[RELEASE] Echo Speaks V4

I'm sorry for bringing this up. I know it's been brought up many times before. However, from a traditional deployment of the cookie server that's definitely changed. In addition, I have wiped out my Raspberry Pi server and dedicated it to this deployment and it's been months and I still cannot get the locally installed server to work.

Let's start with the native server on Heroku. After attempting to try to get the locally installed cookie server to work I attempted to fall back to Heroku. I did a redeploy on that platform like I've done many times before and it would not find the URL and give you the following view.

When I reviewed the app settings the URL is not presented the same way as it used to be. If you're already deployed this is not an issue only for new deployments.

If I follow that link the server does appear.

So, I don't see a way around this problem.

And now for the locally installed server, I've tried with docker without docker and either I'm in a login loop where it doesn't take the cookie, or it does and then reverts back to being unconfigured or back to Heroku.

I rebuilt my Raspberry Pi to follow the following setup for Portainer / Docker-Compose.
And I cannot get the Hubitat application to recognize the login. According to the log files with the container I do get a cookie, but the Hubitat application does not see it.
image

The page does display with the container popup.

Login

Back to

image

I have tried everything, and I cannot seem to find where the problem is.
Now, I cannot move forward or backward. I’m now just stuck.

Did you follow my tips at the bottom of the guide?

  • If you get red error text that you still need to login:
    • Restart the docker container
    • Click the Amazon Login Page button
    • Log in again (do not click Save at bottom)
    • Close the server login tab, and the main app should refresh
1 Like

Drove me crazy but it seems like that's working. Thank you very much.
I hope I don't have the secondary issue where it loses its mind and it won't refresh the cookie or thinks it's back to heruko.

Thanks!

Firstly thanks for so quickly adding the new echo pop device after I submitted the details!

I have an issue with using the “search Sirius XM” function on a WHA speaker group. Does anyone have any suggestions? It won’t play, if I select speakers individually it will. I was thinking of using an Alexa routine and calling that.. any other ideas?

Thanks

Everyone should know,, you have to enter a credit card for this.
As it is not "free" cheap but not free.

1 Like

Its free if you host your own cookie server: [GUIDE] Echo Speaks Server on Docker (or without)

3 Likes

If you are referring to the Heroku service then there are quite a few post about that. It use to be free,but they started charging back in Nov of last year.

This app is free if you use your own a local server, as I do and many others. There are a number of post about how to do that as well.

3 Likes

Yes that is what I was speaking of.
Thanks.

I see tons of errors in the logs out of the blue (running vanilla service on Heroku).

It started with 408 errors on cookie refresh (for 2 weeks now), and today I also got this:

app:682023-07-24 10:24:20.656 PMerrorEchoApp (v4.2.2.0) | wakeupServerResp Exception: java.lang.Exception: No response data exists for async request
app:682023-07-24 10:24:20.652 PMerrorEchoApp (v4.2.2.0) | wakeupServerResp | Stack Trace: java.lang.Exception: No response data exists for async request at hubitat.scheduling.AsyncResponse.getData(AsyncResponse.groovy:71) at user_app_tonesto7_Echo_Speaks_163.wakeupServerResp(user_app_tonesto7_Echo_Speaks_163.groovy:2228)
app:682023-07-24 10:24:20.624 PMerrorEchoApp (v4.2.2.0) | wakeupServerResp Server may be down / unreachable
app:682023-07-24 10:24:20.613 PMwarn EchoApp (v4.2.2.0) | wakeupServerResp: 503

Any ideas about what's going on?

According to one of the devs up above those errors are from the app trying to connect with the cookie server, which is failing. Same errors were posted above and discussed.

Thanks for the reference, @jtp10181. I saw those discussions. Most of them are about troubles people get setting up standalone servers. I use the vanilla Heroku, and the code is up to date (according to github, at least), so I am unsure how to debug the server errors.

P.S. That's why I advocated separate the threads :wink:

The 503 response indicates that your cookie server is not responding. It seems that your Heroku instance is not responding to requests.

That's what I figured... But it looks good from what I can tell. What are the debugging procedures for it?

That actually isn't good. You should be seeing that you are logged into Amazon. You need to complete that step. Click on the Goto Login Page button and complete the steps to login to your Amazon account. Step 7 in the guide below.

Duh, obviously, I was logged in (not sure when I logged out). I logged back in, and forced cookie refresh, but the cookie is still not refreshed:

It may take a bit for that page to refresh. You might need to exit and go back in.

I see it trying:

If you keep getting those wakeup server response errors, you still have something going on with the Heroku side of things. It has been quite some time since I used Heroku and I no longer have an account, so I can't go and look. I would go back through the guide I posted and make sure you have everything set correctly.

It's been working for at least a few years, so I am sure the setup is good. Heroku logs show the app crashes, and when it recovers, I am logged out (that's how I ended up not being logged in). Maybe Heroku service troubles though... :man_shrugging:t3:

You might want to consider standing up you own cookie server. It is relatively easy to do if you have the resources and follow the guide/s posted by @jtp10181.