mhu/shimatta-kenkyusho
2
0
Fork 0
Code Issues 13 Pull Requests 2 Releases Wiki Activity

WIP: issue/23-add-docu #27

Draft
mhu wants to merge 4 commits from issue/23-add-docu into develop
pull from: issue/23-add-docu
merge into: mhu:develop
Conversation 0 Commits 4 Files Changed 3 +190
3 changed files with 190 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

185
README.md Normal file
View File

@ -0,0 +1,185 @@
# Shimatta Kenkyusho Parts Database
## Installation
### Prerequisites
Shimatta Kenkyusho (しまった・研究所) is a Django based web application. It is highly recommended to run it using the supplied docker setup. This removes the need of any special installation on the host system. This guide assumes, that `nginx` is running on the host system and can serve as a reverse proxy and webserver. For easiest download, it is recommended to clone the desired release with `git`
Install the requirements:
**For Debian / Ubuntu:**
```
# apt-get update
# apt-get install docker docker-compose-plugin nginx git
```
**For Arch based Systems:**
```
# pacman -S nginx docker docker-compose git
```
### Setup Shimatta Kenkyusho
Clone this repository:
```
$ git clone https://git.shimatta.de/mhu/shimatta-kenkyusho.git
```
> Note: Shimatta Kenkyusho is currently not stable yet and the newest verison is in the `develop` branch. This will change once actual releases are done and merged to the `master` branch. You will be able to get the latest stable version from the `master` branch or a repsective tag. For now, the `develop` is recommended.
Change directory into the `shimatta-kenkyusho` folder cloned by git.
Copy the `example.env` file to `.env` and edit it according to your needs:
The following settings are required to be adapted:
- `DJANGO_STATIC_VOL`: The directory the application will extract its static data into, which needs to be served by your webserver. See the example reverse proxy setup for more details.
- `DJANGO_MEDIA_VOL`: The directory all media files like images uploaded to the application are stored here. This folder must be served by your webserver on the configured media URL.
- `PGDATA_VOL`: The directory, the postgres database will store its files.
- `PORT`: The TCP/IP port that the whole setup will listen on. Use a reverse proxy to forward to this port. Do not directly expose it to the internet!
- `DJANGO_SECRET_KEY`: Provide a secret, and randomly generated key. Do not share this with anybody!
- `DJANGO_ALLOWED_HOST`: Set this to the domain, the application will be reached at. E.g: `lab.example.com`
- `DJANGO_MEDIA_URL`: Set this to the media URL at which your webserver serves the `DJANGO_MEDIA_VOL` diretory. E.g: `media.lab.example.com/` Note the **slash at the end**. It is important.
Once the environment is set up, the docker containers can be built and started. Run
```
$ docker compose build
```
This will generate two container images:
1. `shimatta-kenkyusho-shimatta-kenkyusho-web`: The django application
2. `postgres`: A alpine based docker container containing the postgres database.
Start the application as a service with
```
$ docker compose up -d
```
> Note: The initial startup might need a minute because the whole database etc needs to be initialized first.
Use
```
$ docker ps
```
to check if the `shimatta-kenkyusho-shimatta-kenkyusho-db` and the `shimatta-kenkyusho-shimatta-kenkyusho-web` container are running and report a *healthy status*.
### Setup Initial Login User
> TODO
### Example Reverse Proxy Setup Using nginx
Once the setup is configured the reverse proxy setup is needed. This setup serves three purposes:
1. Redirect incoming requests to the django application running on the port `PORT` configured in the `.env`
2. Serve static files ath the URL: (e.g. `lab.example.com/static`). See `ALLOWED_HOST` configuration.
3. Serve the media volume at the media URL (e.g. `media.lab.example.com`). See `DJANGO_MEDIA_URL`
Example nginx configuration for `nginx >v2.25` with SSL and http2 / http3 support:
```
# Force redirection from http to https for application
server {
listen 80;
listen [::]:80;
server_name lab.example.com; # This must match your ALLOWED_HOST. Adapt domain.
allow all;
return 301 https://lab.example.com$request_uri; # Adapt domain
}
# Force redirection from http to https for media url
server {
listen 80;
listen [::]:80;
server_name media.lab.example.com; # Adapt domain name according to DJANGO_MEDIA_URL
allow all;
return 301 https://media.lab.example.com$request_uri; # Adapt domain name
}
# Reverse Proxy for application
server {
listen 443 ssl;
listen [::]:443 ssl;
http2 on;
# Add this for HTTP3. If your nginx is older than 2.25 this might not be available
######################################################################################
# listen 443 quic reuseport;
# listen [::]:443 quic reuseport;
# Enable QUIC and HTTP/3
# ssl_early_data on;
# add_header Alt-Svc 'h3=":443"; ma=86400';
#######################################################################################
server_name lab.example.com; # Adapt domain
# Use letsencrypt as SSL certificate provider.
ssl_certificate /etc/letsencrypt/live/lab.example.com/fullchain.pem;
ssl_certificate_key /etc/letsencrypt/live/lab.example.com/privkey.pem;
ssl_protocols TLSv1.3;
ssl_prefer_server_ciphers on;
ssl_session_cache shared:SSL:1m;
ssl_session_timeout 5m;
ssl_ciphers HIGH:!aNULL:!MD5;
location / {
allow all;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-Proto $scheme;
proxy_pass http://127.0.0.1:8000; # Adapt PORT from .env
}
location /static/ {
# Adapt path to static volume here. Note the slash at the end
alias /path/to/DJANGO_STATIC_VOL/;
allow all;
}
client_max_body_size 60m;
}
# Serve the media files
server {
listen 443 ssl;
listen [::]:443 ssl;
# Add this for HTTP3. If your nginx is older than 2.25 this might not be available
######################################################################################
# listen 443 quic reuseport;
# listen [::]:443 quic reuseport;
# Enable QUIC and HTTP/3
# ssl_early_data on;
# add_header Alt-Svc 'h3=":443"; ma=86400';
#######################################################################################
http2 on;
server_name media.lab.example.com; # Adapt according to DJANGO_MEDIA_URL
# Use letsencrypt as SSL certificate provider.
ssl_certificate /etc/letsencrypt/live/media.lab.example.com/fullchain.pem;
ssl_certificate_key /etc/letsencrypt/live/media.lab.example.com/privkey.pem;
ssl_protocols TLSv1.3;
ssl_prefer_server_ciphers on;
ssl_session_cache shared:SSL:1m;
ssl_session_timeout 5m;
ssl_ciphers HIGH:!aNULL:!MD5;
error_page 502 /lab_down.html;
allow all;
root /path/to /DJANGO_MEDIA_VOL/; # Adapt this to the volume provided.
}
```
Congratulations. Your shimatta kenkyusho installation is now fully setup.
> Note that, the `compose.yaml` contains a restart-policy. By default the contianers will restart automatically, even after a reboot of the host machine, if the docker service is enabled.
## Backup and Restore
> TODO
## Debugging and Development
> Todo

5
compose.yaml
View File

@ -1,5 +1,9 @@
x-op-restart-policy: &restart_policy
restart: unless-stopped
services:
shimatta-kenkyusho-web:
<<: *restart_policy
build: .
volumes:
- "${DJANGO_STATIC_VOL:-./run/static}:/var/static"
@ -30,6 +34,7 @@ services:
start_period: 30s
shimatta-kenkyusho-db:
<<: *restart_policy
image: postgres:16.5-alpine
environment:
POSTGRES_PASSWORD: "${DJANGO_POSTGRESQL_PW:-p4ssw0rd}"

0
.env.example → example.env
View File