Tag Archives: development board

Sensors Monitoring – Nodered Configuration

At this point we set nodered to read the messages on Mosquitto and insert the data into InfluxDB and MySql.

Below the json file to be imported into nodered

Nodered Json

This flow is functional to the explanation of the functioning of the architecture as a whole; the flow must be optimized by programming nodered so as to automatically generate the information of the nodes from which the data arrives without resorting to different nodes for each sensor.

To execute the import proceed as follows, after accessing nodered

http://dashboardora.local/nodered

and after clicking on the menu on the right, choose to import the json from the clipboard

importFlowsAt the next screen, choose to import the json file into a new flow

flowOnce imported, set the credentials to access MySql and InfluxDB, selecting Configuration nodes from the menu on the right and setting the user account and password

configureNodesThe connections to the dbs and mosquitto are shown

configureNodesDBBelow is the MySql setting, accessed by clicking the mouse twice on ESPMYSQL

MySqlSettingUpdate button updates the node.

To make the flows operational click on Deploy

deploy

Now, with nodered we take the contents of the various mqtt topics and insert them into MySql and InfluxDB.

In the next article we display these data in Grafana.

Sensors Monitoring – InfluxDB Configuration

After MySql tables creation

Sensors Monitoring – MySql Configuration

we have to create a database  on InfluxDB.

The first step is to activate an administrative user on the system, as described in the link

Authentication and Authorization

InfluxDB  is accessed from the local client by running the influx command from a shell

Run

In the file /etc/infuxdb/influxdb.conf, using the root user or the sudo command, the auth-enabled parameter is set from false to true

Restart influxdb

Using a command shell on Debian  create the database DATAESP with the following commands

Enter the admin credentials to the auth command and proceed to create the database

Below is the result of the commands listed above

datainfluxdb

For the management of the Influxdb database, it is possible to use Chronograph

Chronograph

The Chronograph download is available at the link

Chronograph Download

Installation can be performed in two ways.

Application start

Log in with the monitor user and in a command shell

Start the chronograf application

 

Start as a Service

Log in with the monitor user; use a command shell to execute

In the /opt/chronograf/usr/lib/chronograf/scripts folder there are scripts for the automatic start of chronograf.

Here are the operations performed in our system:

  • Copy /opt/chronograf/usr/lib/chronograf/scripts/init.sh into /etc/init.d/chronograf and assign execution permissions

  • Copy /opt/chronograf/usr/lib/chronograf/scripts/chronogaf.service  into /etc/systemd/system/chronograf.service

Change in both files the user and path of the various chronograf components, i.e. change the user to monitor and the paths that point to /opt/chronograf

Below is the link to the modified files

Automatic Startup Chronograf

After the modifications run

To enable it at startup and start it run

After starting chronograf in one of the ways described above, it can be accessed via browser and port 8888

http://dashboardora.local:8888

ChronografPress Get Started and set the connection and the credentials of the admin user with relative password defined before

Chronograf2conSet up the Influxdb dashboard

ChronografDashboardDo not set the Kapacitor component

Chronograf3Complete the configuration and by clicking on View All Connections

Chronograf4we access to the newly created connection

ChronografConnections

This component can be useful later to manage and monitor InfluxDB.

In the next article, we ‘ll define the rules on Nodered to populate MySql and InfluxDB databases with the data sent by the sensors

Sensors Monitoring – Nodered Configuration

 

 

 

 

Sensors Monitoring – MySql Configuration

After the software installation on Orange Pi PC we proceed to the their configuration

Sensors Monitoring – Envinronment Setup 2

Creating Tables on MySql

On MySql 3 tables have been created containing the values of temperature, humidity and pressure of the sensors.

Here is the MySql model

MySql Model

to import into MySql Workbench

ESPModelloOnce imported, access the Database-> Forward Engineering menu and import the tables on your MySql system on the Orange Pi PC, by entering the MySql root user and relative password

ESPImportAt the next screen we  use the connection created on MySql Workbench to connect to the database

ESPImport1Leave everything by default and enter the root password when prompted.

The tables are now available on the database

ESPTablesIn the next article we configure InfluxDb to store data in real time sent by the sensors

Sensors Monitoring – InfluxDB Configuration

Sensors Monitoring – Envinronment Setup 2

After the software components installation in the previous article

Sensors Monitoring – Eninronment Setup

we finish this operation with this article.

Infludb

The Influxdb is installed as indicated in the link

Influxdb

For Debian 9 we execute the following commands

Check that everything works correctly by connecting with the infludb client

infux

Nodered

We proceed to the installation of node.js

Then we proceed to  nodered installation, as indicated at the link

Nodered installation

To run nodered at the start up, we use the script available at the following link

Nodered autostart

We did not run the automated script, but executed the following steps

Edit /etc/init.d/nodered and change the user from pi to monitor, ie change the parameter

in

Run

Start  nodered

Check that nodered works by accessing the following link with the browser

http://dashboardora.local:1880

nodered

Restart the system and check that nodered starts automatically.

Proceed to the installation of the nodered add-ons

 

Grafana

We proceed to download the version for arm

Grafana for ARM

Enable and start Grafana

Access the system using the url

http://dashboardora.local:3000

Log in with the user admin and password admin; change the password

GrafanaOnce the new password has been set, the setup interface is available

Grafana2In the next articles we will execute the settings to be able to view the sensors data.

Reverse Proxy installation

To access the system through port 80 of the http server we use nginx as a reverse proxy

nginx

We disable apache

We proceed to the installation of nginx

Everything must be configured to access the two systems using the following links

http://dashboardora.local/nodered for nodered

http://dashboardora.local/grafana for grafana

For Grafana we follow the official documentation

Grafana proxy

We change the file /etc/grafana/grafana.ini by modifying the domain and root_url parameters; in our file the parameteres are modified  as follows

Then we configure nginx, creating the servizi file with the following contents in the /etc/nginx/sites-available folder

Here is the link of the nginx configuration

nginx configuration

We run the following command

and start nginx and grafana services

At this point it is possible to access the two services via the links

http://dashboardora.local/nodered

nodered-nginx

http://dashboardora.local/grafana

grafana-nginxIn the next article we start the configuration of the various modules needed for the sensors monitoring

Sensors Monitoring – MySql Configuration

Sensors Monitoring – Envinronment Setup

In setting up the infrastructure for monitoring sensors made up of ESP8266 devices in previously published articles

Mqtt with Beaglebone ed ESP8266 – Freeboard

Freeboard had been set up to display the data coming from the sensors.

In this new series of articles, we  set up Grafana

Grafana Dashboard

as  environment to view the data of the various sensors.

We implement the environment, in this case,  on an Orange Pi PC board, but other arm or pc system can be used.

Below the architecture used in this series of articles

archGrafana

In these articles we’ll use the maximum authorizations for each environment, we will not use authorizations on mosquitto and no encrypted connections will be used. In a test environment this setting is fine, but in an operating environment it is necessary to set up the correct authorizations and possibly the encryptions in the connections in all environments.

We have installed our compiled version of Debian 9 Stretch for Orange PI PC. The image used is the following

Image Download

The monitor user was created during the Image setup, while the hostname was renamed in dashboardora. If not installed proceed to install the avahi daemon

In this way the services can be accessed by publishing the hostname, i.e. dashboardora.local.

Using the armbian utilities  disable ip6 on the system; to carry out this setting launch armbian-config

ipv6Disable1Proceed to diasble IPV6

ipv6Disable2

As a first step we install the  mqtt broker mosquitto

Install the http services and the databases Mysql and MySqlWorkbench

Apache2

and various prerequisites

Start apache

and check that the server responds to the web address, which in our case corresponds to http://dashboardora.local

apache

MySql

Debian 9  comes with MariaDB; you must set the password for the database administration user root with the following command

We set the root password and the ability to log in from other hosts as root

mariadb-secureThe other chosen options are listed below

mariadb-secure2

We install phpmyadmin for database management

At the next screen choose apache2

phpmyadminFinally choose NO

phpmyadmin1

Run the commands listed below

Check that the system works correctly by accessing via web browser by adding to the apache phpmyadmin address, in our case

http://dashboardora.local/phpmyadmin

phpmyadmin2

Log in with the database root credentials.

Mariadb must be enabled to accept connections from other hosts. To do this, go to the folder

and edit the 50-server.cnf file as root, changing the setting

in

To allow the root database user to be able to access from an external client, it is necessary to log on mariadb locally with his client and enable this option; therefore the following is performed

 

Mysql Workbench Installation on PC

Go to the following link

MySql Workbench

and download the version for your operating system; proceed with the installation.

For Linux the package is also available in the distribution repositories.

After configuring the connection, you access the MySql system control screen.

MySQLWorkbenchIn the next article we will continue the installation of the other components to be able to monitor the sensors

Sensors Monitoring – Envinronment Setup 2

ESP32 – IDF Sdk with Eclipse

After configuring the IDF development environment in the previous article

ESP32 – IDF Sdk

let’s see now how to use Eclipse as IDE for code development. Instructions are available at the link

Eclipse Setup

Download the  C/C++ Eclipse  for Linux

Eclipse Download

Unpack the file and move it to /opt/

Run ecplise

Set the workspace for eclipse

eclipse-workspaceOnce the IDE is open, select File -> Import -> C / C ++ – -> Existing Code as Makefile Project

eclipse-importPress Next; on the next screen select the folder containing the Makefile of a sample project in the “Existing Code Location” field and choose CrossGcc as “Toolchain for indexer settings” as shown below

eclipse-import2Once the project is loaded, using the mouse right button on the project name, select the properties option

eclipse-project-propertiesUnder C / C ++ Build -> Envinronment add the variable

eclipse-batchAdd the IDF_PATH variable pointing to the downloaded IDF SDK in the same section

eclipse-idf-pathEdit the PATH variable and add the path to the bin folder of the cross compiler /home/sviluppo/Esp32/espressif/esp/xtensa-esp32-elf/bin

eclipse-pathAccess “C / C ++ General” -> “Preprocessor Include Paths”; choose the Providers tab and click on “CDT Cross GCC Built-in Compiler Settings”

eclipse-cross-cdtReplace $ {COMMAND}

eclipse-cross-cdt1with xtensa-esp32-elf-gcc

eclipse-cross-cdt2Choose now, always in the providers section,  “CDT GCC Build Output Parser”

eclipse-cross-cdt3and insert at the beginning of the “Compiler command pattern” xtensa-esp32-elf-

eclipse-cross-cdt4Save the configurations.

The project settings should always be made by make menuconfig in the unix shell. Access the shell and set the LED to be switched on/off  the value 2 as GPIO; this is because in our case the card is a doit whose design is illustrated in this document


Access to Example configuration

blink-configurationand set the gpio value to 2

blink-gpioSave the configuration. Go back to Eclipse and proceed with cleaning with

eclipse-clean-blinkand subsequent compilation with Project -> Build

eclipse-build-blinkTo run firmware flash from eclipse;, select the project and with the right button choose Build Targets -> Create …

eclipse-create-targetEnter flash as value in the target name field, leaving everything else as default

eclipse-create-target-flashAt this point we proceed to upload the image by double-clicking on the new flash item within the project targets

eclipse-run-flashThe settings to run the flash must be set with make menuconfig, as seen in the previous article

 

ESP32 – IDF Sdk

This article describes how to set the Espressif development environment for the ESP32.

From the link to the resources for the soc

https://www.espressif.com/en/products/hardware/esp32/resources

we access the guide for compiling the firmware on the Windows, OSX and Linux platforms

http://esp-idf.readthedocs.io/en/latest/get-started/index.html

In this article we use Linux Debian 9 as operating system, configured as described previously  here

MicroPython – Python ib embedded devices

We  use the IDF github. On Debian 9, with the user used to connect to the system, we create  the Esp32 folder under the user’s Home and  the espressif folder inside it

Install the prerequisites

Download the cross compiler

Create the esp folder and unpack the toolchain

Add the bin folder of the cross compiler to the PATH

Add the user used for the operations to the dialout group

Clone the IDF repository from  $HOME/Esp32/espressif

Add the IDF_PATH environment variable

Create the project folder under $HOME/Esp32/espressif and copy an example from di IDF examples

Run the configuration

Access Serial flasher config -> and set the connection parameters; in our case we leave the default

idf-serial_flasher-engSave the configuration and build the firmware

To upload the firmware on the device run

To upload only the application run

while to build only the application run

To run the program monitor run

hello_world-engHere are the other examples available with sdk

https://github.com/espressif/esp-idf/tree/f586f5e/examples

The following link contains the references of the IDF API

http://esp-idf.readthedocs.io/en/latest/api-reference/index.html

You can also generate documentation from the github source. Go into  ~/Esp32/espressif/esp-idf/docs/en

Install the prerequisites

Generate documentantion in html format

make-html-eng

The documentation is available under  _build / html.

In the next article we will configure Eclipse as IDE  to build our applications

ESP32 -IDF Sdk with Eclipse

 

ESP32 – Nodemcu firmware compiling

After  the compilation of the MicroPython firmware for ESP32

ESP32 – MicroPython compiling for ESP32

this article shows how to generate the Nodemcu firmware for the ESP32 MCU.

The firmware is not yet complete as the one existing for ESP8266.

The instructions can be found on the github at the link

https://github.com/nodemcu/nodemcu-firmware/tree/dev-esp32

The compilation documentation is available at the link

https://nodemcu.readthedocs.io/en/dev-esp32/en/build/

The operating system used is Debian 9 previously encountered for the esptool and Adafruit-ampy utilities

MicroPython – Python on embedded devices

Log on  with the operating system user on Debian 9, sviluppo in our case; under the home folder  create the Esp32 folder and inside this folder the  nodemcu  folder

Clone the repository

To update the cloned repository follow the description on the compilation link, i.e.

Go into  nodemcu-firmware-esp32 folder and run

nodemcu-settingsEnable the bluetooth  in Component config–>Bluetooth

nodemcu-bluetoothEnable various modules in Component config–>Nodemcu modules

nodemcu-modulesIn the flash settings we set the default values for our ESP32,  Serial flasher config->

nodemcu-flashSave the configuration and exit. To build the firmware run

At the end of the compilation the command to upload the firmware  is shown; the same operation is executed  by simply running

For informations about the partitions of the various ESP32 versions, refer to the documentation

http://api-guides/partition-tables.html

In the next article we’ll  show how to make an application using the Espressif SDK

ESP32 – IDF Sdk

ESP32 – MicroPython compiling for ESP32

We begin now a series of articles about the Espressif ESP32 MCU

ESP32 Wifi-Bluetooth Module

After having analyzed in the previous articles  MicroPython  for ESP8266, in this we start to treat MicroPython on ESP32.

The following shows how to generate the MicroPython image from the source code for the ESP32 board.

The operating system is Debian 9, previously encountered for the esptool and Adafruit-ampy utilities

MicroPython – Python on embedded devices

Log on  with the operating system user on Debian 9, sviluppo in our case; under the home folder  create the Micropython folder and inside this folder the  esp32 folder

The MicroPython source code is available at the following link

Github Micropython

Instructions to build the firmware for the ESP32 are available at the link

MicroPython esp32

Clone the repository

Go into  micropython/ports/esp32 folder

Run

The command releases the git hash for the supported espidf

At the time of our compilation the result is as follows

Go into  /home/sviluppo/Micropython/esp32 folder and clone ESPIDF

Checkout with the hash obtained above

Run

We follow the instructions at the link

Cross compiler

to install the cross compiler; install the prerequisites for building

Download the cross compiler

Create the  esp folder under /home/sviluppo/Micropython/esp32

Unpack the cross compiler in esp

Add the binary folder of the cross compiler to the PATH envinronment variable

Go into the following folder

and create the makefile file with the following code inside

Go into the micropython folder available in /home/sviluppo/Micropython/esp32

Run

to add external dependencies.

Next step is the MicroPython cross compiler build

Finally we can build the ESP32  firmware with the following commands

The firmware is generated under ports/esp32/build with the name firmware.bin.

Proceed to load the firmware with the commands

In our case, the firmware upload was successful with the command

It is also possible to load the firmware.bin image with the commands

where  /dev/ttyXXX is replaced with the actual serial port to which we connect the ESP32 board.

In the next article we’ll compile the nodemcu firmware for ESP32

ESP32 – Nodemcu firmware compiling

MicroPython – MicroPython compiling for ESP8266 with additional modules

After the firmware build of the MicroPython firmware for esp8266 in the previous article

MicroPython – MicroPython compiling for ESP8266

we now show how build a module inside the firmware; this option optimizes the module execution and minimizes the size of the same on the device; the module is frozen into the firmware.

With the firmware compiled with the defaults, the following modules are available, listed in the REPL console with the command

standardmodulesNow let’s try to compile the firmware with the mqtt modules already inside. For this purpose, the module sources are copied to the umqtt folder under modules

umqttfolderWe proceed to recompile the firmware. From micropython/ports/esp8266 we run the following commands

We can see from the compilation log, that these modules are placed in the frozen state

frozenmodulesAfter uploading this firmware we obtain the following list of modules inside the firmware

firmwarewithmodulesIt is also possible to create the firmware with the modules not compiled and optimized, by copying the sources in the scripts folder under esp8266; in the specific case the sources must be placed as shown below

The modules are always visible in the firmware, but not optimized, so if you need some module it is convenient to copy it on the device during the coding  and in the last phase of the project compile the firmware with the frozen modules.