Publicación: Reconfiguración dinámica de redes de laboratorios. Autenticación y monitorización con Kubernetes
Cargando...
Fecha
2020-10
Autores
Editor/a
Director/a
Tutor/a
Coordinador/a
Prologuista
Revisor/a
Ilustrador/a
Derechos de acceso
info:eu-repo/semantics/openAccess
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad Nacional de Educación a Distancia (España). Escuela Técnica Superior de Ingeniería Informática. Departamento de Sistemas de Comunicación y Control
Resumen
La irrupción de las infraestructuras y plataformas como servicio han supuesto una revolución en muchos ámbitos, y abre un mundo de nuevas posibilidades que antes simplemente no eran posible. Si además se combinan con miniordenadores de reducido tamaño y peso, pero buenas prestaciones, nos da como resultado la posibilidad de poder crear clústeres de ordenadores de un tamaño inferior a la torre de un ordenador de sobremesa, pero con un gran potencial. Y, si se plantean este tipo de clústeres para actividades o aplicaciones de laboratorio, nos da como resultado una herramienta muy potente para la práctica y formación de alumnos. Pensando en esto, el Proyecto eNMoLabs está desarrollando y gestionando una serie de trabajos que permitan poner en marcha un Lab of Things basado en servicios, un laboratorio con diferentes funcionalidades integradas y que pueda ser accesible de forma remota para trabajar en él. Es dentro de este Proyecto donde se enmarca este trabajo, en el que crearemos un clúster de kubernetes sobre una serie de dispositivos de bajo coste llamados Raspberry Pi. Sobre este clúster configuraremos una aplicación que permita la autenticación y autorización de usuarios por medio de las credenciales de la universidad, y con una gestión de permisos que aseguren que los usuarios autenticados estén asignados a los grupos que correspondan, que a su vez estos grupos se corresponden con roles para controlar los permisos de lo que podrán hacer en el clúster. Esta capa de seguridad se hará siguiendo uno de los estándares actuales de seguridad como es OIDC. Posteriormente nos centraremos en la preparación y configuración de una monitorización general del clúster tanto a nivel de recopilación y presentación de métricas de uso y estado como a nivel de alertas en caso de que se produzcan situaciones anómalas. Y esto se hace con herramientas actuales muy populares como Prometheus o Grafana. Esto permitirá conocer el estado general del clúster y poder configurar notificaciones cuando correspondan, haciendo que la operación del clúster sea más sencilla. De esta forma acabaremos con un clúster que podemos considerar base para que a partir de él se pueda empezar a usar para los diferentes usos que el laboratorio permita sabiendo que los requisitos de seguridad y monitorización están ya cubiertos.
The irruption of the infrastructures and the platforms as a service has marked a milestone in many fields and open a world of new possibilities that till then were just simply not possible. If additionally they are combined with microcomputers with reduced size and weight, although good outputs, we'll get as result the possibility of being able to create computer clusters with a size smaller than a conventional computer tower but with great potential. And, if this kind of cluster is meant to be used as laboratories, we'd get a very powerful tool for the students' tasks and education. Thinking about this, the eNMoLabs Project is developing and managing a set of works with the aim of initiating a Lab of Things based on services, a laboratory with several different integrated functionalities that can be accessed remotely to work on it. Inside this Project is where the current work is defined. We'll create a Kubernetes cluster over a set of low-cost devices called Raspberry Pi. Over this cluster, we'll set an application that allows the user authentication and authorization with the university credentials, and with a permissions management that ensures that the authenticated users are assigned to the groups where they belong to, which at the same time corresponds to roles that are set to control the permissions of the things that can be done inside the cluster. This security layer will be done following one of the current security standards, as OIDC is. Later we'll focus on the preparation and configuration of general monitoring for the cluster, both to gather and present usage and status metrics, and to prepare alerts that will raise in unexpected events. And this is done using tools that are very common currently and popular like Prometheus and Grafana. This will allow us to know the general status of the cluster and be able to set notifications when corresponding, doing the cluster operation simpler. In this way, we'll end up with a cluster that we can consider as a base that can be used for the different uses that the laboratory permit, knowing that the requirements related to security and monitoring are already covered.
The irruption of the infrastructures and the platforms as a service has marked a milestone in many fields and open a world of new possibilities that till then were just simply not possible. If additionally they are combined with microcomputers with reduced size and weight, although good outputs, we'll get as result the possibility of being able to create computer clusters with a size smaller than a conventional computer tower but with great potential. And, if this kind of cluster is meant to be used as laboratories, we'd get a very powerful tool for the students' tasks and education. Thinking about this, the eNMoLabs Project is developing and managing a set of works with the aim of initiating a Lab of Things based on services, a laboratory with several different integrated functionalities that can be accessed remotely to work on it. Inside this Project is where the current work is defined. We'll create a Kubernetes cluster over a set of low-cost devices called Raspberry Pi. Over this cluster, we'll set an application that allows the user authentication and authorization with the university credentials, and with a permissions management that ensures that the authenticated users are assigned to the groups where they belong to, which at the same time corresponds to roles that are set to control the permissions of the things that can be done inside the cluster. This security layer will be done following one of the current security standards, as OIDC is. Later we'll focus on the preparation and configuration of general monitoring for the cluster, both to gather and present usage and status metrics, and to prepare alerts that will raise in unexpected events. And this is done using tools that are very common currently and popular like Prometheus and Grafana. This will allow us to know the general status of the cluster and be able to set notifications when corresponding, doing the cluster operation simpler. In this way, we'll end up with a cluster that we can consider as a base that can be used for the different uses that the laboratory permit, knowing that the requirements related to security and monitoring are already covered.
Descripción
Categorías UNESCO
Palabras clave
kubernetes, Raspberry Pi, OIDC, autenticación y autorización, monitorización, Prometheus, grafana, alertas, authentication and authorization, monitoring, alerts
Citación
Centro
Facultades y escuelas::E.T.S. de Ingeniería Informática
Departamento
No procede