Hi
I need to prepare a demo application for a potential client, which will allow us to get into paid subscription with Blynk. Our target is the StartUp package, although i’m suspecting that Blynk Faces maybe needed at the end.
The application concerns monitoring and control of several refrigeration controllers, each one equipped with an ESP8266 board (NODEMCU) which will be the interface between controller and Blynk Server.
The application will use Tiles with Templates. I’m not in a position at this time to tell if Dynamic or Static auth token provisioning shall be used but Blynk.Inject framework will be used through the Device info menu of the Template.
So, the first question is, will BlynkProvisioning work from my local server or do i have to host this demo project in Blynk Cloud.
Second, i’m about to include BlynkProvisioning library into my project and i’m thinking to paste all these header files residing into /examples/Blynk.Inject/Template_ESP8266, into my projects directory. Is this the right way?
We already implemented projects like this and they work really well. StartUp plan is the most suitable I think, but if you are ok with Android only and expect low volume of devices, you can check out Demo Package.
The application will use Tiles with Templates.
This is how it should be
I’m not in a position at this time to tell if Dynamic or Static auth token provisioning shall be used but Blynk.Inject framework will be used through the Device info menu of the Template.
When creating a template you get/create template ID which is later used in the firmware file Settings.h. Then Blynk does everything for you.
will BlynkProvisioning work from my local server or do i have to host this demo project in Blynk Cloud.
Blynk keeps up a dedicated server for you. So no worries here. It’s yours, together with all the data.
Second, i’m about to include BlynkProvisioning library into my project and i’m thinking to paste all these header files residing into /examples/Blynk.Inject/Template_ESP8266, into my projects directory. Is this the right way?
That’s correct. However, once you subscribe, an engineer from Blynk will be assigned to you and will guide you through the process. They will also review you code (Blynk part of it) if needed.
You can email us to iot@blynk.cc or simply sign up and we will take from there.
Blynk keeps up a dedicated server for you. So no worries here. It’s yours, together with all the data.
@kostas some clarification - we do this because we have an automated process of deploying new server and it costs us nothing. Otherwise (hosting on the end users server) we have to explain a lot of configurations things like “mail configuration”, “firebase configuration”, "domain configuration’, “monitoring approaches”, “backups” etc. And there is always a chance end user will make a mistake.
@Dmitriy i will completely agree with you.
When someone requests a subscription fee as we hope we will agree with the “client” if the project succeed, he needs to have as less infrastructure problems as possible. At the end of the day, they will ask for connectivity success, recovery from a server shut down, data backup and several other details, which only with your cloud server can be secured. Plus the fact of compliance between server-library-Application.
We keep our own local Server for demonstration purposes as well as development and/or improvements.
The question if BlynkProvisioning will run in a local server has to do with practical reasons during the erection of the demonstration project.
If no, i can easily use my Blynk Cloud account. The only inconvenience is the login-logout process from local to cloud and vice versa.
Dear @Pavel
I’m here all the time actually, though silent watcher of the lists.
Thank you very much for your information. it helps a lot to the realization of this demo project.
A target is always accompanied with one or more challenges.
Our challenge is that with such a project we will be able to “win” the subscription for a long time, which then allow us to attract more clients of this kind.
On the other hand, that means we will have to present a working demo before we get into subscription relationship with Blynk.
Fortunately Tile-Template model is well understood, while most Templates have been already developed as stand alone applications, so, its easy to rewrite.
What remaining to be done is mainly BlynkProvisioning and minor changes to the UI. We used to use WifiManager for the standalone ESPs but the seamless integration of provisioning with the Application is unbeatable. I will get into this right now!
One more thing which is essential in this kind of projects, is that number of devices and consequently Tiles is not constant. It depends on client’s demands and should be configured during the setup, case by case. The worst case scenario would be to use a static grid of empty tiles with equal number of auth-tokens and fill it per demand!
The ideal would be to be able inside the application to arrange the grid and consequently number of tiles per wish on the fly, even better to be able to add tabs, as well as add-remove devices accordingly.
I don’t know if such a facility can be realized in the StartUp package, or Blynk.Faces or something completely new needed.
In commercial-ready app number of tiles changes with every added new device. There is no need to plan it upfront. Check this help article, I hope it will give more understanding of that.
So basically, if you have 3 clients, they create 3 accounts, and under each of these accounts there can be a different number of devices(tiles) as they purchase or remove them. Moreover, devices can be of different type.
Inside of the tile there can be pre-defined tabs if they are a part of the template.
Dear @Pavel thank you very much for your time.
Indeed it looks like i’m fully covered.
Furthermore i will proceed to the offer based on Startup subscription.
Hi
I 've started evaluation through myApps and preview.
Well. Tiles with templates is fantastic!
The ability to add-remove or even move devices from project to project without all this mesh of auth-token handling, is unique (dynamic provisioning).
The only “drawback” is that i have to work with 2.4.2 version of esp8266 library instead of the 2.5.2 current version, in order to avoid “ISR not in IRAM” cause of continuous restart of NodeMCU.
Furthermore i was wondering how many devices can be added in a single tiles grid frame.
Additionally BlynkProvisioning uses some EEPROM space in order to save its working parameters.
I also use EEPROM to store configuration of my device, and it looks like there is an overlap.
I was wondering what is the area occupied by BlynkProvisioning.
Our engineers can answer the tech questions. E.g. @vshymanskyy
Re: tiles
1 tile = 1 device and vice versa. There is no way currently to make 1 tile to contain multiple devices. Think about tiles as a list of devices.
Dear @Pavel, i’m aware that tile is a single device container
I was wondering how many devices (consequently tiles) could be added at maximum (via the process “Add Device”)
@Pavel , that’s a good question!
Having too many is not practical to view at a glance.
Let’s say two tabs with 15 tiles each (thirty devices total) could be feasible?
Dynamic provisioning is fine, but is there any way to retrieve auth-tokens of the devices (ESP8266 in my case) following provisioning.
I mean, in MyApps preview i was not able to find the tokens given to the provisionned devices.
Besides mobile app i use some own software like
historical data downloader, processor and reporting tool, or a monitoring and configuration tool. They both communicate with the Blynk Server via REST API. Auth-token is a mandatory part of the URL, unless there is some other approach in business subscription
Without knowing all the details I think it can be achieved by using our Enterprise plan (which is next version of Blynk), but I don’t know if it fits your budget.
As I mentioned, you can get tokens from the server anyway.