Merge remote-tracking branch 'origin/develop' into matrix-org-hotfixes

pull/13598/head
Erik Johnston 2022-07-06 12:11:24 +01:00
commit 8881bd77d4
111 changed files with 1569 additions and 534 deletions

View File

@ -0,0 +1,36 @@
#!/bin/sh
#
# Common commands to set up Complement's prerequisites in a GitHub Actions CI run.
#
# Must be called after Synapse has been checked out to `synapse/`.
#
set -eu
alias block='{ set +x; } 2>/dev/null; func() { echo "::group::$*"; set -x; }; func'
alias endblock='{ set +x; } 2>/dev/null; func() { echo "::endgroup::"; set -x; }; func'
block Set Go Version
# The path is set via a file given by $GITHUB_PATH. We need both Go 1.17 and GOPATH on the path to run Complement.
# See https://docs.github.com/en/actions/using-workflows/workflow-commands-for-github-actions#adding-a-system-path
# Add Go 1.17 to the PATH: see https://github.com/actions/virtual-environments/blob/main/images/linux/Ubuntu2004-Readme.md#environment-variables-2
echo "$GOROOT_1_17_X64/bin" >> $GITHUB_PATH
# Add the Go path to the PATH: We need this so we can call gotestfmt
echo "~/go/bin" >> $GITHUB_PATH
endblock
block Install Complement Dependencies
sudo apt-get -qq update && sudo apt-get install -qqy libolm3 libolm-dev
go get -v github.com/haveyoudebuggedit/gotestfmt/v2/cmd/gotestfmt@latest
endblock
block Install custom gotestfmt template
mkdir .gotestfmt/github -p
cp synapse/.ci/complement_package.gotpl .gotestfmt/github/package.gotpl
endblock
block Check out Complement
# Attempt to check out the same branch of Complement as the PR. If it
# doesn't exist, fallback to HEAD.
synapse/.ci/scripts/checkout_complement.sh
endblock

View File

@ -328,85 +328,21 @@ jobs:
- arrangement: monolith
database: Postgres
- arrangement: workers
database: Postgres
steps:
# The path is set via a file given by $GITHUB_PATH. We need both Go 1.17 and GOPATH on the path to run Complement.
# See https://docs.github.com/en/actions/using-workflows/workflow-commands-for-github-actions#adding-a-system-path
- name: "Set Go Version"
run: |
# Add Go 1.17 to the PATH: see https://github.com/actions/virtual-environments/blob/main/images/linux/Ubuntu2004-Readme.md#environment-variables-2
echo "$GOROOT_1_17_X64/bin" >> $GITHUB_PATH
# Add the Go path to the PATH: We need this so we can call gotestfmt
echo "~/go/bin" >> $GITHUB_PATH
- name: "Install Complement Dependencies"
run: |
sudo apt-get update && sudo apt-get install -y libolm3 libolm-dev
go get -v github.com/haveyoudebuggedit/gotestfmt/v2/cmd/gotestfmt@latest
- name: Run actions/checkout@v2 for synapse
uses: actions/checkout@v2
with:
path: synapse
- name: "Install custom gotestfmt template"
run: |
mkdir .gotestfmt/github -p
cp synapse/.ci/complement_package.gotpl .gotestfmt/github/package.gotpl
# Attempt to check out the same branch of Complement as the PR. If it
# doesn't exist, fallback to HEAD.
- name: Checkout complement
run: synapse/.ci/scripts/checkout_complement.sh
- name: Prepare Complement's Prerequisites
run: synapse/.ci/scripts/setup_complement_prerequisites.sh
- run: |
set -o pipefail
POSTGRES=${{ (matrix.database == 'Postgres') && 1 || '' }} COMPLEMENT_DIR=`pwd`/complement synapse/scripts-dev/complement.sh -json 2>&1 | gotestfmt
shell: bash
name: Run Complement Tests
# We only run the workers tests on `develop` for now, because they're too slow to wait for on PRs.
# Sadly, you can't have an `if` condition on the value of a matrix, so this is a temporary, separate job for now.
# GitHub Actions doesn't support YAML anchors, so it's full-on duplication for now.
complement-developonly:
if: "${{ !failure() && !cancelled() && (github.ref == 'refs/heads/develop') }}"
needs: linting-done
runs-on: ubuntu-latest
name: "Complement Workers (develop only)"
steps:
# The path is set via a file given by $GITHUB_PATH. We need both Go 1.17 and GOPATH on the path to run Complement.
# See https://docs.github.com/en/actions/using-workflows/workflow-commands-for-github-actions#adding-a-system-path
- name: "Set Go Version"
run: |
# Add Go 1.17 to the PATH: see https://github.com/actions/virtual-environments/blob/main/images/linux/Ubuntu2004-Readme.md#environment-variables-2
echo "$GOROOT_1_17_X64/bin" >> $GITHUB_PATH
# Add the Go path to the PATH: We need this so we can call gotestfmt
echo "~/go/bin" >> $GITHUB_PATH
- name: "Install Complement Dependencies"
run: |
sudo apt-get -qq update && sudo apt-get install -qqy libolm3 libolm-dev
go get -v github.com/haveyoudebuggedit/gotestfmt/v2/cmd/gotestfmt@latest
- name: Run actions/checkout@v2 for synapse
uses: actions/checkout@v2
with:
path: synapse
- name: "Install custom gotestfmt template"
run: |
mkdir .gotestfmt/github -p
cp synapse/.ci/complement_package.gotpl .gotestfmt/github/package.gotpl
# Attempt to check out the same branch of Complement as the PR. If it
# doesn't exist, fallback to HEAD.
- name: Checkout complement
run: synapse/.ci/scripts/checkout_complement.sh
- run: |
set -o pipefail
WORKERS=1 COMPLEMENT_DIR=`pwd`/complement synapse/scripts-dev/complement.sh -json 2>&1 | gotestfmt
POSTGRES=${{ (matrix.database == 'Postgres') && 1 || '' }} WORKERS=${{ (matrix.arrangement == 'workers') && 1 || '' }} COMPLEMENT_DIR=`pwd`/complement synapse/scripts-dev/complement.sh -json 2>&1 | gotestfmt
shell: bash
name: Run Complement Tests

View File

@ -96,6 +96,51 @@ jobs:
/logs/results.tap
/logs/**/*.log*
complement:
if: "${{ !failure() && !cancelled() }}"
runs-on: ubuntu-latest
strategy:
fail-fast: false
matrix:
include:
- arrangement: monolith
database: SQLite
- arrangement: monolith
database: Postgres
- arrangement: workers
database: Postgres
steps:
- name: Run actions/checkout@v2 for synapse
uses: actions/checkout@v2
with:
path: synapse
- name: Prepare Complement's Prerequisites
run: synapse/.ci/scripts/setup_complement_prerequisites.sh
# This step is specific to the 'Twisted trunk' test run:
- name: Patch dependencies
run: |
set -x
DEBIAN_FRONTEND=noninteractive sudo apt-get install -yqq python3 pipx
pipx install poetry==1.1.12
poetry remove -n twisted
poetry add -n --extras tls git+https://github.com/twisted/twisted.git#trunk
poetry lock --no-update
# NOT IN 1.1.12 poetry lock --check
working-directory: synapse
- run: |
set -o pipefail
TEST_ONLY_SKIP_DEP_HASH_VERIFICATION=1 POSTGRES=${{ (matrix.database == 'Postgres') && 1 || '' }} WORKERS=${{ (matrix.arrangement == 'workers') && 1 || '' }} COMPLEMENT_DIR=`pwd`/complement synapse/scripts-dev/complement.sh -json 2>&1 | gotestfmt
shell: bash
name: Run Complement Tests
# open an issue if the build fails, so we know about it.
open-issue:
if: failure()
@ -103,6 +148,7 @@ jobs:
- mypy
- trial
- sytest
- complement
runs-on: ubuntu-latest

View File

@ -1,11 +1,19 @@
Synapse 1.62.0 (2022-07-05)
===========================
No significant changes since 1.62.0rc3.
Authors of spam-checker plugins should consult the [upgrade notes](https://github.com/matrix-org/synapse/blob/release-v1.62/docs/upgrade.md#upgrading-to-v1620) to learn about the enriched signatures for spam checker callbacks, which are supported with this release of Synapse.
Synapse 1.62.0rc3 (2022-07-04)
==============================
Bugfixes
--------
- Update the version of the [ldap3 plugin](https://github.com/matrix-org/matrix-synapse-ldap3/) includled in matrix.org docker images and debian packages to 0.2.1. This fixes [problems involving usernames that have uppercase characters](https://github.com/matrix-org/matrix-synapse-ldap3/pull/163). ([\#13156](https://github.com/matrix-org/synapse/issues/13156))
- Fix unread counts for users on small servers. Introduced in v1.62.0rc1. ([\#13168](https://github.com/matrix-org/synapse/issues/13168))
- Update the version of the [ldap3 plugin](https://github.com/matrix-org/matrix-synapse-ldap3/) included in the `matrixdotorg/synapse` DockerHub images and the Debian packages hosted on `packages.matrix.org` to 0.2.1. This fixes [a bug](https://github.com/matrix-org/matrix-synapse-ldap3/pull/163) with usernames containing uppercase characters. ([\#13156](https://github.com/matrix-org/synapse/issues/13156))
- Fix a bug introduced in Synapse 1.62.0rc1 affecting unread counts for users on small servers. ([\#13168](https://github.com/matrix-org/synapse/issues/13168))
Synapse 1.62.0rc2 (2022-07-01)

1
changelog.d/13028.misc Normal file
View File

@ -0,0 +1 @@
Add type annotations to `tests.utils`.

1
changelog.d/13029.doc Normal file
View File

@ -0,0 +1 @@
Add an explanation of the `--report-stats` argument to the docs.

View File

@ -0,0 +1 @@
Implement [MSC3827](https://github.com/matrix-org/matrix-spec-proposals/pull/3827): Filtering of /publicRooms by room type.

3
changelog.d/13077.doc Normal file
View File

@ -0,0 +1,3 @@
Clean up references to sample configuration and redirect users to the configuration manual instead.

1
changelog.d/13079.misc Normal file
View File

@ -0,0 +1 @@
Enable Complement testing in the 'Twisted Trunk' CI runs.

1
changelog.d/13086.doc Normal file
View File

@ -0,0 +1 @@
Add documentation for anonymised homeserver statistics collection.

1
changelog.d/13100.misc Normal file
View File

@ -0,0 +1 @@
Faster room joins: Handle race between persisting an event and un-partial stating a room.

1
changelog.d/13103.misc Normal file
View File

@ -0,0 +1 @@
Add missing type hints to `synapse.logging`.

1
changelog.d/13113.misc Normal file
View File

@ -0,0 +1 @@
Raise a `DependencyError` on missing dependencies instead of a `ConfigError`.

1
changelog.d/13116.doc Normal file
View File

@ -0,0 +1 @@
Fix wrong section header for `allow_public_rooms_over_federation` in the homeserver config documentation.

1
changelog.d/13119.misc Normal file
View File

@ -0,0 +1 @@
Reduce DB usage of `/sync` when a large number of unread messages have recently been sent in a room.

View File

@ -0,0 +1 @@
Add a rate limit for local users sending invites.

1
changelog.d/13127.misc Normal file
View File

@ -0,0 +1 @@
Improve startup times in Complement test runs against workers, particularly in CPU-constrained environments.

1
changelog.d/13129.misc Normal file
View File

@ -0,0 +1 @@
Only one-line SQL statements for logging and tracing.

1
changelog.d/13131.bugfix Normal file
View File

@ -0,0 +1 @@
Fix application service not being able to join remote federated room without a profile set.

1
changelog.d/13134.misc Normal file
View File

@ -0,0 +1 @@
Apply ratelimiting earlier in processing of /send request.

1
changelog.d/13135.misc Normal file
View File

@ -0,0 +1 @@
Enforce type annotations for `tests.test_server`.

1
changelog.d/13136.misc Normal file
View File

@ -0,0 +1 @@
Add type annotations to `tests.server`.

1
changelog.d/13139.doc Normal file
View File

@ -0,0 +1 @@
Add a link to the configuration manual from the homeserver sample config documentation.

1
changelog.d/13143.misc Normal file
View File

@ -0,0 +1 @@
Add support to `complement.sh` for skipping the docker build.

1
changelog.d/13144.misc Normal file
View File

@ -0,0 +1 @@
Faster joins: skip waiting for full state when processing incoming events over federation.

1
changelog.d/13145.misc Normal file
View File

@ -0,0 +1 @@
Improve exception handling when processing events received over federation.

View File

@ -0,0 +1 @@
Improve validation logic in Synapse's REST endpoints.

1
changelog.d/13152.misc Normal file
View File

@ -0,0 +1 @@
Add the ability to set the log level using the `SYNAPSE_TEST_LOG_LEVEL` environment when using `complement.sh`.

1
changelog.d/13153.misc Normal file
View File

@ -0,0 +1 @@
Reduce DB usage of `/sync` when a large number of unread messages have recently been sent in a room.

1
changelog.d/13157.misc Normal file
View File

@ -0,0 +1 @@
Enable Complement testing in the 'Twisted Trunk' CI runs.

1
changelog.d/13159.misc Normal file
View File

@ -0,0 +1 @@
Improve and fix type hints.

1
changelog.d/13166.doc Normal file
View File

@ -0,0 +1 @@
Add missing links to config options.

1
changelog.d/13167.misc Normal file
View File

@ -0,0 +1 @@
Update config used by Complement to allow device name lookup over federation.

1
changelog.d/13194.bugfix Normal file
View File

@ -0,0 +1 @@
Fix bug where rows were not deleted from `event_push_actions` table on large servers. Introduced in v1.62.0.

6
debian/changelog vendored
View File

@ -1,3 +1,9 @@
matrix-synapse-py3 (1.62.0) stable; urgency=medium
* New Synapse release 1.62.0.
-- Synapse Packaging team <packages@matrix.org> Tue, 05 Jul 2022 11:14:15 +0100
matrix-synapse-py3 (1.62.0~rc3) stable; urgency=medium
* New Synapse release 1.62.0rc3.

View File

@ -62,7 +62,13 @@ WORKDIR /synapse
# Copy just what we need to run `poetry export`...
COPY pyproject.toml poetry.lock /synapse/
RUN /root/.local/bin/poetry export --extras all -o /synapse/requirements.txt
# If specified, we won't verify the hashes of dependencies.
# This is only needed if the hashes of dependencies cannot be checked for some
# reason, such as when a git repository is used directly as a dependency.
ARG TEST_ONLY_SKIP_DEP_HASH_VERIFICATION
RUN /root/.local/bin/poetry export --extras all -o /synapse/requirements.txt ${TEST_ONLY_SKIP_DEP_HASH_VERIFICATION:+--without-hashes}
###
### Stage 1: builder
@ -85,6 +91,7 @@ RUN \
openssl \
rustc \
zlib1g-dev \
git \
&& rm -rf /var/lib/apt/lists/*
# To speed up rebuilds, install all of the dependencies before we copy over

View File

@ -67,6 +67,13 @@ The following environment variables are supported in `generate` mode:
* `UID`, `GID`: the user id and group id to use for creating the data
directories. If unset, and no user is set via `docker run --user`, defaults
to `991`, `991`.
* `SYNAPSE_LOG_LEVEL`: the log level to use (one of `DEBUG`, `INFO`, `WARNING` or `ERROR`).
Defaults to `INFO`.
* `SYNAPSE_LOG_SENSITIVE`: if set and the log level is set to `DEBUG`, Synapse
will log sensitive information such as access tokens.
This should not be needed unless you are a developer attempting to debug something
particularly tricky.
## Postgres

View File

@ -59,6 +59,9 @@ if [[ -n "$SYNAPSE_COMPLEMENT_USE_WORKERS" ]]; then
synchrotron, \
appservice, \
pusher"
# Improve startup times by using a launcher based on fork()
export SYNAPSE_USE_EXPERIMENTAL_FORKING_LAUNCHER=1
else
# Empty string here means 'main process only'
export SYNAPSE_WORKER_TYPES=""

View File

@ -81,6 +81,8 @@ rc_invites:
federation_rr_transactions_per_room_per_second: 9999
allow_device_name_lookup_over_federation: true
## Experimental Features ##
experimental_features:

View File

@ -1,3 +1,24 @@
{% if use_forking_launcher %}
[program:synapse_fork]
command=/usr/local/bin/python -m synapse.app.complement_fork_starter
{{ main_config_path }}
synapse.app.homeserver
--config-path="{{ main_config_path }}"
--config-path=/conf/workers/shared.yaml
{%- for worker in workers %}
-- {{ worker.app }}
--config-path="{{ main_config_path }}"
--config-path=/conf/workers/shared.yaml
--config-path=/conf/workers/{{ worker.name }}.yaml
{%- endfor %}
stdout_logfile=/dev/stdout
stdout_logfile_maxbytes=0
stderr_logfile=/dev/stderr
stderr_logfile_maxbytes=0
autorestart=unexpected
exitcodes=0
{% else %}
[program:synapse_main]
command=/usr/local/bin/prefix-log /usr/local/bin/python -m synapse.app.homeserver
--config-path="{{ main_config_path }}"
@ -13,7 +34,7 @@ autorestart=unexpected
exitcodes=0
{% for worker in workers %}
{% for worker in workers %}
[program:synapse_{{ worker.name }}]
command=/usr/local/bin/prefix-log /usr/local/bin/python -m {{ worker.app }}
--config-path="{{ main_config_path }}"
@ -27,4 +48,5 @@ stdout_logfile_maxbytes=0
stderr_logfile=/dev/stderr
stderr_logfile_maxbytes=0
{% endfor %}
{% endfor %}
{% endif %}

View File

@ -2,7 +2,11 @@ version: 1
formatters:
precise:
{% if include_worker_name_in_log_line %}
format: '{{ worker_name }} | %(asctime)s - %(name)s - %(lineno)d - %(levelname)s - %(request)s - %(message)s'
{% else %}
format: '%(asctime)s - %(name)s - %(lineno)d - %(levelname)s - %(request)s - %(message)s'
{% endif %}
handlers:
{% if LOG_FILE_PATH %}
@ -45,11 +49,17 @@ handlers:
class: logging.StreamHandler
formatter: precise
{% if not SYNAPSE_LOG_SENSITIVE %}
{#
If SYNAPSE_LOG_SENSITIVE is unset, then override synapse.storage.SQL to INFO
so that DEBUG entries (containing sensitive information) are not emitted.
#}
loggers:
synapse.storage.SQL:
# beware: increasing this to DEBUG will make synapse log sensitive
# information such as access tokens.
level: INFO
{% endif %}
root:
level: {{ SYNAPSE_LOG_LEVEL or "INFO" }}

View File

@ -26,6 +26,13 @@
# * SYNAPSE_TLS_CERT: Path to a TLS certificate in PEM format.
# * SYNAPSE_TLS_KEY: Path to a TLS key. If this and SYNAPSE_TLS_CERT are specified,
# Nginx will be configured to serve TLS on port 8448.
# * SYNAPSE_USE_EXPERIMENTAL_FORKING_LAUNCHER: Whether to use the forking launcher,
# only intended for usage in Complement at the moment.
# No stability guarantees are provided.
# * SYNAPSE_LOG_LEVEL: Set this to DEBUG, INFO, WARNING or ERROR to change the
# log level. INFO is the default.
# * SYNAPSE_LOG_SENSITIVE: If unset, SQL and SQL values won't be logged,
# regardless of the SYNAPSE_LOG_LEVEL setting.
#
# NOTE: According to Complement's ENTRYPOINT expectations for a homeserver image (as defined
# in the project's README), this script may be run multiple times, and functionality should
@ -35,7 +42,7 @@ import os
import subprocess
import sys
from pathlib import Path
from typing import Any, Dict, List, Mapping, MutableMapping, NoReturn, Set
from typing import Any, Dict, List, Mapping, MutableMapping, NoReturn, Optional, Set
import yaml
from jinja2 import Environment, FileSystemLoader
@ -525,6 +532,7 @@ def generate_worker_files(
"/etc/supervisor/conf.d/synapse.conf",
workers=worker_descriptors,
main_config_path=config_path,
use_forking_launcher=environ.get("SYNAPSE_USE_EXPERIMENTAL_FORKING_LAUNCHER"),
)
# healthcheck config
@ -548,18 +556,25 @@ def generate_worker_log_config(
Returns: the path to the generated file
"""
# Check whether we should write worker logs to disk, in addition to the console
extra_log_template_args = {}
extra_log_template_args: Dict[str, Optional[str]] = {}
if environ.get("SYNAPSE_WORKERS_WRITE_LOGS_TO_DISK"):
extra_log_template_args["LOG_FILE_PATH"] = "{dir}/logs/{name}.log".format(
dir=data_dir, name=worker_name
)
extra_log_template_args["LOG_FILE_PATH"] = f"{data_dir}/logs/{worker_name}.log"
extra_log_template_args["SYNAPSE_LOG_LEVEL"] = environ.get("SYNAPSE_LOG_LEVEL")
extra_log_template_args["SYNAPSE_LOG_SENSITIVE"] = environ.get(
"SYNAPSE_LOG_SENSITIVE"
)
# Render and write the file
log_config_filepath = "/conf/workers/{name}.log.config".format(name=worker_name)
log_config_filepath = f"/conf/workers/{worker_name}.log.config"
convert(
"/conf/log.config",
log_config_filepath,
worker_name=worker_name,
**extra_log_template_args,
include_worker_name_in_log_line=environ.get(
"SYNAPSE_USE_EXPERIMENTAL_FORKING_LAUNCHER"
),
)
return log_config_filepath

View File

@ -110,7 +110,11 @@ def generate_config_from_template(
log_config_file = environ["SYNAPSE_LOG_CONFIG"]
log("Generating log config file " + log_config_file)
convert("/conf/log.config", log_config_file, environ)
convert(
"/conf/log.config",
log_config_file,
{**environ, "include_worker_name_in_log_line": False},
)
# Hopefully we already have a signing key, but generate one if not.
args = [

View File

@ -69,6 +69,7 @@
- [Federation](usage/administration/admin_api/federation.md)
- [Manhole](manhole.md)
- [Monitoring](metrics-howto.md)
- [Reporting Anonymised Statistics](usage/administration/monitoring/reporting_anonymised_statistics.md)
- [Understanding Synapse Through Grafana Graphs](usage/administration/understanding_synapse_through_grafana_graphs.md)
- [Useful SQL for Admins](usage/administration/useful_sql_for_admins.md)
- [Database Maintenance Tools](usage/administration/database_maintenance_tools.md)

View File

@ -124,9 +124,8 @@ Body parameters:
- `address` - string. Value of third-party ID.
belonging to a user.
- `external_ids` - array, optional. Allow setting the identifier of the external identity
provider for SSO (Single sign-on). Details in
[Sample Configuration File](../usage/configuration/homeserver_sample_config.html)
section `sso` and `oidc_providers`.
provider for SSO (Single sign-on). Details in the configuration manual under the
sections [sso](../usage/configuration/config_documentation.md#sso) and [oidc_providers](../usage/configuration/config_documentation.md#oidc_providers).
- `auth_provider` - string. ID of the external identity provider. Value of `idp_id`
in the homeserver configuration. Note that no error is raised if the provided
value is not in the homeserver configuration.

View File

@ -70,82 +70,61 @@ on save as they take a while and can be very resource intensive.
- Avoid wildcard imports (`from synapse.types import *`) and
relative imports (`from .types import UserID`).
## Configuration file format
## Configuration code and documentation format
The [sample configuration file](./sample_config.yaml) acts as a
When adding a configuration option to the code, if several settings are grouped into a single dict, ensure that your code
correctly handles the top-level option being set to `None` (as it will be if no sub-options are enabled).
The [configuration manual](usage/configuration/config_documentation.md) acts as a
reference to Synapse's configuration options for server administrators.
Remember that many readers will be unfamiliar with YAML and server
administration in general, so that it is important that the file be as
easy to understand as possible, which includes following a consistent
format.
administration in general, so it is important that when you add
a configuration option the documentation be as easy to understand as possible, which
includes following a consistent format.
Some guidelines follow:
- Sections should be separated with a heading consisting of a single
line prefixed and suffixed with `##`. There should be **two** blank
lines before the section header, and **one** after.
- Each option should be listed in the file with the following format:
- A comment describing the setting. Each line of this comment
should be prefixed with a hash (`#`) and a space.
- Each option should be listed in the config manual with the following format:
- The name of the option, prefixed by `###`.
The comment should describe the default behaviour (ie, what
- A comment which describes the default behaviour (i.e. what
happens if the setting is omitted), as well as what the effect
will be if the setting is changed.
Often, the comment end with something like "uncomment the
following to <do action>".
- A line consisting of only `#`.
- A commented-out example setting, prefixed with only `#`.
- An example setting, using backticks to define the code block
For boolean (on/off) options, convention is that this example
should be the *opposite* to the default (so the comment will end
with "Uncomment the following to enable [or disable]
<feature>." For other options, the example should give some
non-default value which is likely to be useful to the reader.
should be the *opposite* to the default. For other options, the example should give
some non-default value which is likely to be useful to the reader.
- There should be a blank line between each option.
- Where several settings are grouped into a single dict, *avoid* the
convention where the whole block is commented out, resulting in
comment lines starting `# #`, as this is hard to read and confusing
to edit. Instead, leave the top-level config option uncommented, and
follow the conventions above for sub-options. Ensure that your code
correctly handles the top-level option being set to `None` (as it
will be if no sub-options are enabled).
- Lines should be wrapped at 80 characters.
- Use two-space indents.
- `true` and `false` are spelt thus (as opposed to `True`, etc.)
- Use single quotes (`'`) rather than double-quotes (`"`) or backticks
(`` ` ``) to refer to configuration options.
- There should be a horizontal rule between each option, which can be achieved by adding `---` before and
after the option.
- `true` and `false` are spelt thus (as opposed to `True`, etc.)
Example:
---
### `modules`
Use the `module` sub-option to add a module under `modules` to extend functionality.
The `module` setting then has a sub-option, `config`, which can be used to define some configuration
for the `module`.
Defaults to none.
Example configuration:
```yaml
## Frobnication ##
# The frobnicator will ensure that all requests are fully frobnicated.
# To enable it, uncomment the following.
#
#frobnicator_enabled: true
# By default, the frobnicator will frobnicate with the default frobber.
# The following will make it use an alternative frobber.
#
#frobincator_frobber: special_frobber
# Settings for the frobber
#
frobber:
# frobbing speed. Defaults to 1.
#
#speed: 10
# frobbing distance. Defaults to 1000.
#
#distance: 100
modules:
- module: my_super_module.MySuperClass
config:
do_thing: true
- module: my_other_super_module.SomeClass
config: {}
```
---
Note that the sample configuration is generated from the synapse code
and is maintained by a script, `scripts-dev/generate_sample_config.sh`.
Making sure that the output from this script matches the desired format
is left as an exercise for the reader!

View File

@ -309,6 +309,10 @@ The above will run a monolithic (single-process) Synapse with SQLite as the data
- Passing `POSTGRES=1` as an environment variable to use the Postgres database instead.
- Passing `WORKERS=1` as an environment variable to use a workerised setup instead. This option implies the use of Postgres.
To increase the log level for the tests, set `SYNAPSE_TEST_LOG_LEVEL`, e.g:
```sh
SYNAPSE_TEST_LOG_LEVEL=DEBUG COMPLEMENT_DIR=../complement ./scripts-dev/complement.sh -run TestImportHistoricalMessages
```
### Prettier formatting with `gotestfmt`

View File

@ -49,9 +49,8 @@ as follows:
* For other installation mechanisms, see the documentation provided by the
maintainer.
To enable the JSON web token integration, you should then add a `jwt_config` section
to your configuration file (or uncomment the `enabled: true` line in the
existing section). See [sample_config.yaml](./sample_config.yaml) for some
To enable the JSON web token integration, you should then add a `jwt_config` option
to your configuration file. See the [configuration manual](usage/configuration/config_documentation.md#jwt_config) for some
sample settings.
## How to test JWT as a developer

View File

@ -13,8 +13,10 @@ environments where untrusted users have shell access.
## Configuring the manhole
To enable it, first uncomment the `manhole` listener configuration in
`homeserver.yaml`. The configuration is slightly different if you're using docker.
To enable it, first add the `manhole` listener configuration in your
`homeserver.yaml`. You can find information on how to do that
in the [configuration manual](usage/configuration/config_documentation.md#manhole_settings).
The configuration is slightly different if you're using docker.
#### Docker config

View File

@ -49,9 +49,9 @@ clients.
## Server configuration
Support for this feature can be enabled and configured in the
`retention` section of the Synapse configuration file (see the
[sample file](https://github.com/matrix-org/synapse/blob/v1.36.0/docs/sample_config.yaml#L451-L518)).
Support for this feature can be enabled and configured by adding a the
`retention` in the Synapse configuration file (see
[configuration manual](usage/configuration/config_documentation.md#retention)).
To enable support for message retention policies, set the setting
`enabled` in this section to `true`.
@ -65,8 +65,8 @@ message retention policy configured in its state. This allows server
admins to ensure that messages are never kept indefinitely in a server's
database.
A default policy can be defined as such, in the `retention` section of
the configuration file:
A default policy can be defined as such, by adding the `retention` option in
the configuration file and adding these sub-options:
```yaml
default_policy:
@ -86,8 +86,8 @@ Purge jobs are the jobs that Synapse runs in the background to purge
expired events from the database. They are only run if support for
message retention policies is enabled in the server's configuration. If
no configuration for purge jobs is configured by the server admin,
Synapse will use a default configuration, which is described in the
[sample configuration file](https://github.com/matrix-org/synapse/blob/v1.36.0/docs/sample_config.yaml#L451-L518).
Synapse will use a default configuration, which is described here in the
[configuration manual](usage/configuration/config_documentation.md#retention).
Some server admins might want a finer control on when events are removed
depending on an event's room's policy. This can be done by setting the
@ -137,8 +137,8 @@ the server's database.
### Lifetime limits
Server admins can set limits on the values of `max_lifetime` to use when
purging old events in a room. These limits can be defined as such in the
`retention` section of the configuration file:
purging old events in a room. These limits can be defined under the
`retention` option in the configuration file:
```yaml
allowed_lifetime_min: 1d

View File

@ -45,8 +45,8 @@ as follows:
maintainer.
To enable the OpenID integration, you should then add a section to the `oidc_providers`
setting in your configuration file (or uncomment one of the existing examples).
See [sample_config.yaml](./sample_config.yaml) for some sample settings, as well as
setting in your configuration file.
See the [configuration manual](usage/configuration/config_documentation.md#oidc_providers) for some sample settings, as well as
the text below for example configurations for specific providers.
## Sample configs

View File

@ -66,8 +66,8 @@ in Synapse can be deactivated.
**NOTE**: This has an impact on security and is for testing purposes only!
To deactivate the certificate validation, the following setting must be made in
[homserver.yaml](../usage/configuration/homeserver_sample_config.md).
To deactivate the certificate validation, the following setting must be added to
your [homserver.yaml](../usage/configuration/homeserver_sample_config.md).
```yaml
use_insecure_ssl_client_just_for_testing_do_not_use: true

View File

@ -232,7 +232,9 @@ python -m synapse.app.homeserver \
--report-stats=[yes|no]
```
... substituting an appropriate value for `--server-name`.
... substituting an appropriate value for `--server-name` and choosing whether
or not to report usage statistics (hostname, Synapse version, uptime, total
users, etc.) to the developers via the `--report-stats` argument.
This command will generate you a config file that you can then customise, but it will
also generate a set of keys for you. These keys will allow your homeserver to
@ -405,11 +407,11 @@ The recommended way to do so is to set up a reverse proxy on port
Alternatively, you can configure Synapse to expose an HTTPS port. To do
so, you will need to edit `homeserver.yaml`, as follows:
- First, under the `listeners` section, uncomment the configuration for the
TLS-enabled listener. (Remove the hash sign (`#`) at the start of
each line). The relevant lines are like this:
- First, under the `listeners` option, add the configuration for the
TLS-enabled listener like so:
```yaml
listeners:
- port: 8448
type: http
tls: true
@ -417,9 +419,11 @@ so, you will need to edit `homeserver.yaml`, as follows:
- names: [client, federation]
```
- You will also need to uncomment the `tls_certificate_path` and
`tls_private_key_path` lines under the `TLS` section. You will need to manage
provisioning of these certificates yourself.
- You will also need to add the options `tls_certificate_path` and
`tls_private_key_path`. to your configuration file. You will need to manage provisioning of
these certificates yourself.
- You can find more information about these options as well as how to configure synapse in the
[configuration manual](../usage/configuration/config_documentation.md).
If you are using your own certificate, be sure to use a `.pem` file that
includes the full certificate chain including any intermediate certificates

View File

@ -0,0 +1,81 @@
# Reporting Anonymised Statistics
When generating your Synapse configuration file, you are asked whether you
would like to report anonymised statistics to Matrix.org. These statistics
provide the foundation a glimpse into the number of Synapse homeservers
participating in the network, as well as statistics such as the number of
rooms being created and messages being sent. This feature is sometimes
affectionately called "phone-home" stats. Reporting
[is optional](../../configuration/config_documentation.md#report_stats)
and the reporting endpoint
[can be configured](../../configuration/config_documentation.md#report_stats_endpoint),
in case you would like to instead report statistics from a set of homeservers
to your own infrastructure.
This documentation aims to define the statistics available and the
homeserver configuration options that exist to tweak it.
## Available Statistics
The following statistics are sent to the configured reporting endpoint:
| Statistic Name | Type | Description |
|----------------------------|--------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| `memory_rss` | int | The memory usage of the process (in kilobytes on Unix-based systems, bytes on MacOS). |
| `cpu_average` | int | CPU time in % of a single core (not % of all cores). |
| `homeserver` | string | The homeserver's server name. |
| `server_context` | string | An arbitrary string used to group statistics from a set of homeservers. |
| `timestamp` | int | The current time, represented as the number of seconds since the epoch. |
| `uptime_seconds` | int | The number of seconds since the homeserver was last started. |
| `python_version` | string | The Python version number in use (e.g "3.7.1"). Taken from `sys.version_info`. |
| `total_users` | int | The number of registered users on the homeserver. |
| `total_nonbridged_users` | int | The number of users, excluding those created by an Application Service. |
| `daily_user_type_native` | int | The number of native users created in the last 24 hours. |
| `daily_user_type_guest` | int | The number of guest users created in the last 24 hours. |
| `daily_user_type_bridged` | int | The number of users created by Application Services in the last 24 hours. |
| `total_room_count` | int | The total number of rooms present on the homeserver. |
| `daily_active_users` | int | The number of unique users[^1] that have used the homeserver in the last 24 hours. |
| `monthly_active_users` | int | The number of unique users[^1] that have used the homeserver in the last 30 days. |
| `daily_active_rooms` | int | The number of rooms that have had a (state) event with the type `m.room.message` sent in them in the last 24 hours. |
| `daily_active_e2ee_rooms` | int | The number of rooms that have had a (state) event with the type `m.room.encrypted` sent in them in the last 24 hours. |
| `daily_messages` | int | The number of (state) events with the type `m.room.message` seen in the last 24 hours. |
| `daily_e2ee_messages` | int | The number of (state) events with the type `m.room.encrypted` seen in the last 24 hours. |
| `daily_sent_messages` | int | The number of (state) events sent by a local user with the type `m.room.message` seen in the last 24 hours. |
| `daily_sent_e2ee_messages` | int | The number of (state) events sent by a local user with the type `m.room.encrypted` seen in the last 24 hours. |
| `r30_users_all` | int | The number of 30 day retained users, defined as users who have created their accounts more than 30 days ago, where they were last seen at most 30 days ago and where those two timestamps are over 30 days apart. Includes clients that do not fit into the below r30 client types. |
| `r30_users_android` | int | The number of 30 day retained users, as defined above. Filtered only to clients with "Android" in the user agent string. |
| `r30_users_ios` | int | The number of 30 day retained users, as defined above. Filtered only to clients with "iOS" in the user agent string. |
| `r30_users_electron` | int | The number of 30 day retained users, as defined above. Filtered only to clients with "Electron" in the user agent string. |
| `r30_users_web` | int | The number of 30 day retained users, as defined above. Filtered only to clients with "Mozilla" or "Gecko" in the user agent string. |
| `r30v2_users_all` | int | The number of 30 day retained users, with a revised algorithm. Defined as users that appear more than once in the past 60 days, and have more than 30 days between the most and least recent appearances in the past 60 days. Includes clients that do not fit into the below r30 client types. |
| `r30v2_users_android` | int | The number of 30 day retained users, as defined above. Filtered only to clients with ("riot" or "element") and "android" (case-insensitive) in the user agent string. |
| `r30v2_users_ios` | int | The number of 30 day retained users, as defined above. Filtered only to clients with ("riot" or "element") and "ios" (case-insensitive) in the user agent string. |
| `r30v2_users_electron` | int | The number of 30 day retained users, as defined above. Filtered only to clients with ("riot" or "element") and "electron" (case-insensitive) in the user agent string. |
| `r30v2_users_web` | int | The number of 30 day retained users, as defined above. Filtered only to clients with "mozilla" or "gecko" (case-insensitive) in the user agent string. |
| `cache_factor` | int | The configured [`global factor`](../../configuration/config_documentation.md#caching) value for caching. |
| `event_cache_size` | int | The configured [`event_cache_size`](../../configuration/config_documentation.md#caching) value for caching. |
| `database_engine` | string | The database engine that is in use. Either "psycopg2" meaning PostgreSQL is in use, or "sqlite3" for SQLite3. |
| `database_server_version` | string | The version of the database server. Examples being "10.10" for PostgreSQL server version 10.0, and "3.38.5" for SQLite 3.38.5 installed on the system. |
| `log_level` | string | The log level in use. Examples are "INFO", "WARNING", "ERROR", "DEBUG", etc. |
[^1]: Native matrix users and guests are always counted. If the
[`track_puppeted_user_ips`](../../configuration/config_documentation.md#track_puppeted_user_ips)
option is set to `true`, "puppeted" users (users that an Application Service have performed
[an action on behalf of](https://spec.matrix.org/v1.3/application-service-api/#identity-assertion))
will also be counted. Note that an Application Service can "puppet" any user in their
[user namespace](https://spec.matrix.org/v1.3/application-service-api/#registration),
not only users that the Application Service has created. If this happens, the Application Service
will additionally be counted as a user (irrespective of `track_puppeted_user_ips`).
## Using a Custom Statistics Collection Server
If statistics reporting is enabled, the endpoint that Synapse sends metrics to is configured by the
[`report_stats_endpoint`](../../configuration/config_documentation.md#report_stats_endpoint) config
option. By default, statistics are sent to Matrix.org.
If you would like to set up your own statistics collection server and send metrics there, you may
consider using one of the following known implementations:
* [Matrix.org's Panopticon](https://github.com/matrix-org/panopticon)
* [Famedly's Barad-dûr](https://gitlab.com/famedly/company/devops/services/barad-dur)

View File

@ -317,7 +317,7 @@ Example configuration:
allow_public_rooms_without_auth: true
```
---
### `allow_public_rooms_without_auth`
### `allow_public_rooms_over_federation`
If set to true, allows any other homeserver to fetch the server's public
rooms directory via federation. Defaults to false.
@ -591,7 +591,7 @@ Example configuration:
dummy_events_threshold: 5
```
---
Config option `delete_stale_devices_after`
### `delete_stale_devices_after`
An optional duration. If set, Synapse will run a daily background task to log out and
delete any device that hasn't been accessed for more than the specified amount of time.
@ -1843,7 +1843,7 @@ Example configuration:
turn_shared_secret: "YOUR_SHARED_SECRET"
```
----
Config options: `turn_username` and `turn_password`
### `turn_username` and `turn_password`
The Username and password if the TURN server needs them and does not use a token.
@ -2999,7 +2999,7 @@ This setting has the following sub-options:
* `localdb_enabled`: Set to false to disable authentication against the local password
database. This is ignored if `enabled` is false, and is only useful
if you have other `password_providers`. Defaults to true.
* `pepper`: Set the value here to a secret random string for extra security. # Uncomment and change to a secret random string for extra security.
* `pepper`: Set the value here to a secret random string for extra security.
DO NOT CHANGE THIS AFTER INITIAL SETUP!
* `policy`: Define and enforce a password policy, such as minimum lengths for passwords, etc.
Each parameter is optional. This is an implementation of MSC2000. Parameters are as follows:
@ -3373,7 +3373,7 @@ alias_creation_rules:
action: deny
```
---
Config options: `room_list_publication_rules`
### `room_list_publication_rules`
The `room_list_publication_rules` option controls who can publish and
which rooms can be published in the public room list.

View File

@ -9,6 +9,9 @@ a real homeserver.yaml. Instead, if you are starting from scratch, please genera
a fresh config using Synapse by following the instructions in
[Installation](../../setup/installation.md).
Documentation for all configuration options can be found in the
[Configuration Manual](./config_documentation.md).
```yaml
{{#include ../../sample_config.yaml}}
```

View File

@ -4,5 +4,5 @@ Synapse supports authenticating users via the [Central Authentication
Service protocol](https://en.wikipedia.org/wiki/Central_Authentication_Service)
(CAS) natively.
Please see the `cas_config` and `sso` sections of the [Synapse configuration
file](../../../configuration/homeserver_sample_config.md) for more details.
Please see the [cas_config](../../../configuration/config_documentation.md#cas_config) and [sso](../../../configuration/config_documentation.md#sso)
sections of the configuration manual for more details.

View File

@ -56,7 +56,6 @@ exclude = (?x)
|tests/server.py
|tests/server_notices/test_resource_limits_server_notices.py
|tests/test_metrics.py
|tests/test_server.py
|tests/test_state.py
|tests/test_terms_auth.py
|tests/util/caches/test_cached_call.py
@ -74,7 +73,6 @@ exclude = (?x)
|tests/util/test_lrucache.py
|tests/util/test_rwlock.py
|tests/util/test_wheel_timer.py
|tests/utils.py
)$
[mypy-synapse.federation.transport.client]
@ -89,9 +87,6 @@ disallow_untyped_defs = False
[mypy-synapse.logging.opentracing]
disallow_untyped_defs = False
[mypy-synapse.logging.scopecontextmanager]
disallow_untyped_defs = False
[mypy-synapse.metrics._reactor_metrics]
disallow_untyped_defs = False
# This module imports select.epoll. That exists on Linux, but doesn't on macOS.
@ -131,6 +126,9 @@ disallow_untyped_defs = True
[mypy-tests.federation.transport.test_client]
disallow_untyped_defs = True
[mypy-tests.utils]
disallow_untyped_defs = True
;; Dependencies without annotations
;; Before ignoring a module, check to see if type stubs are available.

View File

@ -54,7 +54,7 @@ skip_gitignore = true
[tool.poetry]
name = "matrix-synapse"
version = "1.62.0rc3"
version = "1.62.0"
description = "Homeserver for the Matrix decentralised comms protocol"
authors = ["Matrix.org Team and Contributors <packages@matrix.org>"]
license = "Apache-2.0"

View File

@ -14,12 +14,18 @@
# By default Synapse is run in monolith mode. This can be overridden by
# setting the WORKERS environment variable.
#
# A regular expression of test method names can be supplied as the first
# argument to the script. Complement will then only run those tests. If
# no regex is supplied, all tests are run. For example;
# You can optionally give a "-f" argument (for "fast") before any to skip
# rebuilding the docker images, if you just want to rerun the tests.
#
# Remaining commandline arguments are passed through to `go test`. For example,
# you can supply a regular expression of test method names via the "-run"
# argument:
#
# ./complement.sh -run "TestOutboundFederation(Profile|Send)"
#
# Specifying TEST_ONLY_SKIP_DEP_HASH_VERIFICATION=1 will cause `poetry export`
# to not emit any hashes when building the Docker image. This then means that
# you can use 'unverifiable' sources such as git repositories as dependencies.
# Exit if a line returns a non-zero exit code
set -e
@ -32,6 +38,37 @@ echo_if_github() {
fi
}
# Helper to print out the usage instructions
usage() {
cat >&2 <<EOF
Usage: $0 [-f] <go test arguments>...
Run the complement test suite on Synapse.
-f Skip rebuilding the docker images, and just use the most recent
'complement-synapse:latest' image
For help on arguments to 'go test', run 'go help testflag'.
EOF
}
# parse our arguments
skip_docker_build=""
while [ $# -ge 1 ]; do
arg=$1
case "$arg" in
"-h")
usage
exit 1
;;
"-f")
skip_docker_build=1
;;
*)
# unknown arg: presumably an argument to gotest. break the loop.
break
esac
shift
done
# enable buildkit for the docker builds
export DOCKER_BUILDKIT=1
@ -49,21 +86,25 @@ if [[ -z "$COMPLEMENT_DIR" ]]; then
echo "Checkout available at 'complement-${COMPLEMENT_REF}'"
fi
# Build the base Synapse image from the local checkout
echo_if_github "::group::Build Docker image: matrixdotorg/synapse"
docker build -t matrixdotorg/synapse -f "docker/Dockerfile" .
echo_if_github "::endgroup::"
if [ -z "$skip_docker_build" ]; then
# Build the base Synapse image from the local checkout
echo_if_github "::group::Build Docker image: matrixdotorg/synapse"
docker build -t matrixdotorg/synapse \
--build-arg TEST_ONLY_SKIP_DEP_HASH_VERIFICATION \
-f "docker/Dockerfile" .
echo_if_github "::endgroup::"
# Build the workers docker image (from the base Synapse image we just built).
echo_if_github "::group::Build Docker image: matrixdotorg/synapse-workers"
docker build -t matrixdotorg/synapse-workers -f "docker/Dockerfile-workers" .
echo_if_github "::endgroup::"
# Build the workers docker image (from the base Synapse image we just built).
echo_if_github "::group::Build Docker image: matrixdotorg/synapse-workers"
docker build -t matrixdotorg/synapse-workers -f "docker/Dockerfile-workers" .
echo_if_github "::endgroup::"
# Build the unified Complement image (from the worker Synapse image we just built).
echo_if_github "::group::Build Docker image: complement/Dockerfile"
docker build -t complement-synapse \
-f "docker/complement/Dockerfile" "docker/complement"
echo_if_github "::endgroup::"
# Build the unified Complement image (from the worker Synapse image we just built).
echo_if_github "::group::Build Docker image: complement/Dockerfile"
docker build -t complement-synapse \
-f "docker/complement/Dockerfile" "docker/complement"
echo_if_github "::endgroup::"
fi
export COMPLEMENT_BASE_IMAGE=complement-synapse
@ -104,6 +145,18 @@ else
test_tags="$test_tags,faster_joins"
fi
if [[ -n "$SYNAPSE_TEST_LOG_LEVEL" ]]; then
# Set the log level to what is desired
export PASS_SYNAPSE_LOG_LEVEL="$SYNAPSE_TEST_LOG_LEVEL"
# Allow logging sensitive things (currently SQL queries & parameters).
# (This won't have any effect if we're not logging at DEBUG level overall.)
# Since this is just a test suite, this is fine and won't reveal anyone's
# personal information
export PASS_SYNAPSE_LOG_SENSITIVE=1
fi
# Run the tests!
echo "Images built; running complement"
cd "$COMPLEMENT_DIR"

View File

@ -268,6 +268,9 @@ class MockHomeserver:
def get_instance_name(self) -> str:
return "master"
def should_send_federation(self) -> bool:
return False
class Porter:
def __init__(

View File

@ -259,3 +259,13 @@ class ReceiptTypes:
READ: Final = "m.read"
READ_PRIVATE: Final = "org.matrix.msc2285.read.private"
FULLY_READ: Final = "m.fully_read"
class PublicRoomsFilterFields:
"""Fields in the search filter for `/publicRooms` that we understand.
As defined in https://spec.matrix.org/v1.3/client-server-api/#post_matrixclientv3publicrooms
"""
GENERIC_SEARCH_TERM: Final = "generic_search_term"
ROOM_TYPES: Final = "org.matrix.msc3827.room_types"

View File

@ -106,7 +106,9 @@ def register_sighup(func: Callable[P, None], *args: P.args, **kwargs: P.kwargs)
def start_worker_reactor(
appname: str,
config: HomeServerConfig,
run_command: Callable[[], None] = reactor.run,
# Use a lambda to avoid binding to a given reactor at import time.
# (needed when synapse.app.complement_fork_starter is being used)
run_command: Callable[[], None] = lambda: reactor.run(),
) -> None:
"""Run the reactor in the main process
@ -141,7 +143,9 @@ def start_reactor(
daemonize: bool,
print_pidfile: bool,
logger: logging.Logger,
run_command: Callable[[], None] = reactor.run,
# Use a lambda to avoid binding to a given reactor at import time.
# (needed when synapse.app.complement_fork_starter is being used)
run_command: Callable[[], None] = lambda: reactor.run(),
) -> None:
"""Run the reactor in the main process

View File

@ -0,0 +1,190 @@
# Copyright 2022 The Matrix.org Foundation C.I.C.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
# ## What this script does
#
# This script spawns multiple workers, whilst only going through the code loading
# process once. The net effect is that start-up time for a swarm of workers is
# reduced, particularly in CPU-constrained environments.
#
# Before the workers are spawned, the database is prepared in order to avoid the
# workers racing.
#
# ## Stability
#
# This script is only intended for use within the Synapse images for the
# Complement test suite.
# There are currently no stability guarantees whatsoever; especially not about:
# - whether it will continue to exist in future versions;
# - the format of its command-line arguments; or
# - any details about its behaviour or principles of operation.
#
# ## Usage
#
# The first argument should be the path to the database configuration, used to
# set up the database. The rest of the arguments are used as follows:
# Each worker is specified as an argument group (each argument group is
# separated by '--').
# The first argument in each argument group is the Python module name of the application
# to start. Further arguments are then passed to that module as-is.
#
# ## Example
#
# python -m synapse.app.complement_fork_starter path_to_db_config.yaml \
# synapse.app.homeserver [args..] -- \
# synapse.app.generic_worker [args..] -- \
# ...
# synapse.app.generic_worker [args..]
#
import argparse
import importlib
import itertools
import multiprocessing
import sys
from typing import Any, Callable, List
from twisted.internet.main import installReactor
class ProxiedReactor:
"""
Twisted tracks the 'installed' reactor as a global variable.
(Actually, it does some module trickery, but the effect is similar.)
The default EpollReactor is buggy if it's created before a process is
forked, then used in the child.
See https://twistedmatrix.com/trac/ticket/4759#comment:17.
However, importing certain Twisted modules will automatically create and
install a reactor if one hasn't already been installed.
It's not normally possible to re-install a reactor.
Given the goal of launching workers with fork() to only import the code once,
this presents a conflict.
Our work around is to 'install' this ProxiedReactor which prevents Twisted
from creating and installing one, but which lets us replace the actual reactor
in use later on.
"""
def __init__(self) -> None:
self.___reactor_target: Any = None
def _install_real_reactor(self, new_reactor: Any) -> None:
"""
Install a real reactor for this ProxiedReactor to forward lookups onto.
This method is specific to our ProxiedReactor and should not clash with
any names used on an actual Twisted reactor.
"""
self.___reactor_target = new_reactor
def __getattr__(self, attr_name: str) -> Any:
return getattr(self.___reactor_target, attr_name)
def _worker_entrypoint(
func: Callable[[], None], proxy_reactor: ProxiedReactor, args: List[str]
) -> None:
"""
Entrypoint for a forked worker process.
We just need to set up the command-line arguments, create our real reactor
and then kick off the worker's main() function.
"""
sys.argv = args
from twisted.internet.epollreactor import EPollReactor
proxy_reactor._install_real_reactor(EPollReactor())
func()
def main() -> None:
"""
Entrypoint for the forking launcher.
"""
parser = argparse.ArgumentParser()
parser.add_argument("db_config", help="Path to database config file")
parser.add_argument(
"args",
nargs="...",
help="Argument groups separated by `--`. "
"The first argument of each group is a Synapse app name. "
"Subsequent arguments are passed through.",
)
ns = parser.parse_args()
# Split up the subsequent arguments into each workers' arguments;
# `--` is our delimiter of choice.
args_by_worker: List[List[str]] = [
list(args)
for cond, args in itertools.groupby(ns.args, lambda ele: ele != "--")
if cond and args
]
# Prevent Twisted from installing a shared reactor that all the workers will
# inherit when we fork(), by installing our own beforehand.
proxy_reactor = ProxiedReactor()
installReactor(proxy_reactor)
# Import the entrypoints for all the workers.
worker_functions = []
for worker_args in args_by_worker:
worker_module = importlib.import_module(worker_args[0])
worker_functions.append(worker_module.main)
# We need to prepare the database first as otherwise all the workers will
# try to create a schema version table and some will crash out.
from synapse._scripts import update_synapse_database
update_proc = multiprocessing.Process(
target=_worker_entrypoint,
args=(
update_synapse_database.main,
proxy_reactor,
[
"update_synapse_database",
"--database-config",
ns.db_config,
"--run-background-updates",
],
),
)
print("===== PREPARING DATABASE =====", file=sys.stderr)
update_proc.start()
update_proc.join()
print("===== PREPARED DATABASE =====", file=sys.stderr)
# At this point, we've imported all the main entrypoints for all the workers.
# Now we basically just fork() out to create the workers we need.
# Because we're using fork(), all the workers get a clone of this launcher's
# memory space and don't need to repeat the work of loading the code!
# Instead of using fork() directly, we use the multiprocessing library,
# which uses fork() on Unix platforms.
processes = []
for (func, worker_args) in zip(worker_functions, args_by_worker):
process = multiprocessing.Process(
target=_worker_entrypoint, args=(func, proxy_reactor, worker_args)
)
process.start()
processes.append(process)
# Be a good parent and wait for our children to die before exiting.
for process in processes:
process.join()
if __name__ == "__main__":
main()

View File

@ -21,7 +21,7 @@ from typing import Any, Callable, Dict, Optional
import attr
from synapse.types import JsonDict
from synapse.util.check_dependencies import DependencyException, check_requirements
from synapse.util.check_dependencies import check_requirements
from ._base import Config, ConfigError
@ -159,12 +159,7 @@ class CacheConfig(Config):
self.track_memory_usage = cache_config.get("track_memory_usage", False)
if self.track_memory_usage:
try:
check_requirements("cache_memory")
except DependencyException as e:
raise ConfigError(
e.message # noqa: B306, DependencyException.message is a property
)
check_requirements("cache_memory")
expire_caches = cache_config.get("expire_caches", True)
cache_entry_ttl = cache_config.get("cache_entry_ttl", "30m")

View File

@ -145,7 +145,7 @@ class EmailConfig(Config):
raise ConfigError(
'The config option "trust_identity_server_for_password_resets" '
'has been replaced by "account_threepid_delegate". '
"Please consult the sample config at docs/sample_config.yaml for "
"Please consult the configuration manual at docs/usage/configuration/config_documentation.md for "
"details and update your config file."
)

View File

@ -87,3 +87,6 @@ class ExperimentalConfig(Config):
# MSC3715: dir param on /relations.
self.msc3715_enabled: bool = experimental.get("msc3715_enabled", False)
# MSC3827: Filtering of /publicRooms by room type
self.msc3827_enabled: bool = experimental.get("msc3827_enabled", False)

View File

@ -15,14 +15,9 @@
from typing import Any
from synapse.types import JsonDict
from synapse.util.check_dependencies import check_requirements
from ._base import Config, ConfigError
MISSING_AUTHLIB = """Missing authlib library. This is required for jwt login.
Install by running:
pip install synapse[jwt]
"""
from ._base import Config
class JWTConfig(Config):
@ -41,13 +36,7 @@ class JWTConfig(Config):
# that the claims exist on the JWT.
self.jwt_issuer = jwt_config.get("issuer")
self.jwt_audiences = jwt_config.get("audiences")
try:
from authlib.jose import JsonWebToken
JsonWebToken # To stop unused lint.
except ImportError:
raise ConfigError(MISSING_AUTHLIB)
check_requirements("jwt")
else:
self.jwt_enabled = False
self.jwt_secret = None

View File

@ -18,7 +18,7 @@ from typing import Any, Optional
import attr
from synapse.types import JsonDict
from synapse.util.check_dependencies import DependencyException, check_requirements
from synapse.util.check_dependencies import check_requirements
from ._base import Config, ConfigError
@ -57,12 +57,7 @@ class MetricsConfig(Config):
self.sentry_enabled = "sentry" in config
if self.sentry_enabled:
try:
check_requirements("sentry")
except DependencyException as e:
raise ConfigError(
e.message # noqa: B306, DependencyException.message is a property
)
check_requirements("sentry")
self.sentry_dsn = config["sentry"].get("dsn")
if not self.sentry_dsn:

View File

@ -24,7 +24,7 @@ from synapse.types import JsonDict
from synapse.util.module_loader import load_module
from synapse.util.stringutils import parse_and_validate_mxc_uri
from ..util.check_dependencies import DependencyException, check_requirements
from ..util.check_dependencies import check_requirements
from ._base import Config, ConfigError, read_file
DEFAULT_USER_MAPPING_PROVIDER = "synapse.handlers.oidc.JinjaOidcMappingProvider"
@ -41,12 +41,7 @@ class OIDCConfig(Config):
if not self.oidc_providers:
return
try:
check_requirements("oidc")
except DependencyException as e:
raise ConfigError(
e.message # noqa: B306, DependencyException.message is a property
) from e
check_requirements("oidc")
# check we don't have any duplicate idp_ids now. (The SSO handler will also
# check for duplicates when the REST listeners get registered, but that happens
@ -146,7 +141,6 @@ OIDC_PROVIDER_CONFIG_WITH_ID_SCHEMA = {
"allOf": [OIDC_PROVIDER_CONFIG_SCHEMA, {"required": ["idp_id", "idp_name"]}]
}
# the `oidc_providers` list can either be None (as it is in the default config), or
# a list of provider configs, each of which requires an explicit ID and name.
OIDC_PROVIDER_LIST_SCHEMA = {

View File

@ -136,6 +136,11 @@ class RatelimitConfig(Config):
defaults={"per_second": 0.003, "burst_count": 5},
)
self.rc_invites_per_issuer = RateLimitConfig(
config.get("rc_invites", {}).get("per_issuer", {}),
defaults={"per_second": 0.3, "burst_count": 10},
)
self.rc_third_party_invite = RateLimitConfig(
config.get("rc_third_party_invite", {}),
defaults={

View File

@ -21,7 +21,7 @@ import attr
from synapse.config.server import generate_ip_set
from synapse.types import JsonDict
from synapse.util.check_dependencies import DependencyException, check_requirements
from synapse.util.check_dependencies import check_requirements
from synapse.util.module_loader import load_module
from ._base import Config, ConfigError
@ -184,13 +184,7 @@ class ContentRepositoryConfig(Config):
)
self.url_preview_enabled = config.get("url_preview_enabled", False)
if self.url_preview_enabled:
try:
check_requirements("url_preview")
except DependencyException as e:
raise ConfigError(
e.message # noqa: B306, DependencyException.message is a property
)
check_requirements("url_preview")
proxy_env = getproxies_environment()
if "url_preview_ip_range_blacklist" not in config:

View File

@ -18,7 +18,7 @@ from typing import Any, List, Set
from synapse.config.sso import SsoAttributeRequirement
from synapse.types import JsonDict
from synapse.util.check_dependencies import DependencyException, check_requirements
from synapse.util.check_dependencies import check_requirements
from synapse.util.module_loader import load_module, load_python_module
from ._base import Config, ConfigError
@ -76,12 +76,7 @@ class SAML2Config(Config):
if not saml2_config.get("sp_config") and not saml2_config.get("config_path"):
return
try:
check_requirements("saml2")
except DependencyException as e:
raise ConfigError(
e.message # noqa: B306, DependencyException.message is a property
)
check_requirements("saml2")
self.saml2_enabled = True

View File

@ -15,7 +15,7 @@
from typing import Any, List, Set
from synapse.types import JsonDict
from synapse.util.check_dependencies import DependencyException, check_requirements
from synapse.util.check_dependencies import check_requirements
from ._base import Config, ConfigError
@ -40,12 +40,7 @@ class TracerConfig(Config):
if not self.opentracer_enabled:
return
try:
check_requirements("opentracing")
except DependencyException as e:
raise ConfigError(
e.message # noqa: B306, DependencyException.message is a property
)
check_requirements("opentracing")
# The tracer is enabled so sanitize the config

View File

@ -67,6 +67,7 @@ from synapse.replication.http.federation import (
ReplicationFederationSendEduRestServlet,
ReplicationGetQueryRestServlet,
)
from synapse.storage.databases.main.events import PartialStateConflictError
from synapse.storage.databases.main.lock import Lock
from synapse.types import JsonDict, StateMap, get_domain_from_id
from synapse.util import json_decoder, unwrapFirstError
@ -882,9 +883,20 @@ class FederationServer(FederationBase):
logger.warning("%s", errmsg)
raise SynapseError(403, errmsg, Codes.FORBIDDEN)
return await self._federation_event_handler.on_send_membership_event(
origin, event
)
try:
return await self._federation_event_handler.on_send_membership_event(
origin, event
)
except PartialStateConflictError:
# The room was un-partial stated while we were persisting the event.
# Try once more, with full state this time.
logger.info(
"Room %s was un-partial stated during `on_send_membership_event`, trying again.",
room_id,
)
return await self._federation_event_handler.on_send_membership_event(
origin, event
)
async def on_event_auth(
self, origin: str, room_id: str, event_id: str

View File

@ -45,6 +45,7 @@ from synapse.api.errors import (
FederationDeniedError,
FederationError,
HttpResponseException,
LimitExceededError,
NotFoundError,
RequestSendFailed,
SynapseError,
@ -64,6 +65,7 @@ from synapse.replication.http.federation import (
ReplicationCleanRoomRestServlet,
ReplicationStoreRoomOnOutlierMembershipRestServlet,
)
from synapse.storage.databases.main.events import PartialStateConflictError
from synapse.storage.databases.main.events_worker import EventRedactBehaviour
from synapse.storage.state import StateFilter
from synapse.types import JsonDict, StateMap, get_domain_from_id
@ -549,15 +551,29 @@ class FederationHandler:
# https://github.com/matrix-org/synapse/issues/12998
await self.store.store_partial_state_room(room_id, ret.servers_in_room)
max_stream_id = await self._federation_event_handler.process_remote_join(
origin,
room_id,
auth_chain,
state,
event,
room_version_obj,
partial_state=ret.partial_state,
)
try:
max_stream_id = (
await self._federation_event_handler.process_remote_join(
origin,
room_id,
auth_chain,
state,
event,
room_version_obj,
partial_state=ret.partial_state,
)
)
except PartialStateConflictError as e:
# The homeserver was already in the room and it is no longer partial
# stated. We ought to be doing a local join instead. Turn the error into
# a 429, as a hint to the client to try again.
# TODO(faster_joins): `_should_perform_remote_join` suggests that we may
# do a remote join for restricted rooms even if we have full state.
logger.error(
"Room %s was un-partial stated while processing remote join.",
room_id,
)
raise LimitExceededError(msg=e.msg, errcode=e.errcode, retry_after_ms=0)
if ret.partial_state:
# Kick off the process of asynchronously fetching the state for this
@ -1567,11 +1583,6 @@ class FederationHandler:
# we raced against more events arriving with partial state. Go round
# the loop again. We've already logged a warning, so no need for more.
# TODO(faster_joins): there is still a race here, whereby incoming events which raced
# with us will fail to be persisted after the call to `clear_partial_state_room` due to
# having partial state.
# https://github.com/matrix-org/synapse/issues/12988
#
continue
events = await self.store.get_events_as_list(

View File

@ -64,6 +64,7 @@ from synapse.replication.http.federation import (
ReplicationFederationSendEventsRestServlet,
)
from synapse.state import StateResolutionStore
from synapse.storage.databases.main.events import PartialStateConflictError
from synapse.storage.databases.main.events_worker import EventRedactBehaviour
from synapse.storage.state import StateFilter
from synapse.types import (
@ -275,7 +276,16 @@ class FederationEventHandler:
affected=pdu.event_id,
)
await self._process_received_pdu(origin, pdu, state_ids=None)
try:
await self._process_received_pdu(origin, pdu, state_ids=None)
except PartialStateConflictError:
# The room was un-partial stated while we were processing the PDU.
# Try once more, with full state this time.
logger.info(
"Room %s was un-partial stated while processing the PDU, trying again.",
room_id,
)
await self._process_received_pdu(origin, pdu, state_ids=None)
async def on_send_membership_event(
self, origin: str, event: EventBase
@ -306,6 +316,9 @@ class FederationEventHandler:
Raises:
SynapseError if the event is not accepted into the room
PartialStateConflictError if the room was un-partial stated in between
computing the state at the event and persisting it. The caller should
retry exactly once in this case.
"""
logger.debug(
"on_send_membership_event: Got event: %s, signatures: %s",
@ -423,6 +436,8 @@ class FederationEventHandler:
Raises:
SynapseError if the response is in some way invalid.
PartialStateConflictError if the homeserver is already in the room and it
has been un-partial stated.
"""
create_event = None
for e in state:
@ -1084,28 +1099,31 @@ class FederationEventHandler:
state_ids: Normally None, but if we are handling a gap in the graph
(ie, we are missing one or more prev_events), the resolved state at the
event
event. Must not be partial state.
backfilled: True if this is part of a historical batch of events (inhibits
notification to clients, and validation of device keys.)
PartialStateConflictError: if the room was un-partial stated in between
computing the state at the event and persisting it. The caller should retry
exactly once in this case. Will never be raised if `state_ids` is provided.
"""
logger.debug("Processing event: %s", event)
assert not event.internal_metadata.outlier
context = await self._state_handler.compute_event_context(
event,
state_ids_before_event=state_ids,
)
try:
context = await self._state_handler.compute_event_context(
event,
state_ids_before_event=state_ids,
)
context = await self._check_event_auth(
origin,
event,
context,
)
except AuthError as e:
# FIXME richvdh 2021/10/07 I don't think this is reachable. Let's log it
# for now
logger.exception("Unexpected AuthError from _check_event_auth")
# This happens only if we couldn't find the auth events. We'll already have
# logged a warning, so now we just convert to a FederationError.
raise FederationError("ERROR", e.code, e.msg, affected=event.event_id)
if not backfilled and not context.rejected:
@ -1934,6 +1952,9 @@ class FederationEventHandler:
event: The event itself.
context: The event context.
backfilled: True if the event was backfilled.
PartialStateConflictError: if attempting to persist a partial state event in
a room that has been un-partial stated.
"""
# this method should not be called on outliers (those code paths call
# persist_events_and_notify directly.)
@ -1986,6 +2007,10 @@ class FederationEventHandler:
Returns:
The stream ID after which all events have been persisted.
Raises:
PartialStateConflictError: if attempting to persist a partial state event in
a room that has been un-partial stated.
"""
if not event_and_contexts:
return self._store.get_room_max_stream_ordering()
@ -1994,14 +2019,19 @@ class FederationEventHandler:
if instance != self._instance_name:
# Limit the number of events sent over replication. We choose 200
# here as that is what we default to in `max_request_body_size(..)`
for batch in batch_iter(event_and_contexts, 200):
result = await self._send_events(
instance_name=instance,
store=self._store,
room_id=room_id,
event_and_contexts=batch,
backfilled=backfilled,
)
try:
for batch in batch_iter(event_and_contexts, 200):
result = await self._send_events(
instance_name=instance,
store=self._store,
room_id=room_id,
event_and_contexts=batch,
backfilled=backfilled,
)
except SynapseError as e:
if e.code == HTTPStatus.CONFLICT:
raise PartialStateConflictError()
raise
return result["max_stream_id"]
else:
assert self._storage_controllers.persistence

View File

@ -37,6 +37,7 @@ from synapse.api.errors import (
AuthError,
Codes,
ConsentNotGivenError,
LimitExceededError,
NotFoundError,
ShadowBanError,
SynapseError,
@ -53,6 +54,7 @@ from synapse.handlers.directory import DirectoryHandler
from synapse.logging.context import make_deferred_yieldable, run_in_background
from synapse.metrics.background_process_metrics import run_as_background_process
from synapse.replication.http.send_event import ReplicationSendEventRestServlet
from synapse.storage.databases.main.events import PartialStateConflictError
from synapse.storage.databases.main.events_worker import EventRedactBehaviour
from synapse.storage.state import StateFilter
from synapse.types import (
@ -903,6 +905,9 @@ class EventCreationHandler:
await self.clock.sleep(random.randint(1, 10))
raise ShadowBanError()
if ratelimit:
await self.request_ratelimiter.ratelimit(requester, update=False)
# We limit the number of concurrent event sends in a room so that we
# don't fork the DAG too much. If we don't limit then we can end up in
# a situation where event persistence can't keep up, causing
@ -1247,6 +1252,8 @@ class EventCreationHandler:
Raises:
ShadowBanError if the requester has been shadow-banned.
SynapseError(503) if attempting to persist a partial state event in
a room that has been un-partial stated.
"""
extra_users = extra_users or []
@ -1297,24 +1304,35 @@ class EventCreationHandler:
# We now persist the event (and update the cache in parallel, since we
# don't want to block on it).
result, _ = await make_deferred_yieldable(
gather_results(
(
run_in_background(
self._persist_event,
requester=requester,
event=event,
context=context,
ratelimit=ratelimit,
extra_users=extra_users,
try:
result, _ = await make_deferred_yieldable(
gather_results(
(
run_in_background(
self._persist_event,
requester=requester,
event=event,
context=context,
ratelimit=ratelimit,
extra_users=extra_users,
),
run_in_background(
self.cache_joined_hosts_for_event, event, context
).addErrback(
log_failure, "cache_joined_hosts_for_event failed"
),
),
run_in_background(
self.cache_joined_hosts_for_event, event, context
).addErrback(log_failure, "cache_joined_hosts_for_event failed"),
),
consumeErrors=True,
consumeErrors=True,
)
).addErrback(unwrapFirstError)
except PartialStateConflictError as e:
# The event context needs to be recomputed.
# Turn the error into a 429, as a hint to the client to try again.
logger.info(
"Room %s was un-partial stated while persisting client event.",
event.room_id,
)
).addErrback(unwrapFirstError)
raise LimitExceededError(msg=e.msg, errcode=e.errcode, retry_after_ms=0)
return result
@ -1329,6 +1347,9 @@ class EventCreationHandler:
"""Actually persists the event. Should only be called by
`handle_new_client_event`, and see its docstring for documentation of
the arguments.
PartialStateConflictError: if attempting to persist a partial state event in
a room that has been un-partial stated.
"""
# Skip push notification actions for historical messages
@ -1345,16 +1366,21 @@ class EventCreationHandler:
# If we're a worker we need to hit out to the master.
writer_instance = self._events_shard_config.get_instance(event.room_id)
if writer_instance != self._instance_name:
result = await self.send_event(
instance_name=writer_instance,
event_id=event.event_id,
store=self.store,
requester=requester,
event=event,
context=context,
ratelimit=ratelimit,
extra_users=extra_users,
)
try:
result = await self.send_event(
instance_name=writer_instance,
event_id=event.event_id,
store=self.store,
requester=requester,
event=event,
context=context,
ratelimit=ratelimit,
extra_users=extra_users,
)
except SynapseError as e:
if e.code == HTTPStatus.CONFLICT:
raise PartialStateConflictError()
raise
stream_id = result["stream_id"]
event_id = result["event_id"]
if event_id != event.event_id:
@ -1482,6 +1508,10 @@ class EventCreationHandler:
The persisted event. This may be different than the given event if
it was de-duplicated (e.g. because we had already persisted an
event with the same transaction ID.)
Raises:
PartialStateConflictError: if attempting to persist a partial state event in
a room that has been un-partial stated.
"""
extra_users = extra_users or []

View File

@ -25,6 +25,7 @@ from synapse.api.constants import (
GuestAccess,
HistoryVisibility,
JoinRules,
PublicRoomsFilterFields,
)
from synapse.api.errors import (
Codes,
@ -181,6 +182,7 @@ class RoomListHandler:
== HistoryVisibility.WORLD_READABLE,
"guest_can_join": room["guest_access"] == "can_join",
"join_rule": room["join_rules"],
"org.matrix.msc3827.room_type": room["room_type"],
}
# Filter out Nones rather omit the field altogether
@ -239,7 +241,9 @@ class RoomListHandler:
response["chunk"] = results
response["total_room_count_estimate"] = await self.store.count_public_rooms(
network_tuple, ignore_non_federatable=from_federation
network_tuple,
ignore_non_federatable=from_federation,
search_filter=search_filter,
)
return response
@ -508,8 +512,21 @@ class RoomListNextBatch:
def _matches_room_entry(room_entry: JsonDict, search_filter: dict) -> bool:
if search_filter and search_filter.get("generic_search_term", None):
generic_search_term = search_filter["generic_search_term"].upper()
"""Determines whether the given search filter matches a room entry returned over
federation.
Only used if the remote server does not support MSC2197 remote-filtered search, and
hence does not support MSC3827 filtering of `/publicRooms` by room type either.
In this case, we cannot apply the `room_type` filter since no `room_type` field is
returned.
"""
if search_filter and search_filter.get(
PublicRoomsFilterFields.GENERIC_SEARCH_TERM, None
):
generic_search_term = search_filter[
PublicRoomsFilterFields.GENERIC_SEARCH_TERM
].upper()
if generic_search_term in room_entry.get("name", "").upper():
return True
elif generic_search_term in room_entry.get("topic", "").upper():

View File

@ -101,19 +101,33 @@ class RoomMemberHandler(metaclass=abc.ABCMeta):
burst_count=hs.config.ratelimiting.rc_joins_remote.burst_count,
)
# Ratelimiter for invites, keyed by room (across all issuers, all
# recipients).
self._invites_per_room_limiter = Ratelimiter(
store=self.store,
clock=self.clock,
rate_hz=hs.config.ratelimiting.rc_invites_per_room.per_second,
burst_count=hs.config.ratelimiting.rc_invites_per_room.burst_count,
)
self._invites_per_user_limiter = Ratelimiter(
# Ratelimiter for invites, keyed by recipient (across all rooms, all
# issuers).
self._invites_per_recipient_limiter = Ratelimiter(
store=self.store,
clock=self.clock,
rate_hz=hs.config.ratelimiting.rc_invites_per_user.per_second,
burst_count=hs.config.ratelimiting.rc_invites_per_user.burst_count,
)
# Ratelimiter for invites, keyed by issuer (across all rooms, all
# recipients).
self._invites_per_issuer_limiter = Ratelimiter(
store=self.store,
clock=self.clock,
rate_hz=hs.config.ratelimiting.rc_invites_per_issuer.per_second,
burst_count=hs.config.ratelimiting.rc_invites_per_issuer.burst_count,
)
self._third_party_invite_limiter = Ratelimiter(
store=self.store,
clock=self.clock,
@ -258,7 +272,9 @@ class RoomMemberHandler(metaclass=abc.ABCMeta):
if room_id:
await self._invites_per_room_limiter.ratelimit(requester, room_id)
await self._invites_per_user_limiter.ratelimit(requester, invitee_user_id)
await self._invites_per_recipient_limiter.ratelimit(requester, invitee_user_id)
if requester is not None:
await self._invites_per_issuer_limiter.ratelimit(requester)
async def _local_membership_update(
self,
@ -844,10 +860,17 @@ class RoomMemberHandler(metaclass=abc.ABCMeta):
content["membership"] = Membership.JOIN
profile = self.profile_handler
if not content_specified:
content["displayname"] = await profile.get_displayname(target)
content["avatar_url"] = await profile.get_avatar_url(target)
try:
profile = self.profile_handler
if not content_specified:
content["displayname"] = await profile.get_displayname(target)
content["avatar_url"] = await profile.get_avatar_url(target)
except Exception as e:
logger.info(
"Failed to get profile information while processing remote join for %r: %s",
target,
e,
)
if requester.is_guest:
content["kind"] = "guest"
@ -924,11 +947,18 @@ class RoomMemberHandler(metaclass=abc.ABCMeta):
content["membership"] = Membership.KNOCK
profile = self.profile_handler
if "displayname" not in content:
content["displayname"] = await profile.get_displayname(target)
if "avatar_url" not in content:
content["avatar_url"] = await profile.get_avatar_url(target)
try:
profile = self.profile_handler
if "displayname" not in content:
content["displayname"] = await profile.get_displayname(target)
if "avatar_url" not in content:
content["avatar_url"] = await profile.get_avatar_url(target)
except Exception as e:
logger.info(
"Failed to get profile information while processing remote knock for %r: %s",
target,
e,
)
return await self.remote_knock(
remote_room_hosts, room_id, target, content

View File

@ -271,6 +271,9 @@ class StatsHandler:
room_state["is_federatable"] = (
event_content.get(EventContentFields.FEDERATE, True) is True
)
room_type = event_content.get(EventContentFields.ROOM_TYPE)
if isinstance(room_type, str):
room_state["room_type"] = room_type
elif typ == EventTypes.JoinRules:
room_state["join_rules"] = event_content.get("join_rule")
elif typ == EventTypes.RoomHistoryVisibility:

View File

@ -164,6 +164,7 @@ Gotchas
with an active span?
"""
import contextlib
import enum
import inspect
import logging
import re
@ -268,7 +269,7 @@ try:
_reporter: Reporter = attr.Factory(Reporter)
def set_process(self, *args, **kwargs):
def set_process(self, *args: Any, **kwargs: Any) -> None:
return self._reporter.set_process(*args, **kwargs)
def report_span(self, span: "opentracing.Span") -> None:
@ -319,7 +320,11 @@ _homeserver_whitelist: Optional[Pattern[str]] = None
# Util methods
Sentinel = object()
class _Sentinel(enum.Enum):
# defining a sentinel in this way allows mypy to correctly handle the
# type of a dictionary lookup.
sentinel = object()
P = ParamSpec("P")
@ -339,12 +344,12 @@ def only_if_tracing(func: Callable[P, R]) -> Callable[P, Optional[R]]:
return _only_if_tracing_inner
def ensure_active_span(message, ret=None):
def ensure_active_span(message: str, ret=None):
"""Executes the operation only if opentracing is enabled and there is an active span.
If there is no active span it logs message at the error level.
Args:
message (str): Message which fills in "There was no active span when trying to %s"
message: Message which fills in "There was no active span when trying to %s"
in the error log if there is no active span and opentracing is enabled.
ret (object): return value if opentracing is None or there is no active span.
@ -402,7 +407,7 @@ def init_tracer(hs: "HomeServer") -> None:
config = JaegerConfig(
config=hs.config.tracing.jaeger_config,
service_name=f"{hs.config.server.server_name} {hs.get_instance_name()}",
scope_manager=LogContextScopeManager(hs.config),
scope_manager=LogContextScopeManager(),
metrics_factory=PrometheusMetricsFactory(),
)
@ -451,15 +456,15 @@ def whitelisted_homeserver(destination: str) -> bool:
# Could use kwargs but I want these to be explicit
def start_active_span(
operation_name,
child_of=None,
references=None,
tags=None,
start_time=None,
ignore_active_span=False,
finish_on_close=True,
operation_name: str,
child_of: Optional[Union["opentracing.Span", "opentracing.SpanContext"]] = None,
references: Optional[List["opentracing.Reference"]] = None,
tags: Optional[Dict[str, str]] = None,
start_time: Optional[float] = None,
ignore_active_span: bool = False,
finish_on_close: bool = True,
*,
tracer=None,
tracer: Optional["opentracing.Tracer"] = None,
):
"""Starts an active opentracing span.
@ -493,11 +498,11 @@ def start_active_span(
def start_active_span_follows_from(
operation_name: str,
contexts: Collection,
child_of=None,
child_of: Optional[Union["opentracing.Span", "opentracing.SpanContext"]] = None,
start_time: Optional[float] = None,
*,
inherit_force_tracing=False,
tracer=None,
inherit_force_tracing: bool = False,
tracer: Optional["opentracing.Tracer"] = None,
):
"""Starts an active opentracing span, with additional references to previous spans
@ -540,7 +545,7 @@ def start_active_span_from_edu(
edu_content: Dict[str, Any],
operation_name: str,
references: Optional[List["opentracing.Reference"]] = None,
tags: Optional[Dict] = None,
tags: Optional[Dict[str, str]] = None,
start_time: Optional[float] = None,
ignore_active_span: bool = False,
finish_on_close: bool = True,
@ -617,23 +622,27 @@ def set_operation_name(operation_name: str) -> None:
@only_if_tracing
def force_tracing(span=Sentinel) -> None:
def force_tracing(
span: Union["opentracing.Span", _Sentinel] = _Sentinel.sentinel
) -> None:
"""Force sampling for the active/given span and its children.
Args:
span: span to force tracing for. By default, the active span.
"""
if span is Sentinel:
span = opentracing.tracer.active_span
if span is None:
if isinstance(span, _Sentinel):
span_to_trace = opentracing.tracer.active_span
else:
span_to_trace = span
if span_to_trace is None:
logger.error("No active span in force_tracing")
return
span.set_tag(opentracing.tags.SAMPLING_PRIORITY, 1)
span_to_trace.set_tag(opentracing.tags.SAMPLING_PRIORITY, 1)
# also set a bit of baggage, so that we have a way of figuring out if
# it is enabled later
span.set_baggage_item(SynapseBaggage.FORCE_TRACING, "1")
span_to_trace.set_baggage_item(SynapseBaggage.FORCE_TRACING, "1")
def is_context_forced_tracing(
@ -789,7 +798,7 @@ def extract_text_map(carrier: Dict[str, str]) -> Optional["opentracing.SpanConte
# Tracing decorators
def trace(func=None, opname=None):
def trace(func=None, opname: Optional[str] = None):
"""
Decorator to trace a function.
Sets the operation name to that of the function's or that given
@ -822,11 +831,11 @@ def trace(func=None, opname=None):
result = func(*args, **kwargs)
if isinstance(result, defer.Deferred):
def call_back(result):
def call_back(result: R) -> R:
scope.__exit__(None, None, None)
return result
def err_back(result):
def err_back(result: R) -> R:
scope.__exit__(None, None, None)
return result

View File

@ -16,11 +16,15 @@ import logging
from types import TracebackType
from typing import Optional, Type
from opentracing import Scope, ScopeManager
from opentracing import Scope, ScopeManager, Span
import twisted
from synapse.logging.context import current_context, nested_logging_context
from synapse.logging.context import (
LoggingContext,
current_context,
nested_logging_context,
)
logger = logging.getLogger(__name__)
@ -35,11 +39,11 @@ class LogContextScopeManager(ScopeManager):
but currently that doesn't work due to https://twistedmatrix.com/trac/ticket/10301.
"""
def __init__(self, config):
def __init__(self) -> None:
pass
@property
def active(self):
def active(self) -> Optional[Scope]:
"""
Returns the currently active Scope which can be used to access the
currently active Scope.span.
@ -48,19 +52,18 @@ class LogContextScopeManager(ScopeManager):
Tracer.start_active_span() time.
Return:
(Scope) : the Scope that is active, or None if not
available.
The Scope that is active, or None if not available.
"""
ctx = current_context()
return ctx.scope
def activate(self, span, finish_on_close):
def activate(self, span: Span, finish_on_close: bool) -> Scope:
"""
Makes a Span active.
Args
span (Span): the span that should become active.
finish_on_close (Boolean): whether Span should be automatically
finished when Scope.close() is called.
span: the span that should become active.
finish_on_close: whether Span should be automatically finished when
Scope.close() is called.
Returns:
Scope to control the end of the active period for
@ -112,8 +115,8 @@ class _LogContextScope(Scope):
def __init__(
self,
manager: LogContextScopeManager,
span,
logcontext,
span: Span,
logcontext: LoggingContext,
enter_logcontext: bool,
finish_on_close: bool,
):
@ -121,13 +124,13 @@ class _LogContextScope(Scope):
Args:
manager:
the manager that is responsible for this scope.
span (Span):
span:
the opentracing span which this scope represents the local
lifetime for.
logcontext (LogContext):
the logcontext to which this scope is attached.
logcontext:
the log context to which this scope is attached.
enter_logcontext:
if True the logcontext will be exited when the scope is finished
if True the log context will be exited when the scope is finished
finish_on_close:
if True finish the span when the scope is closed
"""

View File

@ -60,6 +60,9 @@ class ReplicationFederationSendEventsRestServlet(ReplicationEndpoint):
{
"max_stream_id": 32443,
}
Responds with a 409 when a `PartialStateConflictError` is raised due to an event
context that needs to be recomputed due to the un-partial stating of a room.
"""
NAME = "fed_send_events"

View File

@ -59,6 +59,9 @@ class ReplicationSendEventRestServlet(ReplicationEndpoint):
{ "stream_id": 12345, "event_id": "$abcdef..." }
Responds with a 409 when a `PartialStateConflictError` is raised due to an event
context that needs to be recomputed due to the un-partial stating of a room.
The returned event ID may not match the sent event if it was deduplicated.
"""

View File

@ -15,11 +15,11 @@
import logging
from typing import TYPE_CHECKING, Tuple
from synapse.api.errors import AuthError, NotFoundError, SynapseError
from synapse.api.errors import AuthError, Codes, NotFoundError, SynapseError
from synapse.http.server import HttpServer
from synapse.http.servlet import RestServlet, parse_json_object_from_request
from synapse.http.site import SynapseRequest
from synapse.types import JsonDict
from synapse.types import JsonDict, RoomID
from ._base import client_patterns
@ -104,6 +104,13 @@ class RoomAccountDataServlet(RestServlet):
if user_id != requester.user.to_string():
raise AuthError(403, "Cannot add account data for other users.")
if not RoomID.is_valid(room_id):
raise SynapseError(
400,
f"{room_id} is not a valid room ID",
Codes.INVALID_PARAM,
)
body = parse_json_object_from_request(request)
if account_data_type == "m.fully_read":
@ -111,6 +118,7 @@ class RoomAccountDataServlet(RestServlet):
405,
"Cannot set m.fully_read through this API."
" Use /rooms/!roomId:server.name/read_markers",
Codes.BAD_JSON,
)
await self.handler.add_account_data_to_room(
@ -130,6 +138,13 @@ class RoomAccountDataServlet(RestServlet):
if user_id != requester.user.to_string():
raise AuthError(403, "Cannot get account data for other users.")
if not RoomID.is_valid(room_id):
raise SynapseError(
400,
f"{room_id} is not a valid room ID",
Codes.INVALID_PARAM,
)
event = await self.store.get_account_data_for_room_and_type(
user_id, room_id, account_data_type
)

View File

@ -95,6 +95,8 @@ class VersionsRestServlet(RestServlet):
"org.matrix.msc3026.busy_presence": self.config.experimental.msc3026_enabled,
# Supports receiving private read receipts as per MSC2285
"org.matrix.msc2285": self.config.experimental.msc2285_enabled,
# Supports filtering of /publicRooms by room type MSC3827
"org.matrix.msc3827": self.config.experimental.msc3827_enabled,
# Adds support for importing historical messages as per MSC2716
"org.matrix.msc2716": self.config.experimental.msc2716_enabled,
# Adds support for jump to date endpoints (/timestamp_to_event) as per MSC3030

View File

@ -249,8 +249,12 @@ class StateHandler:
partial_state = True
logger.debug("calling resolve_state_groups from compute_event_context")
# we've already taken into account partial state, so no need to wait for
# complete state here.
entry = await self.resolve_state_groups_for_events(
event.room_id, event.prev_event_ids()
event.room_id,
event.prev_event_ids(),
await_full_state=False,
)
state_ids_before_event = entry.state
@ -335,7 +339,7 @@ class StateHandler:
@measure_func()
async def resolve_state_groups_for_events(
self, room_id: str, event_ids: Collection[str]
self, room_id: str, event_ids: Collection[str], await_full_state: bool = True
) -> _StateCacheEntry:
"""Given a list of event_ids this method fetches the state at each
event, resolves conflicts between them and returns them.
@ -343,6 +347,8 @@ class StateHandler:
Args:
room_id
event_ids
await_full_state: if true, will block if we do not yet have complete
state at these events.
Returns:
The resolved state
@ -350,7 +356,7 @@ class StateHandler:
logger.debug("resolve_state_groups event_ids %s", event_ids)
state_groups = await self._state_storage_controller.get_state_group_for_events(
event_ids
event_ids, await_full_state=await_full_state
)
state_group_ids = state_groups.values()

View File

@ -315,6 +315,10 @@ class EventsPersistenceStorageController:
if they were deduplicated due to an event already existing that
matched the transaction ID; the existing event is returned in such
a case.
Raises:
PartialStateConflictError: if attempting to persist a partial state event in
a room that has been un-partial stated.
"""
partitioned: Dict[str, List[Tuple[EventBase, EventContext]]] = {}
for event, ctx in events_and_contexts:
@ -363,6 +367,10 @@ class EventsPersistenceStorageController:
latest persisted event. The returned event may not match the given
event if it was deduplicated due to an existing event matching the
transaction ID.
Raises:
PartialStateConflictError: if attempting to persist a partial state event in
a room that has been un-partial stated.
"""
# add_to_queue returns a map from event ID to existing event ID if the
# event was deduplicated. (The dict may also include other entries if
@ -453,6 +461,10 @@ class EventsPersistenceStorageController:
Returns:
A dictionary of event ID to event ID we didn't persist as we already
had another event persisted with the same TXN ID.
Raises:
PartialStateConflictError: if attempting to persist a partial state event in
a room that has been un-partial stated.
"""
replaced_events: Dict[str, str] = {}
if not events_and_contexts:

View File

@ -366,10 +366,11 @@ class LoggingTransaction:
*args: P.args,
**kwargs: P.kwargs,
) -> R:
sql = self._make_sql_one_line(sql)
# Generate a one-line version of the SQL to better log it.
one_line_sql = self._make_sql_one_line(sql)
# TODO(paul): Maybe use 'info' and 'debug' for values?
sql_logger.debug("[SQL] {%s} %s", self.name, sql)
sql_logger.debug("[SQL] {%s} %s", self.name, one_line_sql)
sql = self.database_engine.convert_param_style(sql)
if args:
@ -389,7 +390,7 @@ class LoggingTransaction:
"db.query",
tags={
opentracing.tags.DATABASE_TYPE: "sql",
opentracing.tags.DATABASE_STATEMENT: sql,
opentracing.tags.DATABASE_STATEMENT: one_line_sql,
},
):
return func(sql, *args, **kwargs)

View File

@ -87,7 +87,6 @@ class DataStore(
RoomStore,
RoomBatchStore,
RegistrationStore,
StreamWorkerStore,
ProfileStore,
PresenceStore,
TransactionWorkerStore,
@ -112,6 +111,7 @@ class DataStore(
SearchStore,
TagsStore,
AccountDataStore,
StreamWorkerStore,
OpenIdStore,
ClientIpWorkerStore,
DeviceStore,

View File

@ -25,8 +25,8 @@ from synapse.storage.database import (
LoggingDatabaseConnection,
LoggingTransaction,
)
from synapse.storage.databases.main.events_worker import EventsWorkerStore
from synapse.storage.databases.main.receipts import ReceiptsWorkerStore
from synapse.storage.databases.main.stream import StreamWorkerStore
from synapse.util import json_encoder
from synapse.util.caches.descriptors import cached
@ -122,7 +122,7 @@ def _deserialize_action(actions: str, is_highlight: bool) -> List[Union[dict, st
return DEFAULT_NOTIF_ACTION
class EventPushActionsWorkerStore(ReceiptsWorkerStore, EventsWorkerStore, SQLBaseStore):
class EventPushActionsWorkerStore(ReceiptsWorkerStore, StreamWorkerStore, SQLBaseStore):
def __init__(
self,
database: DatabasePool,
@ -218,7 +218,7 @@ class EventPushActionsWorkerStore(ReceiptsWorkerStore, EventsWorkerStore, SQLBas
retcol="event_id",
)
stream_ordering = self.get_stream_id_for_event_txn(txn, event_id) # type: ignore[attr-defined]
stream_ordering = self.get_stream_id_for_event_txn(txn, event_id)
return self._get_unread_counts_by_pos_txn(
txn, room_id, user_id, stream_ordering
@ -307,12 +307,22 @@ class EventPushActionsWorkerStore(ReceiptsWorkerStore, EventsWorkerStore, SQLBas
actions that have been deleted from `event_push_actions` table.
"""
# If there have been no events in the room since the stream ordering,
# there can't be any push actions either.
if not self._events_stream_cache.has_entity_changed(room_id, stream_ordering):
return 0, 0
clause = ""
args = [user_id, room_id, stream_ordering]
if max_stream_ordering is not None:
clause = "AND ea.stream_ordering <= ?"
args.append(max_stream_ordering)
# If the max stream ordering is less than the min stream ordering,
# then obviously there are zero push actions in that range.
if max_stream_ordering <= stream_ordering:
return 0, 0
sql = f"""
SELECT
COUNT(CASE WHEN notif = 1 THEN 1 END),
@ -1003,8 +1013,7 @@ class EventPushActionsWorkerStore(ReceiptsWorkerStore, EventsWorkerStore, SQLBas
sql = """
SELECT user_id, room_id,
coalesce(old.%s, 0) + upd.cnt,
upd.stream_ordering,
old.user_id
upd.stream_ordering
FROM (
SELECT user_id, room_id, count(*) as cnt,
max(stream_ordering) as stream_ordering
@ -1032,7 +1041,6 @@ class EventPushActionsWorkerStore(ReceiptsWorkerStore, EventsWorkerStore, SQLBas
summaries[(row[0], row[1])] = _EventPushSummary(
unread_count=row[2],
stream_ordering=row[3],
old_user_id=row[4],
notif_count=0,
)
@ -1053,57 +1061,27 @@ class EventPushActionsWorkerStore(ReceiptsWorkerStore, EventsWorkerStore, SQLBas
summaries[(row[0], row[1])] = _EventPushSummary(
unread_count=0,
stream_ordering=row[3],
old_user_id=row[4],
notif_count=row[2],
)
logger.info("Rotating notifications, handling %d rows", len(summaries))
# If the `old.user_id` above is NULL then we know there isn't already an
# entry in the table, so we simply insert it. Otherwise we update the
# existing table.
self.db_pool.simple_insert_many_txn(
self.db_pool.simple_upsert_many_txn(
txn,
table="event_push_summary",
keys=(
"user_id",
"room_id",
"notif_count",
"unread_count",
"stream_ordering",
),
values=[
key_names=("user_id", "room_id"),
key_values=[(user_id, room_id) for user_id, room_id in summaries],
value_names=("notif_count", "unread_count", "stream_ordering"),
value_values=[
(
user_id,
room_id,
summary.notif_count,
summary.unread_count,
summary.stream_ordering,
)
for ((user_id, room_id), summary) in summaries.items()
if summary.old_user_id is None
for summary in summaries.values()
],
)
txn.execute_batch(
"""
UPDATE event_push_summary
SET notif_count = ?, unread_count = ?, stream_ordering = ?
WHERE user_id = ? AND room_id = ?
""",
(
(
summary.notif_count,
summary.unread_count,
summary.stream_ordering,
user_id,
room_id,
)
for ((user_id, room_id), summary) in summaries.items()
if summary.old_user_id is not None
),
)
txn.execute(
"UPDATE event_push_summary_stream_ordering SET stream_ordering = ?",
(rotate_to_stream_ordering,),
@ -1136,7 +1114,7 @@ class EventPushActionsWorkerStore(ReceiptsWorkerStore, EventsWorkerStore, SQLBas
txn.execute(
"""
SELECT stream_ordering FROM event_push_actions
WHERE stream_ordering < ? AND highlight = 0
WHERE stream_ordering <= ? AND highlight = 0
ORDER BY stream_ordering ASC LIMIT 1 OFFSET ?
""",
(
@ -1151,10 +1129,12 @@ class EventPushActionsWorkerStore(ReceiptsWorkerStore, EventsWorkerStore, SQLBas
else:
stream_ordering = max_stream_ordering_to_delete
# We need to use a inclusive bound here to handle the case where a
# single stream ordering has more than `batch_size` rows.
txn.execute(
"""
DELETE FROM event_push_actions
WHERE stream_ordering < ? AND highlight = 0
WHERE stream_ordering <= ? AND highlight = 0
""",
(stream_ordering,),
)
@ -1283,5 +1263,4 @@ class _EventPushSummary:
unread_count: int
stream_ordering: int
old_user_id: str
notif_count: int

View File

@ -16,6 +16,7 @@
import itertools
import logging
from collections import OrderedDict
from http import HTTPStatus
from typing import (
TYPE_CHECKING,
Any,
@ -35,6 +36,7 @@ from prometheus_client import Counter
import synapse.metrics
from synapse.api.constants import EventContentFields, EventTypes, RelationTypes
from synapse.api.errors import Codes, SynapseError
from synapse.api.room_versions import RoomVersions
from synapse.events import EventBase, relation_from_event
from synapse.events.snapshot import EventContext
@ -69,6 +71,24 @@ event_counter = Counter(
)
class PartialStateConflictError(SynapseError):
"""An internal error raised when attempting to persist an event with partial state
after the room containing the event has been un-partial stated.
This error should be handled by recomputing the event context and trying again.
This error has an HTTP status code so that it can be transported over replication.
It should not be exposed to clients.
"""
def __init__(self) -> None:
super().__init__(
HTTPStatus.CONFLICT,
msg="Cannot persist partial state event in un-partial stated room",
errcode=Codes.UNKNOWN,
)
@attr.s(slots=True, auto_attribs=True)
class DeltaState:
"""Deltas to use to update the `current_state_events` table.
@ -154,6 +174,10 @@ class PersistEventsStore:
Returns:
Resolves when the events have been persisted
Raises:
PartialStateConflictError: if attempting to persist a partial state event in
a room that has been un-partial stated.
"""
# We want to calculate the stream orderings as late as possible, as
@ -354,6 +378,9 @@ class PersistEventsStore:
For each room, a list of the event ids which are the forward
extremities.
Raises:
PartialStateConflictError: if attempting to persist a partial state event in
a room that has been un-partial stated.
"""
state_delta_for_room = state_delta_for_room or {}
new_forward_extremities = new_forward_extremities or {}
@ -1304,6 +1331,10 @@ class PersistEventsStore:
Returns:
new list, without events which are already in the events table.
Raises:
PartialStateConflictError: if attempting to persist a partial state event in
a room that has been un-partial stated.
"""
txn.execute(
"SELECT event_id, outlier FROM events WHERE event_id in (%s)"
@ -2215,6 +2246,11 @@ class PersistEventsStore:
txn: LoggingTransaction,
events_and_contexts: Collection[Tuple[EventBase, EventContext]],
) -> None:
"""
Raises:
PartialStateConflictError: if attempting to persist a partial state event in
a room that has been un-partial stated.
"""
state_groups = {}
for event, context in events_and_contexts:
if event.internal_metadata.is_outlier():
@ -2239,19 +2275,37 @@ class PersistEventsStore:
# if we have partial state for these events, record the fact. (This happens
# here rather than in _store_event_txn because it also needs to happen when
# we de-outlier an event.)
self.db_pool.simple_insert_many_txn(
txn,
table="partial_state_events",
keys=("room_id", "event_id"),
values=[
(
event.room_id,
event.event_id,
)
for event, ctx in events_and_contexts
if ctx.partial_state
],
)
try:
self.db_pool.simple_insert_many_txn(
txn,
table="partial_state_events",
keys=("room_id", "event_id"),
values=[
(
event.room_id,
event.event_id,
)
for event, ctx in events_and_contexts
if ctx.partial_state
],
)
except self.db_pool.engine.module.IntegrityError:
logger.info(
"Cannot persist events %s in rooms %s: room has been un-partial stated",
[
event.event_id
for event, ctx in events_and_contexts
if ctx.partial_state
],
list(
{
event.room_id
for event, ctx in events_and_contexts
if ctx.partial_state
}
),
)
raise PartialStateConflictError()
self.db_pool.simple_upsert_many_txn(
txn,

View File

@ -32,12 +32,17 @@ from typing import (
import attr
from synapse.api.constants import EventContentFields, EventTypes, JoinRules
from synapse.api.constants import (
EventContentFields,
EventTypes,
JoinRules,
PublicRoomsFilterFields,
)
from synapse.api.errors import StoreError
from synapse.api.room_versions import RoomVersion, RoomVersions
from synapse.config.homeserver import HomeServerConfig
from synapse.events import EventBase
from synapse.storage._base import SQLBaseStore, db_to_json
from synapse.storage._base import SQLBaseStore, db_to_json, make_in_list_sql_clause
from synapse.storage.database import (
DatabasePool,
LoggingDatabaseConnection,
@ -199,10 +204,29 @@ class RoomWorkerStore(CacheInvalidationWorkerStore):
desc="get_public_room_ids",
)
def _construct_room_type_where_clause(
self, room_types: Union[List[Union[str, None]], None]
) -> Tuple[Union[str, None], List[str]]:
if not room_types or not self.config.experimental.msc3827_enabled:
return None, []
else:
# We use None when we want get rooms without a type
is_null_clause = ""
if None in room_types:
is_null_clause = "OR room_type IS NULL"
room_types = [value for value in room_types if value is not None]
list_clause, args = make_in_list_sql_clause(
self.database_engine, "room_type", room_types
)
return f"({list_clause} {is_null_clause})", args
async def count_public_rooms(
self,
network_tuple: Optional[ThirdPartyInstanceID],
ignore_non_federatable: bool,
search_filter: Optional[dict],
) -> int:
"""Counts the number of public rooms as tracked in the room_stats_current
and room_stats_state table.
@ -210,11 +234,20 @@ class RoomWorkerStore(CacheInvalidationWorkerStore):
Args:
network_tuple
ignore_non_federatable: If true filters out non-federatable rooms
search_filter
"""
def _count_public_rooms_txn(txn: LoggingTransaction) -> int:
query_args = []
room_type_clause, args = self._construct_room_type_where_clause(
search_filter.get(PublicRoomsFilterFields.ROOM_TYPES, None)
if search_filter
else None
)
room_type_clause = f" AND {room_type_clause}" if room_type_clause else ""
query_args += args
if network_tuple:
if network_tuple.appservice_id:
published_sql = """
@ -249,6 +282,7 @@ class RoomWorkerStore(CacheInvalidationWorkerStore):
OR join_rules = '{JoinRules.KNOCK_RESTRICTED}'
OR history_visibility = 'world_readable'
)
{room_type_clause}
AND joined_members > 0
"""
@ -347,8 +381,12 @@ class RoomWorkerStore(CacheInvalidationWorkerStore):
if ignore_non_federatable:
where_clauses.append("is_federatable")
if search_filter and search_filter.get("generic_search_term", None):
search_term = "%" + search_filter["generic_search_term"] + "%"
if search_filter and search_filter.get(
PublicRoomsFilterFields.GENERIC_SEARCH_TERM, None
):
search_term = (
"%" + search_filter[PublicRoomsFilterFields.GENERIC_SEARCH_TERM] + "%"
)
where_clauses.append(
"""
@ -365,6 +403,15 @@ class RoomWorkerStore(CacheInvalidationWorkerStore):
search_term.lower(),
]
room_type_clause, args = self._construct_room_type_where_clause(
search_filter.get(PublicRoomsFilterFields.ROOM_TYPES, None)
if search_filter
else None
)
if room_type_clause:
where_clauses.append(room_type_clause)
query_args += args
where_clause = ""
if where_clauses:
where_clause = " AND " + " AND ".join(where_clauses)
@ -373,7 +420,7 @@ class RoomWorkerStore(CacheInvalidationWorkerStore):
sql = f"""
SELECT
room_id, name, topic, canonical_alias, joined_members,
avatar, history_visibility, guest_access, join_rules
avatar, history_visibility, guest_access, join_rules, room_type
FROM (
{published_sql}
) published
@ -1109,19 +1156,25 @@ class RoomWorkerStore(CacheInvalidationWorkerStore):
return room_servers
async def clear_partial_state_room(self, room_id: str) -> bool:
# this can race with incoming events, so we watch out for FK errors.
# TODO(faster_joins): this still doesn't completely fix the race, since the persist process
# is not atomic. I fear we need an application-level lock.
# https://github.com/matrix-org/synapse/issues/12988
"""Clears the partial state flag for a room.
Args:
room_id: The room whose partial state flag is to be cleared.
Returns:
`True` if the partial state flag has been cleared successfully.
`False` if the partial state flag could not be cleared because the room
still contains events with partial state.
"""
try:
await self.db_pool.runInteraction(
"clear_partial_state_room", self._clear_partial_state_room_txn, room_id
)
return True
except self.db_pool.engine.module.DatabaseError as e:
# TODO(faster_joins): how do we distinguish between FK errors and other errors?
# https://github.com/matrix-org/synapse/issues/12988
logger.warning(
except self.db_pool.engine.module.IntegrityError as e:
# Assume that any `IntegrityError`s are due to partial state events.
logger.info(
"Exception while clearing lazy partial-state-room %s, retrying: %s",
room_id,
e,
@ -1166,6 +1219,7 @@ class _BackgroundUpdates:
POPULATE_ROOM_DEPTH_MIN_DEPTH2 = "populate_room_depth_min_depth2"
REPLACE_ROOM_DEPTH_MIN_DEPTH = "replace_room_depth_min_depth"
POPULATE_ROOMS_CREATOR_COLUMN = "populate_rooms_creator_column"
ADD_ROOM_TYPE_COLUMN = "add_room_type_column"
_REPLACE_ROOM_DEPTH_SQL_COMMANDS = (
@ -1200,6 +1254,11 @@ class RoomBackgroundUpdateStore(SQLBaseStore):
self._background_add_rooms_room_version_column,
)
self.db_pool.updates.register_background_update_handler(
_BackgroundUpdates.ADD_ROOM_TYPE_COLUMN,
self._background_add_room_type_column,
)
# BG updates to change the type of room_depth.min_depth
self.db_pool.updates.register_background_update_handler(
_BackgroundUpdates.POPULATE_ROOM_DEPTH_MIN_DEPTH2,
@ -1569,6 +1628,69 @@ class RoomBackgroundUpdateStore(SQLBaseStore):
return batch_size
async def _background_add_room_type_column(
self, progress: JsonDict, batch_size: int
) -> int:
"""Background update to go and add room_type information to `room_stats_state`
table from `event_json` table.
"""
last_room_id = progress.get("room_id", "")
def _background_add_room_type_column_txn(
txn: LoggingTransaction,
) -> bool:
sql = """
SELECT state.room_id, json FROM event_json
INNER JOIN current_state_events AS state USING (event_id)
WHERE state.room_id > ? AND type = 'm.room.create'
ORDER BY state.room_id
LIMIT ?
"""
txn.execute(sql, (last_room_id, batch_size))
room_id_to_create_event_results = txn.fetchall()
new_last_room_id = None
for room_id, event_json in room_id_to_create_event_results:
event_dict = db_to_json(event_json)
room_type = event_dict.get("content", {}).get(
EventContentFields.ROOM_TYPE, None
)
if isinstance(room_type, str):
self.db_pool.simple_update_txn(
txn,
table="room_stats_state",
keyvalues={"room_id": room_id},
updatevalues={"room_type": room_type},
)
new_last_room_id = room_id
if new_last_room_id is None:
return True
self.db_pool.updates._background_update_progress_txn(
txn,
_BackgroundUpdates.ADD_ROOM_TYPE_COLUMN,
{"room_id": new_last_room_id},
)
return False
end = await self.db_pool.runInteraction(
"_background_add_room_type_column",
_background_add_room_type_column_txn,
)
if end:
await self.db_pool.updates._end_background_update(
_BackgroundUpdates.ADD_ROOM_TYPE_COLUMN
)
return batch_size
class RoomStore(RoomBackgroundUpdateStore, RoomWorkerStore):
def __init__(

View File

@ -16,7 +16,7 @@
import logging
from enum import Enum
from itertools import chain
from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple, cast
from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple, Union, cast
from typing_extensions import Counter
@ -238,6 +238,7 @@ class StatsStore(StateDeltasStore):
* avatar
* canonical_alias
* guest_access
* room_type
A is_federatable key can also be included with a boolean value.
@ -263,6 +264,7 @@ class StatsStore(StateDeltasStore):
"avatar",
"canonical_alias",
"guest_access",
"room_type",
):
field = fields.get(col, sentinel)
if field is not sentinel and (not isinstance(field, str) or "\0" in field):
@ -572,7 +574,7 @@ class StatsStore(StateDeltasStore):
state_event_map = await self.get_events(event_ids, get_prev_content=False) # type: ignore[attr-defined]
room_state = {
room_state: Dict[str, Union[None, bool, str]] = {
"join_rules": None,
"history_visibility": None,
"encryption": None,
@ -581,6 +583,7 @@ class StatsStore(StateDeltasStore):
"avatar": None,
"canonical_alias": None,
"is_federatable": True,
"room_type": None,
}
for event in state_event_map.values():
@ -604,6 +607,9 @@ class StatsStore(StateDeltasStore):
room_state["is_federatable"] = (
event.content.get(EventContentFields.FEDERATE, True) is True
)
room_type = event.content.get(EventContentFields.ROOM_TYPE)
if isinstance(room_type, str):
room_state["room_type"] = room_type
await self.update_room_state(room_id, room_state)

View File

@ -46,10 +46,12 @@ from typing import (
Set,
Tuple,
cast,
overload,
)
import attr
from frozendict import frozendict
from typing_extensions import Literal
from twisted.internet import defer
@ -795,6 +797,24 @@ class StreamWorkerStore(EventsWorkerStore, SQLBaseStore):
)
return RoomStreamToken(topo, stream_ordering)
@overload
def get_stream_id_for_event_txn(
self,
txn: LoggingTransaction,
event_id: str,
allow_none: Literal[False] = False,
) -> int:
...
@overload
def get_stream_id_for_event_txn(
self,
txn: LoggingTransaction,
event_id: str,
allow_none: bool = False,
) -> Optional[int]:
...
def get_stream_id_for_event_txn(
self,
txn: LoggingTransaction,

View File

@ -0,0 +1,19 @@
/* Copyright 2022 The Matrix.org Foundation C.I.C
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
ALTER TABLE room_stats_state ADD room_type TEXT;
INSERT INTO background_updates (update_name, progress_json)
VALUES ('add_room_type_column', '{}');

View File

@ -20,6 +20,7 @@ from typing import Any, Callable, Dict, Generator, Optional
import attr
from frozendict import frozendict
from matrix_common.versionstring import get_distribution_version_string
from typing_extensions import ParamSpec
from twisted.internet import defer, task
from twisted.internet.defer import Deferred
@ -82,6 +83,9 @@ def unwrapFirstError(failure: Failure) -> Failure:
return failure.value.subFailure # type: ignore[union-attr] # Issue in Twisted's annotations
P = ParamSpec("P")
@attr.s(slots=True)
class Clock:
"""
@ -110,7 +114,7 @@ class Clock:
return int(self.time() * 1000)
def looping_call(
self, f: Callable, msec: float, *args: Any, **kwargs: Any
self, f: Callable[P, object], msec: float, *args: P.args, **kwargs: P.kwargs
) -> LoopingCall:
"""Call a function repeatedly.

View File

@ -109,7 +109,7 @@ GLOBAL_ROOT = ListNode["_Node"].create_root_node()
@wrap_as_background_process("LruCache._expire_old_entries")
async def _expire_old_entries(
clock: Clock, expiry_seconds: int, autotune_config: Optional[dict]
clock: Clock, expiry_seconds: float, autotune_config: Optional[dict]
) -> None:
"""Walks the global cache list to find cache entries that haven't been
accessed in the given number of seconds, or if a given memory threshold has been breached.

View File

@ -50,7 +50,7 @@ class LogContextScopeManagerTestCase(TestCase):
# global variables that power opentracing. We create our own tracer instance
# and test with it.
scope_manager = LogContextScopeManager({})
scope_manager = LogContextScopeManager()
config = jaeger_client.config.Config(
config={}, service_name="test", scope_manager=scope_manager
)

Some files were not shown because too many files have changed in this diff Show More