We Keep Coding

Google Assistant API, controlling a light switch connected to Google Home

I have machine A that just cranks out .png files. It gets synced to machine B and I view it on machine B.
Sometimes machine A crashes for whatever reason and stops doing the scheduled jobs, which means then files on machine B will be old.
I want machine B to run a script to see if the file is older than 1 day, and if it is, then reset the power switch on machine A, so that it can be cold booted. The switch is connected to Google Home but understand I have to use the Assistant API.
I have installed the google-assistant-sdk[samples] package. Can someone show me some code on how to query and return all devices then flip the switch on and off on that device?

Unfortunately, many smart home companies are building products for consumers, not developers. Google's SDK is letting developers stream consumer voice requests to their servers and turning that into actions. Gosund, similarly, is only interested in API access for Amazon and Google. They're API is probably not documented for public use.
To answer your specific question, if you want to use the Google Assistant SDK, you would name your switch something like "Server A Switch" and record a short clip of you saying "Turn off Server A Switch" and "Turn on Server A Switch" and send those two google. The way google matches the requests with your particular account is through OAuth2 tokens, which google will give you in exchange for valid sign in credentials.
If Gosund works with Google Assistant, it has a standard OAuth2 server endpoint as well as a Google Assistant compliant API endpoint. I only recommend this if you want to have some fun reverse engineering it.
In your Google Assistant app, if you try adding the Gosund integration, the first screen popup is the url endpoint where you can exchange valid Gosund account credentials for a one-time code which you can then exchange for OAuth2 access and refresh tokens. With the access token you can theoretically control your switch. The commands you'll want to send are standardized by Google. However, you'll have to figure out where to send them. The best bet here is probably to email their developers.
Are you familiar with OAuth2? If not, I don't recommend doing any of the above.
Your other option is to prevent Server A from hardware crashes. This is what I recommend as the least amount of work. You should start with a server that never crashes, keep it that way and add stuff on top of it. If you only have two servers, they should be able to maintain many months of uptime. Run your scheduled jobs using cron or systemctl and have a watchdog that restarts the job when it detects an error. If your job is crashing the server maybe put it in a VM like docker or something, which gives you neat auto-restart capabilities off the bat.
Another hacky thing you can do is schedule your gosund plug to turn off and on once a day through their consumer UI or app, or at whatever frequency you feel like is most optimal.

The google-assistant-sdk is intended for processing audio requests.
From the doc:
Your project captures an utterance (a spoken audio request, such as What's on my calendar?), sends it to the Google Assistant, and receives a spoken audio response in addition to the raw text of the utterance.
While you could use that with some recorded phrases it makes more sense to connect to the switch directly or use a service like IFTTT. What kind of switch is it?

How to stop automatically Billing in Heroku website?

I am kind of scared of filling billing information on the Heroku website because of the automatic billing option in case I forget to remove an account.
Help me with your experience.

If you're still on free tier, it will not charge you. When I upgraded to hobby, I was only charged for how often the service was used (often less than $7). They were very honest - I have had no problems with Heroku. If you elaborate on your concerns, maybe I can provide a more tailored response.
You can stop charging at any time. If you're looking for specific information on how to remove your card/billing information, thus reducing down to the free tier you can change your dyno type in "Resources" on the Heroku web app.

how to design rest api which handle offline data

I have multiple api which we have provided to android developers.
Like :
1) Creating Business card API
2) Creating Contacts API
So these api working fine when app is online. So our requirement is to handle to create business card and contacts when app is offline.
We are following steps but not sure:-
1) Android developer store the business card when app offline and send this data to server using separate offline business card api when app comes online.
2) Same we do for creating contacts offline using offline contact api.
My problem is I want do in one api call to send all data to server and do operation.
Is this approach will right?? Also please suggest what is the best approach to handle offline data. Also how to handle syncing data when app would come online??
Please let me know if I could provide more information.

I'm confused as to how you're approaching this. My understanding is that when the app is offline you want to "queue up" any API requests that are sent.
Your process seems fine however without knowing the terms around the app being "offline" it's hard to understand if this best.
Assuming you're meaning the server(s) holding the application are offline you're correct you want a process in the android app that will store the request until the application becomes online. However, this can be dangerous for end users. They should be receiving a message on the application being offline and to "try again later" as it were. The fear being they submit a request for x new contacts to be queued and then re-submit not realizing the application was offline.
I would suggest you have the android app built to either notify the user of the app being down or provide some very visible notification that requests are queued locally on their phone until the application becomes available and for them to view/modify/delete said locally cached requests until the application becomes available. When the API becomes available a notification can be set for users to release their queue on their device.

Changing a program to be used on a webserver

Say I have code written in python that analyzes files on my computer and returns a result. It works great locally on my HD, but now I'd like to turn it into a mobile app. This means I'll require a server of some kind (cloud for instance) where users can access it.
It is my understanding that all that would be required is a method to grant user credentials and permissions to the patrons so they can access the "run" command in my analysis program. But honestly, I have no ZERO visibility in this area and don't really know where to begin.
I only have two questions:
Users & their credentials are endless, but they all have to share the same analysis program. I don't know much about servers, but wouldn't this method cause long queue times? Generally-speaking what considerations would I have to make in my analysis code to avoid this?
Can someone just point me in the direction of what I'd need to learn in order to answer the above question? This topic is a bottomless pit of information and I don't wanna get trapped.
Thanks.

Django is an MVC Web framework which possesses all features required for doing Web applications with Python. Simply go through the tutorial and you should be up and running in no time, on your local machine.
To deploy there are various options, be it a cloud instance (a lot of providers here, including Rackspace and Amazon, Google for "django web hosting"), or "traditional" server machines (again a lot of providers here).
The "mobile" part is just the user interface. This affects decisions in the presentation part of your application, and you can restricted this to the View part in Django jargon (i.e. the HTML templates) of your Web application. You can look for frameworks which allow the production of aesthetically decent (or better) user interfaces HTML tailored for mobile/tablet devices, e.g. JQueryMobile.
Therefore direction: start with Django -> deploy on a server "somewhere" -> tailor your user interface for mobile devices.

GAE Python2.7 client authentication by certificate

I'm writing an application on GAE that is exposing a RESTlike API to a fixed number of remote servers which may be using any OS/software. At the moment I'm pondering how to identify and authenticate these remote servers painlessly.
I'm trying to avoid having to program too much of this myself for obvious security concerns.
If I'd were on an Apache or nginx I'd use SSL client certificates and let the clients choose whatever they want to contact the API, curl, webapp, whatever.
I understand that, at this time, GAE doesn't provide checking client certificates this way.
Is there any other way to do this in GAE with Py2.7?
If this is not possible or a very big hassle, can anyone point me to another good way to identify and authenticate remote servers in a situation like this?
I can only post two links of what I checked
https://groups.google.com/forum/?fromgroups=#!topic/google-appengine-python/c5PHMrAMAcI
https://sites.google.com/site/oauthgoog/authenticate-google-app-engine-app
plus much more over the last few days. I found some questions which are similar to mine, but none with a satisfying answer, to me at least.

I'd suggest looking into Google Cloud Endpoints, which is currently in the trusted tester phase. I discuss it in more detail in this answer, but the main things you'll get are:
An easy way to define your API
Automatic support for OAuth 2
We're primarily targeting developers building APIs in their application backends, but other "same party" (e.g. you're the API developer and consumer) uses should work equally well.