Currently, GtS only supports using the built-in LE client directly for
TLS. However, admins may still want to use GtS directly (so without a
reverse proxy) but with certificates provided through some other
mechanism. They may have some centralised way of provisioning these
things themselves, or simply prefer to use LE but with a different
challenge like DNS-01 which is not supported by autocert.
This adds support for loading a public/private keypair from disk instead
of using LE and reconfigures the server to use a TLS listener if we
succeed in doing so.
Additionally, being able to load TLS keypair from disk opens up the path
to using a custom CA for testing purposes avoinding the need for a
constellation of containers and something like Pebble or Step CA to
provide LE APIs.
* [bug] Fix nginx fileserver caching example
This updates the example to ensure the nginx proxies the request on to
GTS if the file is not found on disk. This can happen due to media
pruning.
* [chore] Set cache-control in nginx to private
This makes the header match with the backend. For things from the
fileserver it may not be appropriate for anything other than a private
cache (i.e the client) to cache things.
* serve publickey separately from AP, don't throttle it
* update nginx cache documentation, cache main-key too
* throttle public key, but separately from other endpoints
since 2.4.47 (released April 22nd 2021), Apache httpd can ProxyPass to
websockets on the same URL, without mod_rewrite (and, without
mod_proxy_wstunnel).
Lots of these were appearing:
```
*459 connect() failed (111: Connection refused) while connecting to upstream
```
This change resolves it, see https://stackoverflow.com/a/52550758
* [docs] Serve static assets with nginx
This explains how to use nginx to serve static assets and offload GTS
from that responsibility. It also shows how to have nginx add caching
headers to indicate to clients how long they may cache an asset.
* [docs] Move additional nginx config to advanced
This moves a bunch of additional nginx configuration into the Advanced
page instead. It declutters the nginx configuration page.
This explains how nginx can be used to cache webfinger responses and
potentially serve stale responses in case GTS is down. This can be
useful to do in order to ensure webfinger keeps working even if you're
doing some maintenance.
* remove filesystem logging directives from example systemd unit config
* [docs] Update docs to reflect new systemd config
Co-authored-by: tsmethurst <tobi.smethurst@protonmail.com>
* Enable the 'admonitions' Markdown extension for Mkdocs.
The admonitions extension to Python-Markdown allows you to include
rST-style "admonitions" to Markdown documents, for instance,
!!! note
Here's an important note to keep in mind!
In general, the current documentation uses bold text to try to achieve
the same effect, which is a bit harder to notice and makes it difficult
to differentiate between "here's something useful to know" versus "here
there be dragons".
* Add AppArmor profile and documentation for LSM-related sandboxing
This commit adds an AppArmor profile for gotosocial in
examples/apparmor/gotosocial. This will (hopefully) serve as a helpful
security mitigation for people are planning on deploying GTS on a
Debian-family Linux distribution.
I've also updates the documentation to include some information about
deploying GTS with either AppArmor or SELinux (moving the documentation
for the former out of the "binary installation guide" docs).
Since the documentation site only shows the latest version of the
docs, we need the docs to explain how to use the latest stable
release, not just the latest git version.
The NewSignup method was already being called with
requireApproval=false, but it had emailVerified=false as well, which
meant that it was required to use the `admin account confirm` command
to verify the email before the newly-created user could log in.
I think that was probably an oversight; effectively it did require
approval anyway. Changing emailVerified to true allows you to just
create the account and log in immediately, reducing the opportunity
for manual error to sneak in.
Also updated the docs to remove the mention of needing to confirm new
accounts. However, I've left the confirmation command alone because I
think once we have web signups, it will be needed in that context.
* [bugfix] Correctly style inputs and buttons
<input>, <textarea>, and <button> were incorrectly using the system-ui font previously; this commit fixes that. text-align: center; was added to <button> due to an inconsistency with .button where text-align would be off.
* [chore] Update binary installation instructions
This commit updates the example release mentioned in the docs, and mentions Caddy in the reverse proxy options.
* [bugfix] Remove redundant Caddyfile
Caddy automatically upgrades HTTP to HTTPS (see https://caddyserver.com/docs/automatic-https) so the upgrading part of the Caddyfile is redundant.
* Add instructions for working with Caddy 2
Add instructions for working with Caddy 2. Some texts are duplicated from the NGINX part (mainly the configuration part, about systemctl).
* Add new Caddy docs to mkdocs.yml
Adds the new Caddy document to the document tree
* Remove up and downstream headers from configuration
Removed the header statements for the proxy, as proxying those are transparant. Kept the flush_interval directive, disabling the response buffer completely so we can write without delays.
* Update caddy.md
Corrects the opening link
* Apply comments mentioned in PR
@igalic mentioned a few comments to improve these docs. I've processed those in this PR.
* docs: unify nginx explainers and add apache httpd
there are two places where nginx + certbot is explained, unify that into
one place.
Add apache httpd, following the same steps, but using mod_md for
LetsEncrypt
add a note about #453 in both guides.
Link to both, and call the section reverse proxy, instead of NGINX
* restore full nginx.conf from docker.md
* add installation_guide/apache-httpd.md to mkdocs
* exludes docker-volume from exemple/docker-compose
* Adds an docker-compose.yaml example and adds some readme to the
docker.md
* Changes Docker-Compose Example
* Configures docker-compose.yaml and docker.md
* Adds some cosmetics to the Documentation
* Adds UserID:GroupID
* Describes how to create a new user
When I tried to start the server using the command `./gotosocial server start --config-file ./config.yaml`, I got the following error:
```
root@gotosocial-experiment:/gotosocial# ./gotosocial server start --config-file ./config.yaml
Incorrect Usage: flag provided but not defined: -config-file
NAME:
gotosocial server start - start the gotosocial server
USAGE:
gotosocial server start [command options] [arguments...]
OPTIONS:
--help, -h show help (default: false)
FATA[0000] flag provided but not defined: -config-file
```
Putting the flag after the `./gotosocial` but before the `server start` seems to work properly
* start pulling out + replacing urfave and config
* replace many many instances of config
* move more stuff => viper
* properly remove urfave
* move some flags to root command
* add testrig commands to root
* alias config file keys
* start adding cli parsing tests
* reorder viper init
* remove config path alias
* fmt
* change config file keys to non-nested
* we're more or less in business now
* tidy up the common func
* go fmt
* get tests passing again
* add note about the cliparsing tests
* reorganize
* update docs with changes
* structure cmd dir better
* rename + move some files around
* fix dangling comma
