From 9e05c8d3090a956054b5b39347359655236caaef Mon Sep 17 00:00:00 2001 From: Erik Johnston Date: Tue, 11 Sep 2018 10:42:10 +0100 Subject: Change the manhole SSH key to have more bits Newer versions of openssh client refuse to connect to the old key due to its length. --- synapse/util/manhole.py | 44 +++++++++++++++++++++++++++++++------------- 1 file changed, 31 insertions(+), 13 deletions(-) (limited to 'synapse/util') diff --git a/synapse/util/manhole.py b/synapse/util/manhole.py index 14be3c7396..8d0f2a8918 100644 --- a/synapse/util/manhole.py +++ b/synapse/util/manhole.py @@ -19,22 +19,40 @@ from twisted.conch.ssh.keys import Key from twisted.cred import checkers, portal PUBLIC_KEY = ( - "ssh-rsa AAAAB3NzaC1yc2EAAAABIwAAAGEArzJx8OYOnJmzf4tfBEvLi8DVPrJ3/c9k2I/Az" - "64fxjHf9imyRJbixtQhlH9lfNjUIx+4LmrJH5QNRsFporcHDKOTwTTYLh5KmRpslkYHRivcJS" - "kbh/C+BR3utDS555mV" + "ssh-rsa AAAAB3NzaC1yc2EAAAADAQABAAABAQDHhGATaW4KhE23+7nrH4jFx3yLq9OjaEs5" + "XALqeK+7385NlLja3DE/DO9mGhnd9+bAy39EKT3sTV6+WXQ4yD0TvEEyUEMtjWkSEm6U32+C" + "DaS3TW/vPBUMeJQwq+Ydcif1UlnpXrDDTamD0AU9VaEvHq+3HAkipqn0TGpKON6aqk4vauDx" + "oXSsV5TXBVrxP/y7HpMOpU4GUWsaaacBTKKNnUaQB4UflvydaPJUuwdaCUJGTMjbhWrjVfK+" + "jslseSPxU6XvrkZMyCr4znxvuDxjMk1RGIdO7v+rbBMLEgqtSMNqJbYeVCnj2CFgc3fcTcld" + "X2uOJDrJb/WRlHulthCh" ) PRIVATE_KEY = """-----BEGIN RSA PRIVATE KEY----- -MIIByAIBAAJhAK8ycfDmDpyZs3+LXwRLy4vA1T6yd/3PZNiPwM+uH8Yx3/YpskSW -4sbUIZR/ZXzY1CMfuC5qyR+UDUbBaaK3Bwyjk8E02C4eSpkabJZGB0Yr3CUpG4fw -vgUd7rQ0ueeZlQIBIwJgbh+1VZfr7WftK5lu7MHtqE1S1vPWZQYE3+VUn8yJADyb -Z4fsZaCrzW9lkIqXkE3GIY+ojdhZhkO1gbG0118sIgphwSWKRxK0mvh6ERxKqIt1 -xJEJO74EykXZV4oNJ8sjAjEA3J9r2ZghVhGN6V8DnQrTk24Td0E8hU8AcP0FVP+8 -PQm/g/aXf2QQkQT+omdHVEJrAjEAy0pL0EBH6EVS98evDCBtQw22OZT52qXlAwZ2 -gyTriKFVoqjeEjt3SZKKqXHSApP/AjBLpF99zcJJZRq2abgYlf9lv1chkrWqDHUu -DZttmYJeEfiFBBavVYIF1dOlZT0G8jMCMBc7sOSZodFnAiryP+Qg9otSBjJ3bQML -pSTqy7c3a2AScC/YyOwkDaICHnnD3XyjMwIxALRzl0tQEKMXs6hH8ToUdlLROCrP -EhQ0wahUTCk1gKA4uPD6TMTChavbh4K63OvbKg== +MIIEpQIBAAKCAQEAx4RgE2luCoRNt/u56x+Ixcd8i6vTo2hLOVwC6nivu9/OTZS4 +2twxPwzvZhoZ3ffmwMt/RCk97E1evll0OMg9E7xBMlBDLY1pEhJulN9vgg2kt01v +7zwVDHiUMKvmHXIn9VJZ6V6ww02pg9AFPVWhLx6vtxwJIqap9ExqSjjemqpOL2rg +8aF0rFeU1wVa8T/8ux6TDqVOBlFrGmmnAUyijZ1GkAeFH5b8nWjyVLsHWglCRkzI +24Vq41Xyvo7JbHkj8VOl765GTMgq+M58b7g8YzJNURiHTu7/q2wTCxIKrUjDaiW2 +HlQp49ghYHN33E3JXV9rjiQ6yW/1kZR7pbYQoQIDAQABAoIBAQC8KJ0q8Wzzwh5B +esa1dQHZ8+4DEsL/Amae66VcVwD0X3cCN1W2IZ7X5W0Ij2kBqr8V51RYhcR+S+Ek +BtzSiBUBvbKGrqcMGKaUgomDIMzai99hd0gvCCyZnEW1OQhFkNkaRNXCfqiZJ27M +fqvSUiU2eOwh9fCvmxoA6Of8o3FbzcJ+1GMcobWRllDtLmj6lgVbDzuA+0jC5daB +9Tj1pBzu3wn3ufxiS+gBnJ+7NcXH3E73lqCcPa2ufbZ1haxfiGCnRIhFXuQDgxFX +vKdEfDgtvas6r1ahGbc+b/q8E8fZT7cABuIU4yfOORK+MhpyWbvoyyzuVGKj3PKt +KSPJu5CZAoGBAOkoJfAVyYteqKcmGTanGqQnAY43CaYf6GdSPX/jg+JmKZg0zqMC +jWZUtPb93i+jnOInbrnuHOiHAxI8wmhEPed28H2lC/LU8PzlqFkZXKFZ4vLOhhRB +/HeHCFIDosPFlohWi3b+GAjD7sXgnIuGmnXWe2ea/TS3yersifDEoKKjAoGBANsQ +gJX2cJv1c3jhdgcs8vAt5zIOKcCLTOr/QPmVf/kxjNgndswcKHwsxE/voTO9q+TF +v/6yCSTxAdjuKz1oIYWgi/dZo82bBKWxNRpgrGviU3/zwxiHlyIXUhzQu78q3VS/ +7S1XVbc7qMV++XkYKHPVD+nVG/gGzFxumX7MLXfrAoGBAJit9cn2OnjNj9uFE1W6 +r7N254ndeLAUjPe73xH0RtTm2a4WRopwjW/JYIetTuYbWgyujc+robqTTuuOZjAp +H/CG7o0Ym251CypQqaFO/l2aowclPp/dZhpPjp9GSjuxFBZLtiBB3DNBOwbRQzIK +/vLTdRQvZkgzYkI4i0vjNt3JAoGBANP8HSKBLymMlShlrSx2b8TB9tc2Y2riohVJ +2ttqs0M2kt/dGJWdrgOz4mikL+983Olt/0P9juHDoxEEMK2kpcPEv40lnmBpYU7h +s8yJvnBLvJe2EJYdJ8AipyAhUX1FgpbvfxmASP8eaUxsegeXvBWTGWojAoS6N2o+ +0KSl+l3vAoGAFqm0gO9f/Q1Se60YQd4l2PZeMnJFv0slpgHHUwegmd6wJhOD7zJ1 +CkZcXwiv7Nog7AI9qKJEUXLjoqL+vJskBzSOqU3tcd670YQMi1aXSXJqYE202K7o +EddTrx3TNpr1D5m/f+6mnXWrc8u9y1+GNx9yz889xMjIBTBI9KqaaOs= -----END RSA PRIVATE KEY-----""" -- cgit 1.4.1 From 0a81038ea0e308d2944b7deae43060e4982b0abe Mon Sep 17 00:00:00 2001 From: Erik Johnston Date: Fri, 14 Sep 2018 14:39:59 +0100 Subject: Add in flight real time metrics for Measure blocks --- synapse/metrics/__init__.py | 110 +++++++++++++++++++++++++++++++++++++++++++- synapse/util/metrics.py | 22 +++++++++ 2 files changed, 131 insertions(+), 1 deletion(-) (limited to 'synapse/util') diff --git a/synapse/metrics/__init__.py b/synapse/metrics/__init__.py index 550f8443f7..98333b2048 100644 --- a/synapse/metrics/__init__.py +++ b/synapse/metrics/__init__.py @@ -18,8 +18,11 @@ import gc import logging import os import platform +import threading import time +import six + import attr from prometheus_client import Counter, Gauge, Histogram from prometheus_client.core import REGISTRY, GaugeMetricFamily @@ -68,7 +71,7 @@ class LaterGauge(object): return if isinstance(calls, dict): - for k, v in calls.items(): + for k, v in six.iteritems(calls): g.add_metric(k, v) else: g.add_metric([], calls) @@ -87,6 +90,111 @@ class LaterGauge(object): all_gauges[self.name] = self +class InFlightGauge(object): + """Tracks number of things (e.g. requests, Measure blocks, etc) in flight + at any given time. + + Each InFlightGauge will create a metric called `_total` that counts + the number of in flight blocks, as well as a metrics for each item in the + given `sub_metrics` as `_` which will get updated by the + callbacks. + + Args: + name (str) + desc (str) + labels (list[str]) + sub_metrics (list[str]): A list of sub metrics that the callbacks + will update. + """ + + # TODO: Expand this to + + def __init__(self, name, desc, labels, sub_metrics): + self.name = name + self.desc = desc + self.labels = labels + self.sub_metrics = sub_metrics + + # Create a class which have the sub_metrics values as attributes, which + # default to 0 on initialization. Used to pass to registered callbacks. + self._metrics_class = attr.make_class( + "_MetricsEntry", + attrs={x: attr.ib(0) for x in sub_metrics}, + slots=True, + ) + + # Counts number of in flight blocks for a given set of label values + self._registrations = {} + + # Protects access to _registrations + self._lock = threading.Lock() + + self._register_with_collector() + + def register(self, key, callback): + """Registers that we've entered a new block with labels `key`. + + `callback` gets called each time the metrics are collected. The same + value must also be given to `unregister`. + + `callback` gets called with an object that has an attribute per + sub_metric, which should be updated with the necessary values. Note that + the metrics object is shared between all callbacks registered with the + same key. + + Note that `callback` may be called on a separate thread. + """ + with self._lock: + self._registrations.setdefault(key, set()).add(callback) + + def unregister(self, key, callback): + """Registers that we've exited a block with labels `key`. + """ + + with self._lock: + self._registrations.setdefault(key, set()).discard(callback) + + def collect(self): + """Called by prometheus client when it reads metrics. + + Note: may be called by a separate thread. + """ + in_flight = GaugeMetricFamily(self.name + "_total", self.desc, labels=self.labels) + + metrics_by_key = {} + + # We copy so that we don't mutate the list while iterating + with self._lock: + keys = list(self._registrations) + + for key in keys: + with self._lock: + callbacks = set(self._registrations[key]) + + in_flight.add_metric(key, len(callbacks)) + + metrics = self._metrics_class() + metrics_by_key[key] = metrics + for callback in callbacks: + callback(metrics) + + yield in_flight + + for name in self.sub_metrics: + gauge = GaugeMetricFamily("_".join([self.name, name]), "", labels=self.labels) + for key, metrics in six.iteritems(metrics_by_key): + gauge.add_metric(key, getattr(metrics, name)) + yield gauge + + def _register_with_collector(self): + if self.name in all_gauges.keys(): + logger.warning("%s already registered, reregistering" % (self.name,)) + REGISTRY.unregister(all_gauges.pop(self.name)) + + REGISTRY.register(self) + all_gauges[self.name] = self + + # # Detailed CPU metrics # diff --git a/synapse/util/metrics.py b/synapse/util/metrics.py index 97f1267380..4b4ac5f6c7 100644 --- a/synapse/util/metrics.py +++ b/synapse/util/metrics.py @@ -20,6 +20,7 @@ from prometheus_client import Counter from twisted.internet import defer +from synapse.metrics import InFlightGauge from synapse.util.logcontext import LoggingContext logger = logging.getLogger(__name__) @@ -45,6 +46,13 @@ block_db_txn_duration = Counter( block_db_sched_duration = Counter( "synapse_util_metrics_block_db_sched_duration_seconds", "", ["block_name"]) +# Tracks the number of blocks currently active +in_flight = InFlightGauge( + "synapse_util_metrics_block_in_flight", "", + labels=["block_name"], + sub_metrics=["real_time_max", "real_time_sum"], +) + def measure_func(name): def wrapper(func): @@ -82,10 +90,14 @@ class Measure(object): self.start_usage = self.start_context.get_resource_usage() + in_flight.register((self.name,), self._update_in_flight) + def __exit__(self, exc_type, exc_val, exc_tb): if isinstance(exc_type, Exception) or not self.start_context: return + in_flight.unregister((self.name,), self._update_in_flight) + duration = self.clock.time() - self.start block_counter.labels(self.name).inc() @@ -120,3 +132,13 @@ class Measure(object): if self.created_context: self.start_context.__exit__(exc_type, exc_val, exc_tb) + + def _update_in_flight(self, metrics): + """Gets called when processing in flight metrics + """ + duration = self.clock.time() - self.start + + metrics.real_time_max = max(metrics.real_time_max, duration) + metrics.real_time_sum += duration + + # TODO: Add other in flight metrics. -- cgit 1.4.1 From fcfe7a850dada2e65982f2bc805d8bc409f07512 Mon Sep 17 00:00:00 2001 From: Erik Johnston Date: Fri, 14 Sep 2018 19:23:07 +0100 Subject: Add an awful secondary timeout to fix wedged requests This is an attempt to mitigate #3842 by adding yet-another-timeout --- synapse/http/matrixfederationclient.py | 11 ++++++++ synapse/util/async_helpers.py | 51 ++++++++++++++++++++++++++++++++++ 2 files changed, 62 insertions(+) (limited to 'synapse/util') diff --git a/synapse/http/matrixfederationclient.py b/synapse/http/matrixfederationclient.py index da16b5dd8c..13b19f7626 100644 --- a/synapse/http/matrixfederationclient.py +++ b/synapse/http/matrixfederationclient.py @@ -42,6 +42,7 @@ from synapse.api.errors import ( ) from synapse.http.endpoint import matrix_federation_endpoint from synapse.util import logcontext +from synapse.util.async_helpers import timeout_no_seriously from synapse.util.logcontext import make_deferred_yieldable from synapse.util.metrics import Measure @@ -228,6 +229,16 @@ class MatrixFederationHttpClient(object): ) request_deferred.addTimeout(_sec_timeout, self.hs.get_reactor()) + # Sometimes the timeout above doesn't work, so lets hack yet + # another layer of timeouts in in the vain hope that at some + # point the world made sense and this really really really + # should work. + request_deferred = timeout_no_seriously( + request_deferred, + timeout=_sec_timeout * 2, + reactor=self.hs.get_reactor(), + ) + with Measure(self.clock, "outbound_request"): response = yield make_deferred_yieldable( request_deferred, diff --git a/synapse/util/async_helpers.py b/synapse/util/async_helpers.py index 9b3f2f4b96..083e4f4128 100644 --- a/synapse/util/async_helpers.py +++ b/synapse/util/async_helpers.py @@ -438,3 +438,54 @@ def _cancelled_to_timed_out_error(value, timeout): value.trap(CancelledError) raise DeferredTimeoutError(timeout, "Deferred") return value + + +def timeout_no_seriously(deferred, timeout, reactor): + """The in build twisted deferred addTimeout (and the method above) + completely fail to time things out under some unknown circumstances. + + Lets try a different way of timing things out and maybe that will make + things work?! + + TODO: Kill this with fire. + """ + + new_d = defer.Deferred() + + timed_out = [False] + + def time_it_out(): + timed_out[0] = True + deferred.cancel() + + if not new_d.called: + new_d.errback(DeferredTimeoutError(timeout, "Deferred")) + + delayed_call = reactor.callLater(timeout, time_it_out) + + def convert_cancelled(value): + if timed_out[0]: + return _cancelled_to_timed_out_error(value, timeout) + return value + + deferred.addBoth(convert_cancelled) + + def cancel_timeout(result): + # stop the pending call to cancel the deferred if it's been fired + if delayed_call.active(): + delayed_call.cancel() + return result + + deferred.addBoth(cancel_timeout) + + def success_cb(val): + if not new_d.called: + new_d.callback(val) + + def failure_cb(val): + if not new_d.called: + new_d.errback(val) + + deferred.addCallbacks(success_cb, failure_cb) + + return new_d -- cgit 1.4.1 From 24efb2a70dfd8aa100507612836601991e11affc Mon Sep 17 00:00:00 2001 From: Erik Johnston Date: Sat, 15 Sep 2018 11:38:39 +0100 Subject: Fix timeout function Turns out deferred.cancel sometimes throws, so we do that last to ensure that we always do resolve the new deferred. --- synapse/util/async_helpers.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'synapse/util') diff --git a/synapse/util/async_helpers.py b/synapse/util/async_helpers.py index 083e4f4128..40c2946129 100644 --- a/synapse/util/async_helpers.py +++ b/synapse/util/async_helpers.py @@ -456,11 +456,12 @@ def timeout_no_seriously(deferred, timeout, reactor): def time_it_out(): timed_out[0] = True - deferred.cancel() if not new_d.called: new_d.errback(DeferredTimeoutError(timeout, "Deferred")) + deferred.cancel() + delayed_call = reactor.callLater(timeout, time_it_out) def convert_cancelled(value): -- cgit 1.4.1