This PR adds 2 `dendrite-demo` main's, each designed expressly to serve
a Hidden Service/Overlay network.
The first, `dendrite-demo-i2p` add self-configuration for use of
dendrite as an I2P hidden service(eepsite) and to connect to I2P
services(federate) as an I2P client. It further disables the `dendrite`
server from communicating with non-anonymous servers by
federation(because I2P does not canonically have the ability to exit, we
rely on donors for exit traffic), and enables the use of self-signed TLS
certificates([because I2P services are self-authenticating but TLS is
still required for other aspects of the system to work
reliably](https://tor.stackexchange.com/questions/13887/registering-onion-with-certificate-authority)).
This demo turns the system into an "pseudonymous" homeserver which
people can connect to using an I2P-enabled Matrix client(I like `cinny`
and it's what I tested with).
The second, `dendrite-demo-tor` adds self-configuration for the use of
dendrite as an Onion service and to connect to other onion services and
non-anonymous web sites using Tor to obfuscate it's physical location
and providing, optionally, pseudonymity. It also enables the use of
self-signed TLS certificates, for the same reason as with I2P, because
onion services aren't typically eligible for TLS certificates. It has
also been tested with `cinny`.
These services are both pseudonymous like myself, not anonymous. I will
be meeting members of the element team at the CCC assembly shortly to
discuss contributing under my pseudonym.
As none of the other `dendrite-demo` have unit tests I did not add them
to these checkins.
* [*] I have added Go unit tests or [Complement integration
tests](https://github.com/matrix-org/complement) for this PR _or_ I have
justified why this PR doesn't need tests
---------
Co-authored-by: eyedeekay <idk@mulder>
Co-authored-by: Till Faelligen <2353100+S7evinK@users.noreply.github.com>
Not 100% on how you would want to test this; you would need a NATS
server configured with NKey:
https://docs.nats.io/using-nats/developer/connecting/creds
This was tested with Synadia's free NATS SaaS and it does appear to be
working, however there's an issue with how NATS is used in general:
```
time="2024-09-10T14:40:05.105105731Z" level=fatal msg="Unable to add in-memory stream" error="nats: account requires a stream config to have max bytes set" stream=DendriteInputRoomEvent subjects="[DendriteInputRoomEvent DendriteInputRoomEvent.>]"
```
I tried creating the topic manually, however dendrite insists on
deleting/recreating the topic, so getting this to work is an issue I'm
going ot have to deal with later unless somebody gets to it before then.
If you feel more competent than me and wanna draw from this PR as an
example (if you have another way you'd prefer to see this done) go ahead
feel free I just wanna see it get done and I'm not particularly good at
working with golang.
Signed-off-by: `Paige Thompson <paige@paige.bio>`
The internal NATS instance is definitely convenient but it does have one
problem: its lifecycle is tied to the Dendrite process. That means if
Dendrite panics or OOMs, it takes out NATS with it. I suspect this is
sometimes contributing to what people see with stuck streams, as some
operations or state might not be written to disk fully before it gets
interrupted.
Using `SyncAlways` means that NATS will effectively use `O_SYNC` and
block writes on flushes, which should improve resiliency against this
kind of failure considerably. It might affect performance a little but
shouldn't be significant.
Also updates NATS to 2.10.20 as there have been all sorts of fixes since
2.10.7, including better `SyncAlways` handling.
Signed-off-by: Neil Alexander <git@neilalexander.dev>
---------
Signed-off-by: Neil Alexander <git@neilalexander.dev>
Co-authored-by: Neil Alexander <neilalexander@users.noreply.github.com>
Given that #2714 wasn't merged but we are now at a minimum supported Go
version of 1.20 (soon to be 1.21), I wanted to carry over some of the
changes. Namely:
- Fix the log typo
- Simplify build constraints for unix
- Use stdlib atomic package
### Pull Request Checklist
<!-- Please read
https://matrix-org.github.io/dendrite/development/contributing before
submitting your pull request -->
* [x] I have added Go unit tests or [Complement integration
tests](https://github.com/matrix-org/complement) for this PR _or_ I have
justified why this PR doesn't need tests
* [x] Pull request includes a [sign off below using a legally
identifiable
name](https://matrix-org.github.io/dendrite/development/contributing#sign-off)
_or_ I have already signed off privately
Signed-off-by: `0x1a8510f2 <admin@0x1a8510f2.space>`
---------
Co-authored-by: devonh <devon.dmytro@gmail.com>
This should avoid 2 additional DB roundtrips if we don't want to use
relays.
So instead of possibly doing roughly 20k trips to the DB, we are now
"only" doing ~6600.
---------
Co-authored-by: devonh <devon.dmytro@gmail.com>
This brings Dendrite's appservice spec support up to v1.4, from the
previous level of pre-release-spec support only (even r0.1.0 wasn't
supported for pushing transactions 🙃). There are config options to
revert to the old behavior, but the default is v1.4+ only. [Synapse also
does
that](https://element-hq.github.io/synapse/latest/usage/configuration/config_documentation.html#use_appservice_legacy_authorization)
mautrix bridges will drop support for legacy paths and authentication
soon (and possibly also require matrix v1.4 to be advertised, but I
might add some workaround to not require that for dendrite)
Signed-off-by: Tulir Asokan <tulir@maunium.net>
Part of https://github.com/matrix-org/dendrite/issues/3216
The files are basically copied from Synapse, with minor changes to the
called endpoints. We never seem to have had the
`/_matrix/static/client/login/` endpoint, this adds it.
This introduces a new stream the syncAPI produces to once it processed a
`OutputRoomEvent` and the appservices consumes.
This is to work around a race condition where appservices receive an
event before the syncAPI has handled it, this can result in e.g. calls
to `/joined_members` returning a wrong membership list.
This makes the following changes:
- Adds two new metrics observing the usage of the `DeviceListUpdater`
workers
- Makes the number of workers configurable
- Adds a 30s timeout for DB requests when receiving a device list update
over federation
Adds the `org.matrix.msc3575.proxy` field (used for configuring sliding
sync) to /.well-known/matrix/client when Dendrite is serving that
endpoint and `well_known_sliding_sync_proxy` has been configured.
ie. Config values of:
``` yaml
global:
well_known_client_name: https://example.com
well_known_sliding_sync_proxy: https://syncv3.example.com
```
results in a /.well-known/matrix/client of:
``` json
{
"m.homeserver": {
"base_url": "https://example.com"
},
"org.matrix.msc3575.proxy": {
"url": "https://syncv3.example.com"
}
}
```
If `well_known_sliding_sync_proxy` is not provided, the json provided by
/.well-known/matrix/client does not include the proxy field.
ie.
``` json
{
"m.homeserver": {
"base_url": "https://example.com"
}
}
```
This PR adds a config key `room_server.default_config_key` to set the
default room version for the room server.
Signed-off-by: `Sam Wedgwood <sam@wedgwood.dev>`
There are cases where a dendrite instance is unaware of a pseudo ID for
a user, the user is not a member of that room. To represent this case,
we currently use the 'zero' value, which is often not checked and so
causes errors later down the line. To make this case more explict, and
to be consistent with `QueryUserIDForSender`, this PR changes this to
use a pointer (and `nil` to mean no sender ID).
Signed-off-by: `Sam Wedgwood <sam@wedgwood.dev>`
If old messages build up in the input stream and do not get processed
successfully, this can create a significant drift between the stream
first sequence and the consumer ack floors, which results in a slow and
expensive start-up when interest-based retention is in use.
If a message is sat in the stream for 24 hours, it's probably not going
to get processed successfully, so let NATS drop them instead. Dendrite
can reconcile by fetching missing events later if it needs to.
---------
Co-authored-by: Neil Alexander <neilalexander@users.noreply.github.com>
They are fundamentally different concepts, so should be represented as
such. Proto events are exchanged in /make_xxx calls over federation, and
made as "fledgling" events in /createRoom and general event sending.
*Building* events is a reasonably complex VERSION SPECIFIC process which
needs amongst other things, auth event providers, prev events, signing
keys, etc.
Requires https://github.com/matrix-org/gomatrixserverlib/pull/379
Requires https://github.com/matrix-org/gomatrixserverlib/pull/376
This has numerous upsides:
- Less type casting to `*Event` is required.
- Making Dendrite work with `PDU` interfaces means we can swap out Event
impls more easily.
- Tests which represent weird event shapes are easier to write.
Part of a series of refactors on GMSL.
We only use it in a few places currently, enough to get things to
compile and run. We should be using it in much more places.
Similarly, in some places we cast []PDU back to []*Event, we need to not
do that. Likewise, in some places we cast PDU to *Event, we need to not
do that. For now though, hopefully this is a start.
Replaced with types.HeaderedEvent _for now_. In reality we want to move
them all to gmsl.Event and only use HeaderedEvent when we _need_ to
bundle the version/event ID with the event (seriailsation boundaries,
and even then only when we don't have the room version).
Requires https://github.com/matrix-org/gomatrixserverlib/pull/373
Adds tests for `QueryRestrictedJoinAllowed`, `IsServerAllowed` and
`PerformRoomUpgrade`. Refactors the `QueryRoomVersionForRoom` method to
accept a string and return a `gmsl.RoomVersion` instead of req/resp
structs.
Adds some more caching for `GetStateEvent`
This should also fix#2912 by ignoring state events belonging to other
users.
This is the last part of unix socket support to talk to app servers, go
based app services already support unix sockets:
5a68173fe3
```
appservice:
# The address that the homeserver can use to connect to this appservice.
address: unix:///var/snap/matrix/current/whatsapp.socket
# The hostname and port where this appservice should listen.
hostname: /var/snap/matrix/current/whatsapp.socket
port: 0
```
### Pull Request Checklist
<!-- Please read
https://matrix-org.github.io/dendrite/development/contributing before
submitting your pull request -->
* [x] I have added Go unit tests or [Complement integration
tests](https://github.com/matrix-org/complement) for this PR _or_ I have
justified why this PR doesn't need tests
* [x] Pull request includes a [sign off below using a legally
identifiable
name](https://matrix-org.github.io/dendrite/development/contributing#sign-off)
_or_ I have already signed off privately
Signed-off-by: `Boris Rybalkin <ribalkin@gmail.com>`
Preparations to actually remove/replace `BaseDendrite`.
Quite a few changes:
- SyncAPI accepts an `fulltext.Indexer` interface (fulltext is removed
from `BaseDendrite`)
- Caches are removed from `BaseDendrite`
- Introduces a `Router` struct (likely to change)
- also fixes#2903
- Introduces a `sqlutil.ConnectionManager`, which should remove
`base.DatabaseConnection` later on
- probably more
This PR changes the following:
- `StoreEvent` now only stores an event (and possibly prev event),
instead of also doing redactions
- Adds a `MaybeRedactEvent` (pulled out from `StoreEvent`), which should
be called after storing events
- a few other things
