mirror of https://github.com/matrix-org/dendrite.git synced 2024-11-26 16:21:55 -06:00

Fulltext implementation using Bleve (#2675 )

Based on #2480

This actually indexes events based on their event type. They are removed
from the index if we receive a `m.room.redaction` event on the
`OutputRoomEvent` stream.
An admin endpoint is added to reindex all existing events.


Co-authored-by: Neil Alexander <neilalexander@users.noreply.github.com>

2022-09-27 18:06:49 +02:00

5 KiB

Raw Blame History

title	parent	nav_order	permalink
Configuring Dendrite	Installation	7	/installation/configuration

Configuring Dendrite

A YAML configuration file is used to configure Dendrite. Sample configuration files are present in the top level of the Dendrite repository:

You will need to duplicate the sample, calling it dendrite.yaml for example, and then tailor it to your installation. At a minimum, you will need to populate the following sections:

Server name

First of all, you will need to configure the server name of your Matrix homeserver. This must match the domain name that you have selected whilst configuring the domain name delegation.

In the global section, set the server_name to your delegated domain name:

global:
  # ...
  server_name: example.com

Server signing keys

Next, you should tell Dendrite where to find your server signing keys.

In the global section, set the private_key to the path to your server signing key:

global:
  # ...
  private_key: /path/to/matrix_key.pem

JetStream configuration

Monolith deployments can use the built-in NATS Server rather than running a standalone server. If you are building a polylith deployment, or you want to use a standalone NATS Server anyway, you can also configure that too.

Built-in NATS Server (monolith only)

In the global section, under the jetstream key, ensure that no server addresses are configured and set a storage_path to a persistent folder on the filesystem:

global:
  # ...
  jetstream:
    in_memory: false
    storage_path: /path/to/storage/folder
    topic_prefix: Dendrite

Standalone NATS Server (monolith and polylith)

To use a standalone NATS Server instance, you will need to configure addresses field to point to the port that your NATS Server is listening on:

global:
  # ...
  jetstream:
    addresses:
      - localhost:4222
    topic_prefix: Dendrite

You do not need to configure the storage_path when using a standalone NATS Server instance. In the case that you are connecting to a multi-node NATS cluster, you can configure more than one address in the addresses field.

Database connections

Configuring database connections varies based on the database configuration that you chose.

Global connection pool (monolith with a single PostgreSQL database only)

If you are running a monolith deployment and want to use a single connection pool to a single PostgreSQL database, then you must uncomment and configure the database section within the global section:

global:
  # ...
  database:
    connection_string: postgres://user:pass@hostname/database?sslmode=disable
    max_open_conns: 100
    max_idle_conns: 5
    conn_max_lifetime: -1

You must then remove or comment out the database sections from other areas of the configuration file, e.g. under the app_service_api, federation_api, key_server, media_api, mscs, room_server, sync_api and user_api blocks, otherwise these will override the global database configuration.

Per-component connections (all other configurations)

If you are building a polylith deployment, are using SQLite databases or separate PostgreSQL databases per component, then you must instead configure the database sections under each of the component blocks ,e.g. under the app_service_api, federation_api, key_server, media_api, mscs, room_server, sync_api and user_api blocks.

For example, with PostgreSQL:

room_server:
  # ...
  database:
    connection_string: postgres://user:pass@hostname/dendrite_component?sslmode=disable
    max_open_conns: 10
    max_idle_conns: 2
    conn_max_lifetime: -1

... or with SQLite:

room_server:
  # ...
  database:
    connection_string: file:roomserver.db
    max_open_conns: 10
    max_idle_conns: 2
    conn_max_lifetime: -1

Fulltext search

Dendrite supports experimental fulltext indexing using Bleve, it is configured in the sync_api section as follows. Depending on the language most likely to be used on the server, it might make sense to change the language used when indexing, to ensure the returned results match the expectations. A full list of possible languages can be found here.

sync_api:
  # ...
  search:
    enabled: false
    index_path: "./fulltextindex"
    language: "en"

Other sections

There are other options which may be useful so review them all. In particular, if you are trying to federate from your Dendrite instance into public rooms then configuring the key_perspectives (like matrix.org in the sample) can help to improve reliability considerably by allowing your homeserver to fetch public keys for dead homeservers from another living server.

5 KiB Raw Blame History