Remove full_twisted_stacktraces option

The debug 'full_twisted_stacktraces' flag caused synapse to rewrite
twisted deferreds to always fire the callback on the next reactor tick.
This was to force the deferred to always store the stacktraces on
exceptions, and thus be more likely to have a full stacktrace when it
reaches the final error handlers and gets printed to the logs.

Dynamically rewriting things is generally bad, and in particular this
change violates assumptions of various bits of Twisted. This wouldn't
necessarily be so bad, but it turns out this option has been turned on
on some production servers.

Turning the option can cause e.g. #1778.

For now, lets just entirely nuke this option.

1 files changed, 0 insertions, 8 deletions

diff --git a/synapse/config/logger.py b/synapse/config/logger.py
index 63e69a7e0c..77ded0ad25 100644
--- a/synapse/config/logger.py
+++ b/synapse/config/logger.py
@@ -22,7 +22,6 @@ import yaml
 from string import Template
 import os
 import signal
-from synapse.util.debug import debug_deferreds
 
 
 DEFAULT_LOG_CONFIG = Template("""
@@ -71,8 +70,6 @@ class LoggingConfig(Config):
         self.verbosity = config.get("verbose", 0)
         self.log_config = self.abspath(config.get("log_config"))
         self.log_file = self.abspath(config.get("log_file"))
-        if config.get("full_twisted_stacktraces"):
-            debug_deferreds()
 
     def default_config(self, config_dir_path, server_name, **kwargs):
         log_file = self.abspath("homeserver.log")
@@ -88,11 +85,6 @@ class LoggingConfig(Config):
 
         # A yaml python logging config file
         log_config: "%(log_config)s"
-
-        # Stop twisted from discarding the stack traces of exceptions in
-        # deferreds by waiting a reactor tick before running a deferred's
-        # callbacks.
-        # full_twisted_stacktraces: true
         """ % locals()
 
     def read_arguments(self, args):