Use the web page to achieve access to the laboratory log of all platforms, and I usually want to make a log similar to a diary. It is now possible to store text and files that want to save. Now you can deploy a file document storage service in a cluster.

Install server

The system is mainly divided into two parts, reverse proxy service and main service, in these two folders /cmd/proxy and /cmd/server. The main service can run independently without setting up a proxy!

1. git clone get all of these source code.

2. Use commands go build main.go to compile executable file.

Configuration

When deploying the service, almost all settings are in the configuration file, which can be quickly changed and customized.

Config server

server:
  # If you use a proxy for multi-machine deployment, you need to change the
  # address settings and port settings, and set the proxy server address and
  # select an appropriate timeout duration.
  handler:
    # The network address the server is running on.
    host_address: 127.0.0.1
    listen_port: 8090
    # Cookie's valid duration the unit is seconds.
    cookie_duration: 7200
    # Cookie's determines that it can work in which domain.
    access_scope: 127.0.0.1
    # File's storage directory store files uploaded by users.
    storage_directory: ./storage
    # The default prefix when generating download links.
    url_base: http://127.0.0.1:8090/library/download

  token:
    # Token's valid duration the unit is seconds.
    token_duration: 7200
    # Key used for token encryption.
    encryption_key: 20180212
    # The issuer of the token.
    token_issuer: labnote

  proxy:
    proxy_address: 127.0.0.1:11000
    # Choose whether use the proxy server.
    use_proxy: true
    # Timeout duration of the registry.
    timeout: 120

database:
  # Here you can choose between 'mysql' or 'mongo' options.
  type: mongo
  # Database connection command and selected database name.
  command: mongodb://admin:[email protected]:27017
  name: labnote

cache:
  address: 127.0.0.1:6379
  password: ''
  # This controls the size of the connection pool started by redis.
  pool_size: 100

Proxy Config

proxy:
  # When using a proxy, static files will be loaded by the proxy server, and be
  # sure to select a suitable timeout period to ensure the stability of the system.
  universal:
    # The network address the proxy is running on.
    host_address: 127.0.0.1
    listen_port: 11000
    # The http version of this proxy such as "http" and "https".
    http_version: 'http'
  registry:
    # Timeout duration of the registry.
    timeout: 120

database:
  # Here you can choose between 'mysql' or 'mongo' options.
  type: mongo
  # Database connection command and selected database name.
  command: mongodb://admin:[email protected]:27017
  name: labnote

cache:
  # Cache the authentication information of the database.
  address: 127.0.0.1:6379
  password: ''
  # This controls the size of the connection pool started by redis.
  pool_size: 100

Transform Config

// Resumable uploads are also provided, you can change the file blocks size
// according to your network environment.
const chunkSize = 2 * 1024 * 1024;

Running

If you need to use a proxy server, you need to start it before starting the main service. And registration is not available because it only used in our lab or small-scale use. So please enter the user directly in the database in the following format.

{
  "email": "[email protected]",
  "password": "..........."
}

During the server running, you can use the note and file system to write or storage something. There are some UI problem at the file system page, please do not try to break something.

Customized

This project is open source, welcome to customize the code and feedback questions! I try to make the system modular as much as possible to simplify the addition of later modules.

Database

In addition to providing some parameters, data interfaces are provided for use in different databases.

// Database definition of the abstract interface of it.
type Database interface {
	// The init function of this database.
	InitConnection() error
	// The close function of this database.
	CloseConnection() error
	// Verify that the user information is reasonable.
	CheckUserAuth(user *User) (bool, error)
	// Insert one note to this labnote system.
	InsertOneNote(note *Note) error
	// Insert one file to this labnote system.
	InsertOneFile(file *File) error
	// Request all notes from the database.
	GetAllNotes() (*[]Note, error)
	// Request all files from the database.
	GetAllFiles() (*[]File, error)
}

If you want to change the database which the server used, you can change config.yml.

Cache

In order to quickly exchange data and facilitate the storage of file slice information, we use redis as a cache database.

// Cache interface defines the functions of the cache.
type Cache interface {
	// Init function of this cache.
	InitConnection() error
	// Close function of this cache.
	CloseConnection() error
	// Insert a chunk info in a list which key is file's hash.
	InsertOneChunk(hash string, chunk Chunk) error
	// Get the chunks list in the cache.
	GetChunkList(hash string, name string) (*[]Chunk, error)
	// Remove all of chunks after merge or error.
	RemoveAllRecords(hash string) error
	// Init the file's state for the server check.
	ChangeFileState(hash string, saved bool) error
	// Check this file whether exist completely or not.
	CheckFileUpload(hash string) (bool, error)
	// Initialize the location of this file for reverse.
	InitFileLocation(hash string, addr string) error
	// Get the storage location of the file and get where it is.
	GetFileLocation(hash string) (string, error)
}

Like the interface provided by the database, you also can easily change the detail of functions and add functions you want.

Problem

I'm not very good at web UI development, and there may be some problems here. I'm appreciate someone who asks issues and improves the code. And I have an idea to write a distributed file system based on p2p network, but don't know how to do it.