A collection of nodes for making Node-RED more resilient by adding self-healing capabilities. This project is at an early development stage and its usage in production environments is not recommended.

This work is part of an ongoing PhD thesis in Software Engineering and Internet-of-Things at the Faculty of Engineering, University of Porto (FEUP). Work supervised by Prof. Hugo Sereno Ferreia and Prof. João Pascoal Faria. With collaboration of Prof. André Restivo.

Each node has a README.md in its folder with further information about it.

Checks if an action was completed by using sensor acknowledgements.

Balances the distribution of messages through multiple outputs using three different strategies: Round Robin, Weighted Round Robin and Random.

Acts between a node sending a message to another, storing the last one in local context and resending it after restarts, if it's within a specified time to live.

Compensate missing values (detected by disruptions on the periodicity of incoming mesages) with a pre-defined strategy (e.g. average of the last 10 readings, last value or maximum value of the last 10 readings).

Delay a command (message) in order to meet the actuator response capacity (e.g. avoid overload). Similar to rate-limit-messages, but with different strategies.

Enable and disable Node-RED flows during runtime (local or remote instances, using the available REST API).

Provides a heartbeat probe for MQTT and HTTP.

A node to continuously scan the network to find working IPs at ports 8080, 443 and 80.

Kalman noise filter.

Continuosly scan the network to find new or removed devices. Can be combined with a device-registry.

Manage redundant instances of Node-RED (setting a master instance). Works only on the local network (uses n2n communication).

Drop values if they are in or out of a given threshold (e.g. two close temperature readings).

Picks a value (e.g. sensor reading) from an array values based on a pre-defined majority.

Monitors system resources, ranging from battery levels to resources usage.

Checks for reading (value) sanity (e.g. checks if the reading is between the sensor possible output values).

Checks for timing issues on data inputs. There are 3 outputs that refer to data comming on expected time, too slow or too fast. A frequency in seconds along with a margin (float: 0-1) should be provided.

All the devices that are reachable can communicate with this device in order to store their information and current state.

Stores the internal state of all flows, making it available to different Node-RED instances.

• Installing Node-RED (Official Docs): https://nodered.org/docs/getting-started/

• Clone or download this repository.
• In your node-red user directory, typically ~/.node-red (in Windows something like C:\Users\<my_name>\.node_red), run: npm install <path_to_downloaded_folder>/node-red-contrib-self-healing
• Start (or restart) Node-RED.
• Nodes should be available under the SHEN tab of the node palette.

• First run: npm install

• Unit tests for every node: npm run test

• Unit tests with coverage: npm run test-coverage

• Mutation tests: npm run mutate

• Property based tests: npm run test-pbt

• Acceptance tests: npm run test-acceptance

• Installing Costum Nodes -- Official Documentation

If you find this code useful in your research, please consider citing:

Visual Self-healing Modelling for Reliable Internet-of-Things Systems (ICCS 2020)

@inproceedings{DiasICCS2020,
    author="Dias, Joao Pedro and Lima, Bruno and Faria, Joao Pascoal and Restivo, Andre and Ferreira, Hugo Sereno",
    editor="Krzhizhanovskaya, Valeria V. and Zavodszky, Gabor and Lees, Michael H. and Dongarra, Jack J. and Sloot, Peter M. A. and Brissos, Sergio and Teixeira, Joao",
    title="Visual Self-healing Modelling for Reliable Internet-of-Things Systems",
    booktitle="Computational Science -- ICCS 2020",
    year="2020",
    publisher="Springer International Publishing",
    address="Cham",
    pages="357--370",
    isbn="978-3-030-50426-7"
}

A Pattern-Language for Self-Healing Internet-of-Things Systems (EuroPLoP'20)

@inproceedings{DiasEuroplop2020,
    title        = {A Pattern-Language for Self-Healing Internet-of-Things Systems},
    author       = {Dias, Jo\~{a}o Pedro and Sousa, Tiago Boldt and Restivo, André and Ferreira, Hugo Sereno},
    year         = 2020,
    booktitle    = {Proceedings of the 25th European Conference on Pattern Languages of Programs},
    location     = {Irsee, Germany},
    publisher    = {Association for Computing Machinery},
    address      = {New York, NY, USA},
    series       = {EuroPLop ’20},
    doi          = {10.1145/3361149.3361165},
    numpages     = 8
}