In order to circumvent the number of duplicate foo:count metrics increasing
without bounds, it's time for a rearrangement.

The following are all deprecated, and replaced with synapse_util_metrics_block_count:
  synapse_util_metrics_block_timer:count
  synapse_util_metrics_block_ru_utime:count
  synapse_util_metrics_block_ru_stime:count
  synapse_util_metrics_block_db_txn_count:count
  synapse_util_metrics_block_db_txn_duration:count

The following are all deprecated, and replaced with synapse_http_server_response_count:
   synapse_http_server_requests
   synapse_http_server_response_time:count
   synapse_http_server_response_ru_utime:count
   synapse_http_server_response_ru_stime:count
   synapse_http_server_response_db_txn_count:count
   synapse_http_server_response_db_txn_duration:count

The following are renamed (the old metrics are kept for now, but deprecated):

  synapse_util_metrics_block_timer:total ->
     synapse_util_metrics_block_time_seconds

  synapse_util_metrics_block_ru_utime:total ->
     synapse_util_metrics_block_ru_utime_seconds

  synapse_util_metrics_block_ru_stime:total ->
     synapse_util_metrics_block_ru_stime_seconds

  synapse_util_metrics_block_db_txn_count:total ->
     synapse_util_metrics_block_db_txn_count

  synapse_util_metrics_block_db_txn_duration:total ->
     synapse_util_metrics_block_db_txn_duration_seconds

  synapse_http_server_response_time:total ->
     synapse_http_server_response_time_seconds

  synapse_http_server_response_ru_utime:total ->
     synapse_http_server_response_ru_utime_seconds

  synapse_http_server_response_ru_stime:total ->
     synapse_http_server_response_ru_stime_seconds

   synapse_http_server_response_db_txn_count:total ->
      synapse_http_server_response_db_txn_count

   synapse_http_server_response_db_txn_duration:total
      synapse_http_server_response_db_txn_duration_seconds

Make Counter render floats

Prometheus handles all metrics as floats, and sometimes we store non-integer
values in them (notably, durations in seconds), so let's render them as floats
too.

(Note that the standard client libraries also treat Counters as floats.)

Reinstate media download on thumbnail request

We need to actually download the remote media when we get a request for a
thumbnail.

When using synctl with workers, don't start the main synapse automatically

Add /room/{id}/event/{id} to synapse

Do bcrypt hashing in a background thread

Fix publicised groups GET API (singular) over federation

Signed-off-by: Michael Telatynski <7t3chguy@gmail.com>

which was missing its fed client API, since there is no other API
it might as well reuse the bulk one and unwrap it

Signed-off-by: Michael Telatynski <7t3chguy@gmail.com>

Update http request metrics before calling servlet

Make sure that we set the servlet name in the metrics object *before* calling
the servlet, in case the servlet throws an exception.

Remove dead code related to default thumbnails

Turns out that there is a valid usecase for retrieving event by id (notably
having received a push), but event ids should be scoped to room, so /event/{id}
is wrong.

Fix flaky test_rooms UTs

Fix broken config UTs

Make indentation of generated log config consistent

pointless function is pointless

Fix flakiness in the UTs caused by the user_directory being updated in the
background

https://github.com/matrix-org/synapse/pull/2755 broke log-config generation,
which in turn broke the unit tests.

(we had a mix of 2- and 4-space indents)

Fix templating error with unban permission message

Fixes https://github.com/matrix-org/synapse/issues/2759



Signed-off-by: Travis Ralston <travpc@gmail.com>

Remove 'verbosity'/'log_file' from generated cfg

... because these only really exist to confuse people nowadays.

Also bring log config more into line with the generated log config, by making `level_for_storage`
apply to the `synapse.storage.SQL` logger rather than `synapse.storage`.

Better logging when login can't find a 3pid