A simple cache without a database support. The server maintains a key–value associative array of items (json objects). MEMORY_LIMIT setting allows to limit amount of memory that can be used by the server. When the limit is reached, subsequent inserts cause older data to be purged in least recently added order. The logic can be easily improved to purge items in least recently used order.
yarn
yarn run dev
Access http://localhost:8080
PORT and MEMORY_LIMIT in .env file where memory limit corresponds to process.memoryUsage().heapUsed nodejs parameter. After heapUsed reaches MEMORY_LIMIT subsequent inserts will cause older data to be purged in least recently added order. If for some reason memory limit is reached, but cache has no items to remove (the cache is already empty), server will return 507 error.
Healthcheck, current memory usage and memory limit GET http://localhost:8080/
Add item POST http://localhost:8080/v1/items
Get item GET http://localhost:8080/v1/items/{itemId}
Delete item DELETE http://localhost:8080/v1/items/{itemId}
(Testing) See all the items GET http://localhost:8080/v1/items
Project automatically builds Docker image, pushes it to Docker Hub and deploys image container to Digital Ocean Kubernetes cluster on push to master via GitHub actions.
Access Cluster http://64.227.108.161:31997
See deployment history https://github.com/steel8rat/cache/actions
Docker Hub repo https://hub.docker.com/repository/docker/steel8rat/cache
Github actions workflow config .github/workflows/do-k8s-deploy.yml
Digital Ocean Kubernetes config do-k8s/api-server.yml
yarn run test