I didn't bother with a container. I had it running on my Linux box inside of about 10 minutes.
Avoiding Heroku's $5 charge is more than just the charge... it's the principle of the matter. And ensuring you're in control of your own ecosystem (that's Hubitat's major sales pitch - reduced cloud dependence).
Honestly, if the choice was between paying $100 for a local instance and paying Heroku $5/mo indefinitely, I'd choose the local instance any day. That it strips Salesforce of additional customer count / revenue is a nice bonus.
Fair enough, mine is too. But learning a little bit about Linux and docker containers for this and other home automation or server tasks is part of the hobby aspect of all this for me.
To be clear, I have absolutely no background in computer coding or IT.
You are seeing only the problematic aspects here when people ask for help. It’s much harder to say what % of people get it working without a hitch or with minimal hassle.
From what I can tell, where things go sideways for people is often related to the quirks of how docker containers on various platforms interact with the network, which can differ slightly or even greatly.
And as @doctorkb just mentioned, sometimes running the server without docker (there are step-by-step instructions for that here too) is easier and meets the same need.
ETA: but in the end, there’s nothing wrong per se with paying a subscription charge to host this server if that’s what someone decides provides the most value to them. But obviously this is a crowd that, all other things being equal, generally prefers a locally run option to anything in the cloud.
I just wish we had a different topic for those discussions. This topic is completely hijacked by Docker issues by now. I am on heroku and see contant exceptions in the logs, but my call for help was burried in container issues in a matter of minutes. Oh, well.
@jbaruch I totally agree with you. There should be a separate thread on server setup issues and containers etc I too have had questions missed. This thread should be about Echo Speaks itself, how it works, how people use it, getting the most out of it. So much valuable info has been lost, and I'm sure I'm not utilizing not even 5% of what this great project can do.
Yeah I’ve sunk more time than I should have into figuring this out, but on the plus side I’ve learned a ton about docker lately and have been able to apply that toward other projects. So, hobby time isn’t completely wasted.
Anyway, I’ll move docker talk to the other threads to stop hijacking’s this one
The Echo Speaks tile on my dashboard - is there a way to reflect the volume numerically?
I've been tweaking all my echo's and there are small diffs - increments of 5% or so and I can't think of a way to easily monitor them - hmm. maybe a smart tile?
Has anyone figured out how to restart the docker on a QNAP after an update? Mine just updated the cookie and then stopped. I tried the mentioned "docker update --restart=unless-stopped echo-speaks-server" from the docker terminal but it did not like it:
Docker isn't a command available within the container itself. You will need to SSH to your NAS and run the command there. I have some automation happening within NodeRed for another container and the patch should be similar to the following that I use for mine: /share/CACHEDEV1_DATA/.qpkg/container-station/bin/docker update --restart=unless-stopped echo-speaks-server
I haven't tested this myself as I am restarting mine via NodeRed but curious if this will work for you.
Thats old info from when this was first launched as v4, the instructions are for people upgrading from v3 to v4. This was also before the Unmatch feature in HPM, which now could be used to accomplish the same thing without deleting the app.