Propagate opentracing contexts across workers
Also includes some Convenience modifications to opentracing for servlets, notably:
- Add boolean to skip the whitelisting check on inject
extract methods. - useful when injecting into carriers
locally. Otherwise we'd always have to include our
own servername and whitelist our servername
- start_active_span_from_request instead of header
- Add boolean to decide whether to extract context
from a request to a servlet
This type of registration was probably never used. It only includes the
user name in the HMAC but not the password.
Shared secret registration is still available via
client/r0/admin/register.
Signed-off-by: Manuel Stahl <manuel.stahl@awesome-technologies.de>
There's no point doing a raise_from here, because the exception is always
logged at warn with no stacktrace in the caller. Instead, let's try to give
better messages to reduce confusion.
In particular, this means that we won't log 'Failed to connect to remote
server' when we don't even attempt to connect to the remote server due to
blacklisting.
Get rid of the labyrinthine `recoverer_fn` code, and clean up the startup code
(it seemed to be previously inexplicably split between
`ApplicationServiceScheduler.start` and `_Recoverer.start`).
Add some docstrings too.
Hopefully, this will fix a stack overflow when recovering an appservice.
The recursion here leads to a huge chain of deferred callbacks, which then
overflows the stack when the chain completes. `inlineCallbacks` makes a better
job of this if we use iteration instead.
Clean up the code a bit too, while we're there.
Get rid of the labyrinthine `recoverer_fn` code, and clean up the startup code
(it seemed to be previously inexplicably split between
`ApplicationServiceScheduler.start` and `_Recoverer.start`).
Add some docstrings too.
Add authenticated_entity and servlet_names tags.
Functionally:
- Add a tag for authenticated_entity
- Add a tag for servlet_names
Stylistically:
Moved to importing methods directly from opentracing.
Fixes#5833
The emailconfig code was attempting to pull incorrect config file names. This corrects that, while also marking a difference between a config file variable that's a filepath versus a str containing HTML.
This refactors MatrixFederationAgent to move the SRV lookup into the
endpoint code, this has two benefits:
1. Its easier to retry different host/ports in the same way as
HostnameEndpoint.
2. We avoid SRV lookups if we have a free connection in the pool
If we have recently seen a valid well-known for a domain we want to
retry on (non-final) errors a few times, to handle temporary blips in
networking/etc.
is cached and so does not always return a `Deferred`.
`await` does not silently pass-through non-Deferreds like `yield` used to.
Signed-off-by: Olivier Wilkinson (reivilibre) <olivier@librepush.net>
This gives a bit of a grace period where we can attempt to refetch a
remote `well-known`, while still using the cached result if that fails.
Hopefully this will make the well-known resolution a bit more torelant
of failures, rather than it immediately treating failures as "no result"
and caching that for an hour.
* allow devices to be marked as "hidden"
This is a prerequisite for cross-signing, as it allows us to create other things
that live within the device namespace, so they can be used for signatures.
It costs both us and the remote server for us to fetch the well known
for every single request we send, so we add a minimum cache period. This
is set to 5m so that we still honour the basic premise of "refetch
frequently".
When persisting events we calculate new stream orderings up front.
Before we notify about an event all events with lower stream orderings
must have finished being persisted.
This PR moves the assignment of stream ordering till *after* calculated
the new current state and split the batch of events into separate chunks
for persistence. This means that if it takes a long time to calculate
new current state then it will not block events in other rooms being
notified about.
This should help reduce some global pauses in the events stream which
can last for tens of seconds (if not longer), caused by some
particularly expensive state resolutions.
This hopefully addresses #5407 by gracefully handling an empty but
limited TimelineBatch. We also add some logging to figure out how this
is happening.
This is intended as an amendment to #5674 as using M_UNKNOWN as the errcode makes it hard for clients to differentiate between an invalid password and a deactivated user (the problem we were trying to solve in the first place).
M_UNKNOWN was originally chosen as it was presumed than an MSC would have to be carried out to add a new code, but as Synapse often is the testing bed for new MSC implementations, it makes sense to try it out first in the wild and then add it into the spec if it is successful. Thus this PR return a new M_USER_DEACTIVATED code when a deactivated user attempts to login.
The `expire_access_token` didn't do what it sounded like it should do. What it
actually did was make Synapse enforce the 'time' caveat on macaroons used as
access tokens, but since our access token macaroons never contained such a
caveat, it was always a no-op.
(The code to add 'time' caveats was removed back in v0.18.5, in #1656)
Annoyingly, `current_state_events` table can include rejected events,
in which case the membership column will be null. To work around this
lets just always filter out null membership for now.
There was some inconsistent behaviour in the caching layer around how
exceptions were handled - particularly synchronously-thrown ones.
This seems to be most easily handled by pushing the creation of
ObservableDeferreds down from CacheDescriptor to the Cache.
This is a prerequisite for cross-signing, as it allows us to create other things
that live within the device namespace, so they can be used for signatures.
--------
- Fix a regression introduced in v1.2.0rc1 which led to incorrect labels on some prometheus metrics. ([\#5734](https://github.com/matrix-org/synapse/issues/5734))
-----BEGIN PGP SIGNATURE-----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=+JBn
-----END PGP SIGNATURE-----
Merge tag 'v1.2.0rc2' into develop
Bugfixes
--------
- Fix a regression introduced in v1.2.0rc1 which led to incorrect labels on some prometheus metrics. ([\#5734](https://github.com/matrix-org/synapse/issues/5734))
`None` is not a valid event id, so queuing up a database fetch for it seems
like a silly thing to do.
I considered making `get_event` return `None` if `event_id is None`, but then
its interaction with `allow_none` seemed uninituitive, and strong typing ftw.
* Fix servlet metric names
Co-Authored-By: Richard van der Hoff <1389908+richvdh@users.noreply.github.com>
* Remove redundant check
* Cover all return paths
This will allow us to efficiently filter out rooms that have been
forgotten in other queries without having to join against the
`room_memberships` table.
* Add decerators for tracing functions
* Use the new clean contexts
* Context and edu utils
* Move opentracing setters
* Move whitelisting
* Sectioning comments
* Better args wrapper
* Docstrings
Co-Authored-By: Erik Johnston <erik@matrix.org>
* Remove unused methods.
* Don't use global
* One tracing decorator to rule them all.
* Refactor Keyring._start_key_lookups
There's an awful lot of deferreds and dictionaries flying around here. The
whole thing can be made much simpler and achieve the same effect.
* Add a delay to key lookup lock release to fix stack overflow
A tactical call_later here should fix#5723
* changelog
The version of a module isn't going to change over the lifetime of the
process (assuming no funky hot reloading is going on, which it isn't),
so let's just cache the result to avoid spawning lots of git
subprocesses.
Fixes#5672.
* Opentracing survival guide
* Update decorator names in doc
* Doc cleanup
These are all alterations as a result of comments in #5703, it
includes mostly typos and clarifications. The most interesting
changes are:
- Split developer and user docs into two sections
- Add a high level description of OpenTracing
* newsfile
* Move contributer specific info to docstring.
* Sample config.
* Trailing whitespace.
* Update 5703.misc
* Apply suggestions from code review
Mostly just rewording parts of the docs for clarity.
Co-Authored-By: Richard van der Hoff <1389908+richvdh@users.noreply.github.com>
Clean up config settings and dead code.
This is mostly about cleaning up the config format, to bring it into line with our conventions. In particular:
* There should be a blank line after `## Section ##' headings
* There should be a blank line between each config setting
* There should be a `#`-only line between a comment and the setting it describes
* We don't really do the `# #` style commenting-out of whole sections if we can help it
* rename `tracer_enabled` to `enabled`
While we're here, do more config parsing upfront, which makes it easier to use
later on.
Also removes redundant code from LogContextScopeManager.
Also changes the changelog fragment to a `feature` - it's exciting!
* Convert BaseFederationServlet._wrap to async
Empirically, this fixes some lost stacktraces. It should be safe because the
wrapped function is called from JsonResource._async_render, which is already
async.
* Convert the rest of synapse.federation.transport.server to async
We may as well do the whole file while we're here.
* changelog
* flake8
We can now use `_get_events_from_cache_or_db` rather than going right back to
the database, which means that (a) we can benefit from caching, and (b) it
opens the way forward to more extensive checks on the original event.
We now always require the original event to exist before we will serve up a
redaction.
Ensures that redactions are correctly authenticated for recent room versions.
There are a few things going on here:
* `_fetch_event_rows` is updated to return a dict rather than a list of rows.
* Rather than returning multiple copies of an event which was redacted
multiple times, it returns the redactions as a list within the dict.
* It also returns the actual rejection reason, rather than merely the fact
that it was rejected, so that we don't have to query the table again in
`_get_event_from_row`.
* The redaction handling is factored out of `_get_event_from_row`, and now
checks if any of the redactions are valid.
A couple of changes here:
* get rid of a redundant `allow_rejected` condition - we should already have filtered out any rejected
events before we get to that point in the code, and the redundancy is confusing. Instead, let's stick in
an assertion just to make double-sure we aren't leaking rejected events by mistake.
* factor out a `_get_events_from_cache_or_db` method, which is going to be important for a
forthcoming fix to redactions.
First of all, let's get rid of `TOKEN_NOT_FOUND_HTTP_STATUS`. It was a hack we
did at one point when it was possible to return either a 403 or a 401 if the
creds were missing. We always return a 401 in these cases now (thankfully), so
it's not needed.
Let's also stop abusing `AuthError` for these cases. Honestly they have nothing
that relates them to the other places that `AuthError` is used, other than the
fact that they are loosely under the 'Auth' banner. It makes no sense for them
to share exception classes.
Instead, let's add a couple of new exception classes: `InvalidClientTokenError`
and `MissingClientTokenError`, for the `M_UNKNOWN_TOKEN` and `M_MISSING_TOKEN`
cases respectively - and an `InvalidClientCredentialsError` base class for the
two of them.
* Configure and initialise tracer
Includes config options for the tracer and sets up JaegerClient.
* Scope manager using LogContexts
We piggy-back our tracer scopes by using log context.
The current log context gives us the current scope. If new scope is
created we create a stack of scopes in the context.
* jaeger is a dependency now
* Carrier inject and extraction for Twisted Headers
* Trace federation requests on the way in and out.
The span is created in _started_processing and closed in
_finished_processing because we need a meaningful log context.
* Create logcontext for new scope.
Instead of having a stack of scopes in a logcontext we create a new
context for a new scope if the current logcontext already has a scope.
* Remove scope from logcontext if logcontext is top level
* Disable tracer if not configured
* typo
* Remove dependence on jaeger internals
* bools
* Set service name
* :Explicitely state that the tracer is disabled
* Black is the new black
* Newsfile
* Code style
* Use the new config setup.
* Generate config.
* Copyright
* Rename config to opentracing
* Remove user whitelisting
* Empty whitelist by default
* User ConfigError instead of RuntimeError
* Use isinstance
* Use tag constants for opentracing.
* Remove debug comment and no need to explicitely record error
* Two errors a "s(c)entry"
* Docstrings!
* Remove debugging brainslip
* Homeserver Whitlisting
* Better opentracing config comment
* linting
* Inclue worker name in service_name
* Make opentracing an optional dependency
* Neater config retreival
* Clean up dummy tags
* Instantiate tracing as object instead of global class
* Inlcude opentracing as a homeserver member.
* Thread opentracing to the request level
* Reference opetnracing through hs
* Instantiate dummy opentracin g for tests.
* About to revert, just keeping the unfinished changes just in case
* Revert back to global state, commit number:
9ce4a3d9067bf9889b86c360c05ac88618b85c4f
* Use class level methods in tracerutils
* Start and stop requests spans in a place where we
have access to the authenticated entity
* Seen it, isort it
* Make sure to close the active span.
* I'm getting black and blue from this.
* Logger formatting
Co-Authored-By: Erik Johnston <erik@matrix.org>
* Outdated comment
* Import opentracing at the top
* Return a contextmanager
* Start tracing client requests from the servlet
* Return noop context manager if not tracing
* Explicitely say that these are federation requests
* Include servlet name in client requests
* Use context manager
* Move opentracing to logging/
* Seen it, isort it again!
* Ignore twisted return exceptions on context exit
* Escape the scope
* Scopes should be entered to make them useful.
* Nicer decorator names
* Just one init, init?
* Don't need to close something that isn't open
* Docs make you smarter
this is only used in one place, so it's clearer if we inline it and reduce the
API surface.
Also, fixes a buglet where we would create an access token even if we were
about to block the user (we would never return the AT, so the user could never
use it, but it was still created and added to the db.)
A fix for PR #5626, which returned the original event content as part of a call to /relations.
Only problem was that we were attempting to aggregate the relations on top of it when we did so. We now set bundle_aggregations to False in the get_event call.
We also do this when pulling the relation events as well, because edits of edits are not something we'd like to support here.
FederationDeniedError is a subclass of SynapseError, which is a subclass of
CodeMessageException, so if e is a FederationDeniedError, then this check for
FederationDeniedError will never be reached since it will be caught by the
check for CodeMessageException above. The check for CodeMessageException does
almost the same thing as this check (since FederationDeniedError initialises
with code=403 and msg="Federation denied with %s."), so may as well just keep
allowing it to handle this case.
When asking for the relations of an event, include the original event in the response. This will mostly be used for efficiently showing edit history, but could be useful in other circumstances.
Nothing uses this now, so we can remove the dead code, and clean up the
API.
Since we're changing the shape of the return value anyway, we take the
opportunity to give the method a better name.
When a user creates an account and the 'require_auth_for_profile_requests' config flag is set, and a client that performed the registration wants to lookup the newly-created profile, the request will be denied because the user doesn't share a room with themselves yet.
This has never been documented, and I'm not sure it's ever been used outside
sytest.
It's quite a lot of poorly-maintained code, so I'd like to get rid of it.
For now I haven't removed the database table; I suggest we leave that for a
future clearout.
- Put the default window_size back to 1000ms (broken by #5181)
- Make the `rc_federation` config actually do something
- fix an off-by-one error in the 'concurrent' limit
- Avoid creating an unused `_PerHostRatelimiter` object for every single
incoming request
The runtime errors that dealt with local email password resets talked about config options that users may not even have in their config file yet (if upgrading). Instead, the cryptic errors are now replaced with hopefully much more helpful ones.
* SAML2 Improvements and redirect stuff
Signed-off-by: Alexander Trost <galexrt@googlemail.com>
* Code cleanups and simplifications.
Also: share the saml client between redirect and response handlers.
* changelog
* Revert redundant changes to static js
* Move all the saml stuff out to a centralised handler
* Add support for tracking SAML2 sessions.
This allows us to correctly handle `allow_unsolicited: False`.
* update sample config
* cleanups
* update sample config
* rename BaseSSORedirectServlet for consistency
* Address review comments
This would cause emails being sent, but Synapse responding with a 429 when creating the event. The client would then retry, and with bad timing the same scenario would happen again. Some testing I did ended up sending me 10 emails for one single invite because of this.
When a client asks for users whose devices have changed since a token we
used to pull *all* users from the database since the token, which could
easily be thousands of rows for old tokens.
This PR changes this to only check for changes for users the client is
actually interested in.
Fixes#5553
Closes#4583
Does slightly less than #5045, which prevented a room from being upgraded multiple times, one after another. This PR still allows that, but just prevents two from happening at the same time.
Mostly just to mitigate the fact that servers are slow and it can take a moment for the room upgrade to actually complete. We don't want people sending another request to upgrade the room when really they just thought the first didn't go through.
If no `from` param is specified we calculate and use the "current
token" that inlcuded typing, presence, etc. These are unused during
pagination and are not available on workers, so we simply don't
calculate them.
* group the arguments together into a group
* add new names "--generate-missing-config" and "--config-directory" for
existing cmdline options "--generate-keys" and "--keys-dir", which better
reflect their purposes.
Fixes https://github.com/matrix-org/synapse/issues/5431
`jinja2` was being imported even when it wasn't strictly necessary. This made it required to run Synapse, even if the functionality that required it wasn't enabled. This was causing new Synapse installations to crash on startup.
Email modules are now required.
There is a README.txt which always sets off this warning, which is a bit
alarming when you first start synapse. I don't think we need to warn about
this.
If, for some reason, presence updates take a while to persist then it
can trigger clients to tightloop calling `/sync` due to the presence
handler returning updates but not advancing the stream token.
Fixes#5503.
Fixes intermittent errors observed on Apple hardware which were caused by
time.clock() appearing to go backwards when called from different threads.
Also fixes a bug where database activity times were logged as 1/1000 of their
correct ratio due to confusion between milliseconds and seconds.
I had to add quite a lot of logging to diagnose a problem with 3pid
invites - we only logged the one failure which isn't all that
informative.
NB. I'm not convinced the logic of this loop is right: I think it
should just accept a single valid signature from a trusted source
rather than fail if *any* signature is invalid. Also it should
probably not skip the rest of middle loop if a check fails? However,
I'm deliberately not changing the logic here.
Adds new config option `cleanup_extremities_with_dummy_events` which
periodically sends dummy events to rooms with more than 10 extremities.
THIS IS REALLY EXPERIMENTAL.
Fixes that when a user exchanges a 3PID invite for a proper invite over
federation it does not include the `invite_room_state` key.
This was due to synapse incorrectly sending out two invite requests.
* remove 2.7 from CI and publishing
* fill out classifiers and also make it not be installed on 3.5
* some minor bumps so that the old deps work on python 3.5
Moves the warning about password resets being disabled to the point where a user actually tries to reset their password. Is this an appropriate place for it to happen?
Also removed the disabling of msisdn password resets when you don't have an email config, as that just doesn't make sense.
Also change the error a user receives upon disabled passwords to specify that only email-based password reset is disabled.
If we try and send a transaction with lots of EDUs and we run out of
space, we call get_new_device_msgs_for_remote with a limit of 0, which
then failed.
Some keys are stored in the synapse database with a null valid_until_ms
which caused an exception to be thrown when using that key. We fix this
by treating nulls as zeroes, i.e. they keys will match verification
requests with a minimum_valid_until_ms of zero (i.e. don't validate ts)
but will not match requests with a non-zero minimum_valid_until_ms.
Fixes#5391.
It's not really a problem to trust notary responses signed by the old key so
long as we are also doing TLS validation.
This commit adds a check to the config parsing code at startup to check that
we do not have the insecure matrix.org key without tls validation, and refuses
to start without it.
This allows us to remove the rather alarming-looking warning which happens at
runtime.
When we try and calculate a description for a room for with no name but
multiple other users we threw an exception (due to trying to subscript
result of `dict.values()`).
Sends password reset emails from the homeserver instead of proxying to the identity server. This is now the default behaviour for security reasons. If you wish to continue proxying password reset requests to the identity server you must now enable the email.trust_identity_server_for_password_resets option.
This PR is a culmination of 3 smaller PRs which have each been separately reviewed:
* #5308
* #5345
* #5368
There are a few changes going on here:
* We make checking the signature on a key server response optional: if no
verify_keys are specified, we trust to TLS to validate the connection.
* We change the default config so that it does not require responses to be
signed by the old key.
* We replace the old 'perspectives' config with 'trusted_key_servers', which
is also formatted slightly differently.
* We emit a warning to the logs every time we trust a key server response
signed by the old key.
* Fix background updates to handle redactions/rejections
In background updates based on current state delta stream we need to
handle that we may not have all the events (or at least that
`get_events` may raise an exception).
Also:
* rename VerifyKeyRequest->VerifyJsonRequest
* calculate key_ids on VerifyJsonRequest construction
* refactor things to pass around VerifyJsonRequests instead of 4-tuples
FederationClient.get_pdu is called in a loop to fetch a batch of PDUs. A
failure to fetch one should not result in a failure of the whole batch. Add the
missing `continue`.
We have too many things called get_event, and it's hard to figure out what we
mean. Also remove some unused params from the signature, and add some logging.
It takes at least 20 minutes to work through the long_retries schedule (11
attempts, each with a 60 second timeout, and 60 seconds between each request),
so if the notary server isn't returning within the timeout, we'll just end up
blocking whatever request is happening for 20 minutes.
Ain't nobody got time for that.
