes with SIGTERM
2016-10-29T18:14:20.926964+00:00 app[web.1]: [2016-10-29 18:14:20 +0000] [15] [INFO] Worker exiting (pid: 15)
2016-10-29T18:14:20.927527+00:00 app[web.1]: [2016-10-29 18:14:20 +0000] [9] [INFO] Worker exiting (pid: 9)
2016-10-29T18:14:20.928258+00:00 app[web.1]: [2016-10-29 18:14:20 +0000] [3] [INFO] Handling signal: term
2016-10-29T18:14:20.964385+00:00 app[web.1]: [2016-10-29 18:14:20 +0000] [3] [INFO] Shutting down: Master
2016-10-29T18:14:21.075531+00:00 heroku[web.1]: Process exited with status 0

After restarting all Dynos... I get

2016-10-29T18:14:20.964385+00:00 app[web.1]: [2016-10-29 18:14:20 +0000] [3] [INFO] Shutting down: Master
2016-10-29T18:14:21.075531+00:00 heroku[web.1]: Process exited with status 0
2016-10-29T19:47:11.892482+00:00 heroku[web.1]: State changed from down to starting
2016-10-29T19:47:15.235470+00:00 heroku[web.1]: Starting process with command `gunicorn wsgi:application`
2016-10-29T19:47:17.543987+00:00 app[web.1]: [2016-10-29 19:47:17 +0000] [3] [INFO] Starting gunicorn 19.6.0
2016-10-29T19:47:17.544597+00:00 app[web.1]: [2016-10-29 19:47:17 +0000] [3] [INFO] Listening at: http://0.0.0.0:15627 (3)
2016-10-29T19:47:17.544720+00:00 app[web.1]: [2016-10-29 19:47:17 +0000] [3] [INFO] Using worker: sync
2016-10-29T19:47:17.552478+00:00 app[web.1]: [2016-10-29 19:47:17 +0000] [9] [INFO] Booting worker with pid: 9
2016-10-29T19:47:17.569492+00:00 app[web.1]: [2016-10-29 19:47:17 +0000] [10] [INFO] Booting worker with pid: 10
2016-10-29T19:47:18.830088+00:00 heroku[web.1]: State changed from starting to up
2016-10-29T19:47:39.453056+00:00 heroku[router]: at=info method=GET path="/" host=xxxxx-xxxxx-xxxxx.herokuapp.com request_id=8be1ec2d-1420-4a02-ac8f-38fb30ef3154 fwd="68.80.6.140" dyno=web.1 connect=1ms service=3ms status=502 bytes=141

[fonzie]

I'm trying to add custom intents for Kodi such as such as logging out of user profiles and opening external programs like Netflix.

This is what I wrote to try to get the system to log out of a profile but Amazon encounters an error when I try to save the skill.

Code:

{
      "intent": "System.LogOff",
      "slots": []
    },

and for netflix I added:

Code:

{
      "intent": "StartAndroidActivity("com.netflix.ninja")",
      "slots": []
    },

I think the error with the first code is the period and the error with the second code is the dual parenthesis.

How do I overcome these problems? Am I even going about this the right way?

[nawo69]

Is the check new shows broken on the latest version for everyone? as I can't get it to work since updating and if I go back to the other branch it is working again

Sent from my SHIELD Tablet K1

[jingai]

Is the check new shows broken on the latest version for everyone? as I can't get it to work since updating and if I go back to the other branch it is working again

I can check when I get home, though I believe I used it yesterday.

If there is an error from the skill in your server log, please post it so I can see what's going on.

[nawo69]

Application Logs

2016-10-30T19:10:27.477349+00:00 app[web.1]: File "/app/wsgi.py", line 1352, in wsgi_handler2016-10-30T19:10:27.477349+00:00 app[web.1]: response = on_intent(alexa_request, alexa_session)2016-10-30T19:10:27.477350+00:00 app[web.1]: return one_intent[1](intent_slots)2016-10-30T19:10:27.477350+00:00 app[web.1]: File "/app/wsgi.py", line 1273, in on_intent2016-10-30T19:10:27.477351+00:00 app[web.1]: File "/app/wsgi.py", line 117, in alexa_check_new_episodes2016-10-30T19:10:27.477352+00:00 app[web.1]: new_episodes = kodi.GetUnwatchedEpisodes()2016-10-30T19:10:27.477353+00:00 app[web.1]: show_info[show] = GetTvShowDetails(show=show)2016-10-30T19:10:27.477360+00:00 app[web.1]: TypeError: GetTvShowDetails() got an unexpected keyword argument 'show'2016-10-30T19:10:27.477352+00:00 app[web.1]: File "/app/kodi.py", line 540, in GetUnwatchedEpisodes2016-10-30T19:31:35.715549+00:00 heroku[router]: at=info method=GET path="/" host=radiant-thicket-xxxxx.herokuapp.com request_id=29624c3d-775c-4b96-a76a-550217803885 fwd="54.163.75.67" dyno=web.1 connect=1ms service=2ms status=502 bytes=1412016-10-30T20:02:20.999671+00:00 heroku[router]: at=info method=GET path="/" host=radiant-thicket-xxxxx.herokuapp.com request_id=3b9a54c0-084b-4240-9e4f-f39f8616b6e2 fwd="54.205.243.213" dyno=web.1 connect=1ms service=3ms status=502 bytes=141



Sent from my SHIELD Tablet K1

[jingai]

Fixed in 1e59125.

[nawo69]

Fixed in 1e59125.

Thanks mate working fine now.

I do have another issue with the latest versions; I have to ask it twice to play anything... It looks like it is working and alexa confirms it is playing but Kodi just flashes opening then nothing but if I ask the exact same thing again it plays fine.

Again if I roll back to other branch it works fine. Will stick a log on in morning

Sent from my PLK-L01

[jingai]

Hmm. It's not doing that for me. I just checked it with three movies and two show episodes -- they all played on the first attempt.

[nawo69]

Hmm. It's not doing that for me. I just checked it with three movies and two show episodes -- they all played on the first attempt.

Yeah strange... I will do some more testing tonight then as I cant find the log history in Heroku from yesterday anyway.

But I did test a few times last night and c7367ba was fine but 6307d27/1e59125 had to ask twice; though navigation commands like play/pause/stop etc did work fine first time

[jingai]

But I did test a few times last night and c7367ba was fine but 6307d27/1e59125 had to ask twice; though navigation commands like play/pause/stop etc did work fine first time

Would need to see the logs to verify that Kodi is actually receiving the command (put Kodi in debug mode and check its log, too).

But also, what version of Kodi are you running? We did recently switch from creating a temporary playlist to simply passing the movie ID to Player.Open, and it's possible it's got a problem in some version of Kodi.

[nawo69]

But I did test a few times last night and c7367ba was fine but 6307d27/1e59125 had to ask twice; though navigation commands like play/pause/stop etc did work fine first time

Would need to see the logs to verify that Kodi is actually receiving the command (put Kodi in debug mode and check its log, too).

But also, what version of Kodi are you running? We did recently switch from creating a temporary playlist to simply passing the movie ID to Player.Open, and it's possible it's got a problem in some version of Kodi.

Ok will do that tonight.

Im using the latest version of SPMC

[jingai]

I assume it's Jarvis then, so it could potentially be a bug in a) Jarvis, b) the SPMC fork, or c) the Android port.

[nawo69]

I assume it's Jarvis then, so it could potentially be a bug in a) Jarvis, b) the SPMC fork, or c) the Android port.

oh and my library is managed by Emby too if that makes a difference. Lots of logs later then

[fonzie]

My library is managed by emby as well and the heroku server works with it just fine.

Sent from my HTC 10