Make systemd-with-workers doc official (#7234)

Simplify and update this documentation, and make it part of the core dist.
pull/7248/head
Richard van der Hoff 2020-04-08 11:59:26 +01:00 committed by GitHub
parent c11d24d48c
commit cae4121484
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
10 changed files with 134 additions and 194 deletions

1
changelog.d/7234.doc Normal file
View File

@ -0,0 +1 @@
Update the contributed documentation on managing synapse workers with systemd, and bring it into the core distribution.

View File

@ -1,150 +1,2 @@
# Setup Synapse with Workers and Systemd
This is a setup for managing synapse with systemd including support for
managing workers. It provides a `matrix-synapse`, as well as a
`matrix-synapse-worker@` service for any workers you require. Additionally to
group the required services it sets up a `matrix.target`. You can use this to
automatically start any bot- or bridge-services. More on this in
[Bots and Bridges](#bots-and-bridges).
See the folder [system](system) for any service and target files.
The folder [workers](workers) contains an example configuration for the
`federation_reader` worker. Pay special attention to the name of the
configuration file. In order to work with the `matrix-synapse-worker@.service`
service, it needs to have the exact same name as the worker app.
This setup expects neither the homeserver nor any workers to fork. Forking is
handled by systemd.
## Setup
1. Adjust your matrix configs. Make sure that the worker config files have the
exact same name as the worker app. Compare `matrix-synapse-worker@.service` for
why. You can find an example worker config in the [workers](workers) folder. See
below for relevant settings in the `homeserver.yaml`.
2. Copy the `*.service` and `*.target` files in [system](system) to
`/etc/systemd/system`.
3. `systemctl enable matrix-synapse.service` this adds the homeserver
app to the `matrix.target`
4. *Optional.* `systemctl enable
matrix-synapse-worker@federation_reader.service` this adds the federation_reader
app to the `matrix-synapse.service`
5. *Optional.* Repeat step 4 for any additional workers you require.
6. *Optional.* Add any bots or bridges by enabling them.
7. Start all matrix related services via `systemctl start matrix.target`
8. *Optional.* Enable autostart of all matrix related services on system boot
via `systemctl enable matrix.target`
## Usage
After you have setup you can use the following commands to manage your synapse
installation:
```
# Start matrix-synapse, all workers and any enabled bots or bridges.
systemctl start matrix.target
# Restart matrix-synapse and all workers (not necessarily restarting bots
# or bridges, see "Bots and Bridges")
systemctl restart matrix-synapse.service
# Stop matrix-synapse and all workers (not necessarily restarting bots
# or bridges, see "Bots and Bridges")
systemctl stop matrix-synapse.service
# Restart a specific worker (i. e. federation_reader), the homeserver is
# unaffected by this.
systemctl restart matrix-synapse-worker@federation_reader.service
# Add a new worker (assuming all configs are setup already)
systemctl enable matrix-synapse-worker@federation_writer.service
systemctl restart matrix-synapse.service
```
## The Configs
Make sure the `worker_app` is set in the `homeserver.yaml` and it does not fork.
```
worker_app: synapse.app.homeserver
daemonize: false
```
None of the workers should fork, as forking is handled by systemd. Hence make
sure this is present in all worker config files.
```
worker_daemonize: false
```
The config files of all workers are expected to be located in
`/etc/matrix-synapse/workers`. If you want to use a different location you have
to edit the provided `*.service` files accordingly.
## Bots and Bridges
Most bots and bridges do not care if the homeserver goes down or is restarted.
Depending on the implementation this may crash them though. So look up the docs
or ask the community of the specific bridge or bot you want to run to make sure
you choose the correct setup.
Whichever configuration you choose, after the setup the following will enable
automatically starting (and potentially restarting) your bot/bridge with the
`matrix.target`.
```
systemctl enable <yourBotOrBridgeName>.service
```
**Note** that from an inactive synapse the bots/bridges will only be started with
synapse if you start the `matrix.target`, not if you start the
`matrix-synapse.service`. This is on purpose. Think of `matrix-synapse.service`
as *just* synapse, but `matrix.target` being anything matrix related, including
synapse and any and all enabled bots and bridges.
### Start with synapse but ignore synapse going down
If the bridge can handle shutdowns of the homeserver you'll want to install the
service in the `matrix.target` and optionally add a
`After=matrix-synapse.service` dependency to have the bot/bridge start after
synapse on starting everything.
In this case the service file should look like this.
```
[Unit]
# ...
# Optional, this will only ensure that if you start everything, synapse will
# be started before the bot/bridge will be started.
After=matrix-synapse.service
[Service]
# ...
[Install]
WantedBy=matrix.target
```
### Stop/restart when synapse stops/restarts
If the bridge can't handle shutdowns of the homeserver you'll still want to
install the service in the `matrix.target` but also have to specify the
`After=matrix-synapse.service` *and* `BindsTo=matrix-synapse.service`
dependencies to have the bot/bridge stop/restart with synapse.
In this case the service file should look like this.
```
[Unit]
# ...
# Mandatory
After=matrix-synapse.service
BindsTo=matrix-synapse.service
[Service]
# ...
[Install]
WantedBy=matrix.target
```
The documentation for using systemd to manage synapse workers is now part of
the main synapse distribution. See [docs/systemd-with-workers](../../docs/systemd-with-workers).

View File

@ -1,19 +0,0 @@
[Unit]
Description=Synapse Matrix Worker
After=matrix-synapse.service
BindsTo=matrix-synapse.service
[Service]
Type=notify
NotifyAccess=main
User=matrix-synapse
WorkingDirectory=/var/lib/matrix-synapse
EnvironmentFile=/etc/default/matrix-synapse
ExecStart=/opt/venvs/matrix-synapse/bin/python -m synapse.app.%i --config-path=/etc/matrix-synapse/homeserver.yaml --config-path=/etc/matrix-synapse/conf.d/ --config-path=/etc/matrix-synapse/workers/%i.yaml
ExecReload=/bin/kill -HUP $MAINPID
Restart=always
RestartSec=3
SyslogIdentifier=matrix-synapse-%i
[Install]
WantedBy=matrix-synapse.service

View File

@ -1,7 +0,0 @@
[Unit]
Description=Contains matrix services like synapse, bridges and bots
After=network.target
AllowIsolate=no
[Install]
WantedBy=multi-user.target

View File

@ -0,0 +1,67 @@
# Setting up Synapse with Workers and Systemd
This is a setup for managing synapse with systemd, including support for
managing workers. It provides a `matrix-synapse` service for the master, as
well as a `matrix-synapse-worker@` service template for any workers you
require. Additionally, to group the required services, it sets up a
`matrix-synapse.target`.
See the folder [system](system) for the systemd unit files.
The folder [workers](workers) contains an example configuration for the
`federation_reader` worker.
## Synapse configuration files
See [workers.md](../workers.md) for information on how to set up the
configuration files and reverse-proxy correctly. You can find an example worker
config in the [workers](workers) folder.
Systemd manages daemonization itself, so ensure that none of the configuration
files set either `daemonize` or `worker_daemonize`.
The config files of all workers are expected to be located in
`/etc/matrix-synapse/workers`. If you want to use a different location, edit
the provided `*.service` files accordingly.
There is no need for a separate configuration file for the master process.
## Set up
1. Adjust synapse configuration files as above.
1. Copy the `*.service` and `*.target` files in [system](system) to
`/etc/systemd/system`.
1. Run `systemctl deamon-reload` to tell systemd to load the new unit files.
1. Run `systemctl enable matrix-synapse.service`. This will configure the
synapse master process to be started as part of the `matrix-synapse.target`
target.
1. For each worker process to be enabled, run `systemctl enable
matrix-synapse-worker@<worker_name>.service`. For each `<worker_name>`, there
should be a corresponding configuration file
`/etc/matrix-synapse/workers/<worker_name>.yaml`.
1. Start all the synapse processes with `systemctl start matrix-synapse.target`.
1. Tell systemd to start synapse on boot with `systemctl enable matrix-synapse.target`/
## Usage
Once the services are correctly set up, you can use the following commands
to manage your synapse installation:
```sh
# Restart Synapse master and all workers
systemctl restart matrix-synapse.target
# Stop Synapse and all workers
systemctl stop matrix-synapse.target
# Restart the master alone
systemctl start matrix-synapse.service
# Restart a specific worker (eg. federation_reader); the master is
# unaffected by this.
systemctl restart matrix-synapse-worker@federation_reader.service
# Add a new worker (assuming all configs are set up already)
systemctl enable matrix-synapse-worker@federation_writer.service
systemctl restart matrix-synapse.target
```

View File

@ -0,0 +1,20 @@
[Unit]
Description=Synapse %i
# This service should be restarted when the synapse target is restarted.
PartOf=matrix-synapse.target
[Service]
Type=notify
NotifyAccess=main
User=matrix-synapse
WorkingDirectory=/var/lib/matrix-synapse
EnvironmentFile=/etc/default/matrix-synapse
ExecStart=/opt/venvs/matrix-synapse/bin/python -m synapse.app.generic_worker --config-path=/etc/matrix-synapse/homeserver.yaml --config-path=/etc/matrix-synapse/conf.d/ --config-path=/etc/matrix-synapse/workers/%i.yaml
ExecReload=/bin/kill -HUP $MAINPID
Restart=always
RestartSec=3
SyslogIdentifier=matrix-synapse-%i
[Install]
WantedBy=matrix-synapse.target

View File

@ -1,5 +1,8 @@
[Unit]
Description=Synapse Matrix Homeserver
Description=Synapse master
# This service should be restarted when the synapse target is restarted.
PartOf=matrix-synapse.target
[Service]
Type=notify
@ -15,4 +18,4 @@ RestartSec=3
SyslogIdentifier=matrix-synapse
[Install]
WantedBy=matrix.target
WantedBy=matrix-synapse.target

View File

@ -0,0 +1,6 @@
[Unit]
Description=Synapse parent target
After=network.target
[Install]
WantedBy=multi-user.target

View File

@ -10,5 +10,4 @@ worker_listeners:
resources:
- names: [federation]
worker_daemonize: false
worker_log_config: /etc/matrix-synapse/federation-reader-log.yaml

View File

@ -52,24 +52,20 @@ synapse process.)
You then create a set of configs for the various worker processes. These
should be worker configuration files, and should be stored in a dedicated
subdirectory, to allow synctl to manipulate them. An additional configuration
for the master synapse process will need to be created because the process will
not be started automatically. That configuration should look like this:
worker_app: synapse.app.homeserver
daemonize: true
subdirectory, to allow synctl to manipulate them.
Each worker configuration file inherits the configuration of the main homeserver
configuration file. You can then override configuration specific to that worker,
e.g. the HTTP listener that it provides (if any); logging configuration; etc.
You should minimise the number of overrides though to maintain a usable config.
You must specify the type of worker application (`worker_app`). The currently
available worker applications are listed below. You must also specify the
replication endpoints that it's talking to on the main synapse process.
`worker_replication_host` should specify the host of the main synapse,
`worker_replication_port` should point to the TCP replication listener port and
`worker_replication_http_port` should point to the HTTP replication port.
In the config file for each worker, you must specify the type of worker
application (`worker_app`). The currently available worker applications are
listed below. You must also specify the replication endpoints that it's talking
to on the main synapse process. `worker_replication_host` should specify the
host of the main synapse, `worker_replication_port` should point to the TCP
replication listener port and `worker_replication_http_port` should point to
the HTTP replication port.
Currently, the `event_creator` and `federation_reader` workers require specifying
`worker_replication_http_port`.
@ -90,8 +86,6 @@ For instance:
- names:
- client
worker_daemonize: True
worker_pid_file: /home/matrix/synapse/synchrotron.pid
worker_log_config: /home/matrix/synapse/config/synchrotron_log_config.yaml
...is a full configuration for a synchrotron worker instance, which will expose a
@ -101,7 +95,31 @@ by the main synapse.
Obviously you should configure your reverse-proxy to route the relevant
endpoints to the worker (`localhost:8083` in the above example).
Finally, to actually run your worker-based synapse, you must pass synctl the -a
Finally, you need to start your worker processes. This can be done with either
`synctl` or your distribution's preferred service manager such as `systemd`. We
recommend the use of `systemd` where available: for information on setting up
`systemd` to start synapse workers, see
[systemd-with-workers](systemd-with-workers). To use `synctl`, see below.
### Using synctl
If you want to use `synctl` to manage your synapse processes, you will need to
create an an additional configuration file for the master synapse process. That
configuration should look like this:
```yaml
worker_app: synapse.app.homeserver
```
Additionally, each worker app must be configured with the name of a "pid file",
to which it will write its process ID when it starts. For example, for a
synchrotron, you might write:
```yaml
worker_pid_file: /home/matrix/synapse/synchrotron.pid
```
Finally, to actually run your worker-based synapse, you must pass synctl the `-a`
commandline option to tell it to operate on all the worker configurations found
in the given directory, e.g.: