[](https://hub.docker.com/r/ggogel/seafile-server)
[](https://hub.docker.com/r/ggogel/seahub)
[](https://hub.docker.com/r/ggogel/seahub-media)
[](https://hub.docker.com/r/ggogel/seafile-caddy)
- Complete redesign of the [official Docker deployment](https://manual.seafile.com/docker/deploy%20seafile%20with%20docker/) with containerization best-practices in mind.
- The caddy reverse proxy serves as a single entry point to the stack. Everything else runs in an isolated network.
- Using [Alpine Linux](https://alpinelinux.org/about/) based images for the frontend, which is designed with security in mind and comes with proactive security features.
- Reworked Dockerfiles featuring multi-stage builds, allowing for smaller images and faster builds.
- contains the backend component called [seafile-server](https://github.com/haiwen/seafile-server)
- handles storage, some direct client access and seafdav
- *seahub*
- dynamic frontend component called [seahub](https://github.com/haiwen/seahub)
- serves the web-ui
- communicates with seafile-server
- *seahub-media*
- serves static website content as well as avatars and custom logos
- *db*
- the database used by *seafile-server* and *seahub*
- *memcached*
- database cache for *seahub*
- *seafile-caddy*
- reverse proxy that forwards paths to the correct endpoints: *seafile-server*, *seahub* or *seahub-media*
- is the single external entrypoint to the deployment
Volumes:
- *seafile-data*
- shared data volume of *seafile-server* and *seahub*
- also contains configuration and log files
- *seafile-mariadb*
- volume of the *db* service
- stores the database
- *seahub-custom*
- contains custom logos
- stored by *seahub* and served by *seahub-media*
- *seahub-avatars*
- contains user avatars
- stored by *seahub* and served by *seahub-media*
*Note: In the official docker deployment custom and avatars are served by nginx. Seahub alone is not able to serve them for some reason, hence the separate volumes.*
Networks:
- *seafile-net*
- isolated local network used by the services to communicate with each other
If you run a single node swarm and don't want to run multiple replicas, you can use the same compose file. Otherwise refer to [Additional Information / Docker Swarm](#Docker-Swarm-1).
**Important:** The environment variables are only relevant for the first deployment. Existing configuration in the volumes is **not** overwritten.
On a first deployment you need to carefully set those values. Changing them later can be tricky. Refer to the Seafile documentation on how to change configuration values.
This will be used for the SERVICE_URL and FILE_SERVER_ROOT.
Important: Changing those values in the config files later won't have any effect because they are written to the database. Those values have priority over the config files. To change them enter the "System Admin" section on the web-ui. If you encounter issues with file upload, it's likely that those are configured incorrectly.
**If you set up Seafile from scratch you can skip this part.**
The [official Docker deployment](https://manual.seafile.com/docker/deploy%20seafile%20with%20docker/) uses [bind mounts](https://docs.docker.com/storage/bind-mounts/) to the host path instead of actual docker volumes. This was probably chosen to create compatibility between a native install and the docker deployment. This deployment uses [named volumes](https://docs.docker.com/storage/volumes/), which come with several advantages over bind mounts and are the recommended mechanism for persisted storage on Docker. The default path for named volumes on Docker is `/var/lib/docker/volumes/VOLUME_NAME/_data`.
This deployment does by design **not** include a reverse proxy that is capable of https and Let's Encrypt, because usually Docker users already have some docker-based reverse proxy solution deployed, which does exactly that. If you're using Docker for a while already, you probably know what to do and you can skip this section.
The *seafile-server* images contains scripts that will detect if a newer version of *seafile-server* is deployed and will automatically run the migration scripts included in the Seafile package. Upgrade from 7.1 is succesfully tested.
If you are unable to upload user avatars and custom logos, check your `SERVICE_URL` in the admin settings. It should not contain any port number. It should begin with `https` if you use a reverse proxy with https.
In order for LDAP to work, the *seafile-server* needs to be able to establish a connection to the LDAP server. Because the `seafile-net` is defined as `internal: true`, the service won't reach the LDAP server, as long as you don't deploy it in to the same stack and also connect it to `seafile-net`. As a workaround define another network in the `networks` top-level element. Like this:
```
networks:
seafile-net:
internal: true
ext:
```
If not defined otherwise, the network will automatically have external access.
Then hook up *seafile-server* to this network:
```
seafile-server:
...
networks:
- seafile-net
- ext
```
### OAuth
For OAuth the same network problem as with LDAP will occur, but here you will need to hook up the *seahub* service to the external network.
Seafile has a block-based storage backend. This means that every file is associated to one or many blocks. If a file is permanently deleted from the trash bin, those blocks need to be cleared in order to free disk space. This process is called garbage collection.
You can manually run the garbage collection with `docker exec`, where `seafile-server` is the name of the container running *seafile-server*:
```
docker exec -it seafile-server /scripts/gc.sh
```
You can schedule a cron job for garbage collection, by adding the following environment variable to *seafile-server*:
```
- GC_CRON=0 6 * * SUN
```
This would run the garbage collection every sunday at 6AM.
If you want to deploy this stack on a Docker Swarm with multiple nodes or if you want to run replicas of the frontend (clustering), there are several things you have to consider first.
**Important:** You can only deploy multiple replicas of the frontend services *seahub* and *seahub-media*. Deploying replicas of the backend or the database would cause data inconsistency or even data corruption.
#### Storage
In order to make the same volumes available to services running on different nodes, you need an advanced storage solution. This could either be distributed storage like GlusterFS and Ceph or a network storage like a NFS share. The volumes are then usually mounted through storage plugins. The repository [marcelo-ochoa/docker-volume-plugins](https://github.com/marcelo-ochoa/docker-volume-plugins) contains some good storage plugins for Docker Swarm.
#### Network
If you have services running on different nodes, which have to communicate to each other, you have to define their network as an overlay network. This will span the network across the whole Swarm.
Set the endpoint mode for the frontend services *seahub* and *seahub-media* to dnsrr. This will enable *seafile-caddy* to see the IPs of all replicas, instead the default virtual IP (VIP) created by the Swarm routing mesh.
The load balancer, in this case *seafile-caddy*, will then create so called sticky sessions, which means that a client connecting with a certain IP will be forwarded to the same service for the time being. Hashing is based on the header `X-Forwarded-For`. This is better than client ip based hashing, when you have another reverse proxy in front of *seafile-caddy*, which is highly recommended. With client ip based hashing *seafile-caddy* would just forward everything to the same container, as it only sees the IP of the reverse proxy. Instead the X-Forwarder-For header contains the actual client IP.
It is also recommended to use dnsrr mode on the *seafile-server*, when you run multiple replicas of *seahub*. This will enable *seafile-server* to see the actual IPs of the *seahub* replicas when they connect to it, instead of a single virtual IP for all of them. This will circumvent probable IP:PORT overlaps in the TCP connection between *seahub* and *seafile-server* if you run many *seahub* replicas.
You can check out this example and use it as a starting point for your Docker Swarm deployment. It is using [lucaslorentz/caddy-docker-proxy](https://github.com/lucaslorentz/caddy-docker-proxy) as the external reverse proxy and the GlusterFS plugin from [marcelo-ochoa/docker-volume-plugins](https://github.com/marcelo-ochoa/docker-volume-plugins). This resembles my personal production setup.
