matrix-docker-ansible-deploy/docs/howto-server-delegation.md

139 lines
7.7 KiB
Markdown
Raw Normal View History

# Server Delegation
To have a server on a subdomain (e.g. `matrix.<your-domain>`) handle Matrix federation traffic for the base domain (`<your-domain>`), we need to instruct the Matrix network of such a delegation.
By default, this playbook guides you into setting up [Server Delegation via a well-known file](#server-delegation-via-a-well-known-file).
However, that method may have some downsides that are not to your liking. Hence this guide about alternative ways to set up Server Delegation.
It is a complicated matter, so unless you are affected by the [Downsides of well-known-based Server Delegation](#downsides-of-well-known-based-server-delegation), we suggest you stay on the simple/default path.
## Server Delegation via a well-known file
Serving a `/.well-known/matrix/server` file from the base domain is the most straightforward way to set up server delegation, but it suffers from the following problems:
As we already mention in [Configuring DNS](configuring-dns.md) and [Configuring Service Discovery via .well-known](configuring-well-known.md),
this playbook already properly guides you into setting up such delegation by means of a `/.well-known/matrix/server` file served from the base domain (`<your-domain>`).
If this is okay with you, feel free to not read ahead.
### Downsides of well-known-based Server Delegation
Server Delegation by means of a `/.well-known/matrix/server` file is the most straightforward, but suffers from the following downsides:
- you need to have a working HTTPS server for the base domain (`<your-domain>`)
- any downtime on the base domain (`<your-domain>`) or network trouble between the matrix subdomain (`matrix.<your-domain>`) and the base `<domain>` may cause Matrix Federation outages. As the [Server-Server spec says](https://matrix.org/docs/spec/server_server/r0.1.0.html#server-discovery):
> Errors are recommended to be cached for up to an hour, and servers are encouraged to exponentially back off for repeated failures.
If this is not a concern for you, feel free to not read ahead.
Otherwise, you can decide to go against the default for this playbook, and instead set up [Server Delegation via a DNS SRV record (advanced)](#server-delegation-via-a-dns-serv-record-advanced).
## Server Delegation via a DNS SRV record (advanced)
**NOTE**: doing Server Delegation via a DNS SRV record is a more advanced way to do it and is not the default for this playbook.
As per the [Server-Server spec](https://matrix.org/docs/spec/server_server/r0.1.0.html#server-discovery), it's possible to do Server Delegation using only a SRV record (without a `/.well-known/matrix/server` file).
This prevents you from suffering the [Downsides of well-known-based Server Delegation](#downsides-of-well-known-based-server-delegation).
To use DNS SRV record validation, you need to:
- ensure that no `/.well-known/matrix/server` is served from the base domain, as that would interfere with DNS SRV record Server Delegation. To make the playbook **not** generate and serve the file, use the following configuration: `matrix_well_known_matrix_server_enabled: false`.
- ensure that you have a `_matrix._tcp` DNS SRV record for your base domain (`<your-domain>`) with a value of `10 0 8448 matrix.<your-domain>`
- ensure that you are serving the Matrix Federation API (tcp/8448) with a certificate for `<your-domain>` (not `matrix.<your-domain>`!). See below.
### Obtaining certificates
How you can obtain a valid certificate for `<your-domain>` on the `matrix.<your-domain>` server is up to you.
If `<your-domain>` and `matrix.<your-domain>` is the same machine, you can let the playbook obtain the certificate for you by redefining the `matrix_ssl_domains_to_obtain_certificates_for` variable. Example:
```yaml
matrix_ssl_domains_to_obtain_certificates_for:
- '{{ hostname_matrix }}'
- '{{ hostname_riot }}'
- '{{ hostname_identity }}'
```
This way, the playbook would obtain certificates for your base domain as well (referred to by the `hostname_identity` variable).
The certificate files would be available in `/matrix/ssl/config/live/<your-domain>/...`.
If `<your-domain>` and `matrix.<your-domain>` are not the same machine, you can copy over the certificate files manually.
If they get renewed automatically, you may also have to transfer them periodically. How often you do that is up to you, as long as the certificate files don't expire.
### Serving the Federation API with your certificates
Regardless of which method for obtaining certificates you've used, once you've managed to get certificates for your base domain onto the `matrix.<your-domain>` machine you can put them to use.
### Serving the Federation API with your certificates and matrix-nginx-proxy
**If you are using matrix-nginx-proxy**, a reverse-proxy webserver used by default in this playbook, you only need to override the certificates used for the Matrix Federation API. You can do that using:
```yaml
# Adjust paths below to point to your certificate.
#
# NOTE: these are in-container paths. `/matrix/ssl` on the host is mounted into the container
# at the same path (`/matrix/ssl`) by default, so if that's the path you need, it would be seamless.
matrix_nginx_proxy_proxy_matrix_federation_api_ssl_certificate: /matrix/ssl/config/live/<your-domain>/fullchain.pem
matrix_nginx_proxy_proxy_matrix_federation_api_ssl_certificate_key: /matrix/ssl/config/live/<your-domain>/privkey.pem
```
If your files are not in `/matrix/ssl` but in some other location, you would need to mount them into the container:
```yaml
matrix_nginx_proxy_container_additional_volumes:
- src: /some/path/on/the/host
dst: /some/path/inside/the/container
options: ro
```
You then refer to them (for `matrix_nginx_proxy_proxy_matrix_federation_api_ssl_certificate` and `matrix_nginx_proxy_proxy_matrix_federation_api_ssl_certificate_key`) by using `/some/path/inside/the/container`.
2019-02-05 19:38:20 +08:00
Make sure to reload matrix-nginx-proxy once in a while (`systemctl reload matrix-nginx-proxy`), so that newer certificates can kick in.
Reloading doesn't cause any downtime.
### Serving the Federation API with your certificates and another webserver
**If you are NOT using matrix-nginx-proxy**, but rather some other webserver, you can set up reverse-proxying for the `tcp/8448` port by yourself.
Make sure to use the proper certificates for `<your-domain>` (not for `matrix.<your-domain>`) when serving the `tcp/8448` port.
Proxying needs to happen to `127.0.0.1:8048` (unencrypted Synapse federation listener).
2019-02-05 19:38:20 +08:00
Make sure to reload/restart your webserver once in a while, so that newer certificates can kick in.
### Serving the Federation API with your certificates and Synapse handling Federation
**Alternatively**, if you are **NOT using matrix-nginx-proxy** and **would rather not use your own webserver for Federation traffic**, you can let Synapse handle Federation by itself.
To do that, make sure the certificate files are mounted into the Synapse container:
```yaml
matrix_synapse_container_additional_volumes:
- src: /some/path/on/the/host
dst: /some/path/inside/the/container
options: ro
```
You can then tell Synapse to serve Federation traffic over TLS on `tcp/8448`:
```yaml
matrix_synapse_no_tls: false
matrix_synapse_tls_federation_listener_enabled: true
matrix_synapse_tls_certificate_path: /some/path/inside/the/container/certificate.crt
matrix_synapse_tls_private_key_path: /some/path/inside/the/container/private.key
```
2019-02-05 19:38:20 +08:00
Every once in a while (before the certificates expire), you'll need to completely restart Synapse (unless [Synapse becomes capable of reloading certificates without restarting - issue #1180](https://github.com/matrix-org/synapse/issues/1180)). Restarting Synapse can be done like this: `systemctl restart matrix-synapse`. Restarting causes some downtime.