The expansion of the Internet of Things (IoT) over the past decade has significantly changed how we interact with everyday appliances and industrial systems, from smart homes to the Industrial Internet of Things (IIoT). However, as cyberattacks become more frequent, sophisticated, and dynamic, IIoT systems are presenting new security challenges. There is a need for the timely availability of new efficient approaches and accessible/feasible testbeds to support new approach development and validation in real-world scenarios. In this paper, we present a novel security-oriented IIOT testbed that comprises a wide range of physical processes through diverse controllers and devices interacting with each other through various networking protocols. Additionally, we implement and test the performance of machine learning (ML) algorithms, including Random Forest, Decision Tree, Naive Bayes, K-nearest Neighbors, Logistic Regression, Multi-Layer Perceptron Classifier, Recurrent Neural Network, and Transformer for cyber-attack detection. Finally, we collect a new dataset and compare it with another IIoT-based dataset, CICIoT2023, revealing machine learning models’ effectiveness in detecting various cyber-attacks and the feasibility and future work of the proposed testbed.