kopia lustrzana https://github.com/Stopka/fedisearch-compose
30 wiersze
1.3 KiB
Markdown
30 wiersze
1.3 KiB
Markdown
# FediSearch compose
|
|
|
|
This an example setup for Fediverse feed crawler and searcher.
|
|
|
|
Fedisearch consists of 3 app containers:
|
|
1. **Postgres** - SQL database for storing all found feeds
|
|
2. **Fedicrawl** - crawls through the Fediverse network and stores all found nodes and also its feeds to database
|
|
3. **Fedisearch** - web frontend for searching data in database
|
|
|
|
## Installation
|
|
App is designed primarily to be run in docker on linux.
|
|
Thus `docker` and `docker-compose` are mandatory dependencies for this example setup.
|
|
|
|
I don't have prebuilt images in docker hub ready yet (but it's planned), so you have to build them locally.
|
|
Docker compose should build images for you from sources...
|
|
|
|
1. Clone [fedisearch](https://github.com/Stopka/fedisearch) and [fedicrawl](https://github.com/Stopka/fedicrawl) apps to `./build` directory.
|
|
2. Copy `.env.dist` to `.env` and edit it as needed.
|
|
3. Create an `docker-compose.override.yml` file and override configuration by your liking.
|
|
You should at least add proxy (simple example is in `docker-compose.traefik.yml`) with tls termination.
|
|
4. Start the up by command
|
|
```docker compose up -d```
|
|
|
|
## Config
|
|
All configuration is done by env vars.
|
|
Just edit `.env` file.
|
|
|
|
Meaning of variables can be found in the readme files of
|
|
[fedisearch](https://github.com/Stopka/fedisearch) and [fedicrawl](https://github.com/Stopka/fedicrawl) apps.
|