-
Notifications
You must be signed in to change notification settings - Fork 8
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[ISSUE] Could not connect with the skill #40
Comments
I had same problem as you. I moved the other service that was on port 5000 to another port and it started to connect to the skill. But I found another problem that I'm trying to resolve asking for it in #46 Salu2 from Spain |
Hi @adocampo, With debugging set to maximum on the skill do you see any logs on the skill when trying to invoke the skill from the Amazon developer portal? If everything is configured correctly you should see something even if it's just an error. If there are no logs generated by the skill could you temporally disable the service you have running on port 5000, run the skill on that port and see if that resolves the issue. Port 5000 is used by default by the Flask library which serves the app, this can be changed according to their documentation but would require changes to both the skill code and docker file. |
So, the yaml is right? In the Endpoint section, on the URL, I set my navidrome instance (e.g.: https://music.mydomain.com) Once built, I'm going to show how I filled the Now, it would be the time to test the skill, it fails Now, my questions:
|
UPDATED2: I've set an endpoint to the asknavidrome docker service at port 5555, and changed the endpoint URL in the developer's console for that Alexa skill, and rebuilt. Same issue. Testing says I've activated the "HelloWorldIntent", whatever that means... Logs shows something new, though, now are showing every few seconds a line of these
So I decided to test if I could reach it from inside the asknavidrme docker instance
As far as I can see, it misses "u" parameter... I'm on a dead end... any help would be much appreciated. |
Hi @adocampo, Your YAML file looks fine to me, and the skill web service confirms that as it states that it can successfully connect to your Navidrome instance in the debug logs. Also the missing U parameter is nothing to worry about, the same thing happens to me, I suspect it is due to not being authenticated with Navidrome when making the API call. The skill logs in your "UPDATED2" post are interesting, it looks like something other than Amazon is trying to connect to the skill. From what I can see something is connecting to the URL: https://your-asknavidrome-skill-instance.com/api/stats. There is no api/stats page served by the skill so a 404 error is being generated. This is not normal behaviour from Amazon. The Endpoint URL in the Amazon Developer Console you mentioned earlier should be the URL of your skill, this needs to have a valid / Amazon approved SSL / TLS certificate associated with it, a self signed certificate will not work. When you enter the URL make sure it includes the https://our-asknavidrome-skill-instance.com only, I do not believe including a port number in that box is allowed, Amazon must be able to connect to the skill on port 443 Another thing to double check is that your echo device is set to the ES locale, this must match the skill or it will not work. You need to check or set this in the Alexa app on your phone. Could you please set the NAVI_DEBUG option to 3 and then start the skill by saying "Alexa open < your skill name>" You should see output in the skill logs similar to this.
|
Everything sounds like it should work, I'm wondering if locale might be the issue. Where in that app is the screenshot from? When I configured mine I had to go to Echo Dot > Settings > Language. You'd want to choose Espanol, Espana (I apologise for misspelling the words) On the Amazon side, open the developer console and in your skill build menu you should see "Languages" in the "Tools" section. This must match your device, I think the option should be Spanish (ES). If you find you need to change anything you will need to rebuild the skill |
Yes, this is strange. Please don't worry about the version 0.6, I forgot to update the value when I published the 0.8 version. If you put your endpoint URL as it's set in the skill config on Amazon in to a browser do you see a 405 Method Not Allowed error (If you do this means the skill is working)? All your locale settings looks fine too. I'm running out if ideas, but if nothing is showing up in the logs, that means that for some reason Amazon is not able to connect to the asknavidrome skill. If it was connecting you'd see the requests and responses like in the log sample. So the issue is that the Amazon cloud service is not attempting to connect to your skill instance, but I do not know why. |
😞 Well, I need to figure then why aws cannot connect... let's wait if someone else has the same issue... :'( |
Describe the issue
I followed the steps but when testing, it cannot connect to my skill. I know I'm doing something wrong, but I cannot determine what can be. I think documentation can me confusing, at least I found it confusing.
In requirements, is stated:
But in the Troubleshooting section, it states
Nowhere specifies how to configure the web service. I modified the docker-compose.yml to match my alexa skill ID, navidrome URL, user and password, like this
And I set up it as https://naviendpoint.example.com, and I see it's reachable from internet, showing this in a browser
But no luck, when I test it, it says this
The text was updated successfully, but these errors were encountered: