Articles
Managing disks in Linux
Hard disk drives (HDDs) and Solid state drives (SSDs) are currently the most widespread high capacity storage devices. When such a drive is connected to a computer running Linux, it can be found in the device directory <em>/dev/</em> under the name <em>sdX</em>, where <em>X</em> is a letter changing for each drive.
Kubernetes persistent volumes
Applications deployed on a Kubernetes cluster run inside containers. As a consequence, their file system is that of the container, which means that if the container is removed, the data it contained is lost.
Gitlab CI dealing with credentials
GitLab can automatically dockerize applications using the appropriate CI configuration. However, for obvious security reasons, it is bad practice to include credentials in a git repository. Consequently, the CI pipeline is by default not in a position to include credentials in the dockerized application, which most likely prevents the latter from running as intended.
Kubectl pull new version of image without changes to manifest
When using kubectl apply using an already applied and unchanged manifest file, nothing happens on the Kubernetes cluster. However, deployments can be configured so as to always pull a new version image upon restart. This is achieved using the, <code>imagePullPolicy: Always</code> parameter:
Mongoose bulk update upsert
MongoDB's upsert option enables the creation of a new document if the query of an update operationdoesn't match any existing document. Basically, an update command with upsert creates a document if it does notexist already, and update it otherwise.
Docker restart container when docker restarts
Simply add the following flag when using docker run
Gitlab CI commands for TF serving
This is an example .gitlab-ci.yml file which can be used to containerize and deploy a tensorflow model
Kubectl create deplpoyment and service at same time
Simply add entries for both the deployment and the service in the same manifest, separeted by ---
Serving a Keras model using Tensorflow serving and Docker
A Keras model can be created in various ways, for example using the <a href="https://keras.io/getting-started/sequential-model-guide/" rel="noopener noreferrer nofollow">sequential model</a>:
Arduino button debounce function
Physical buttons and switches do not switch between open and close position instantly and perfectly. Instead, they have a tendency to "bounce" between the two states before settling on one. This issue is thoroughly described in circuitdigest's excellent article. Consequently, when using such components with, for example, an Arduino microcontroller, a proper debouncing algorithm must be implemented.