refactor: log-to-syslog in debug_flags will log internal trace msgs to syslog [citest_skip]

If you set __network_debug_flags=log-to-syslog then the internal trace messages which
are returned by the module will also be logged to syslog.  This is useful for when
the network role hangs and times out and is killed before those messages can
be returned.  The trace messages printed to syslog can help debug problems.

Signed-off-by: Rich Megginson <rmeggins@redhat.com>

Author: richm

library/network_connections.py

@@ -1822,6 +1822,8 @@ def log(
             raise AssertionError("idx {0} is less than -1".format(idx))
         self._log_idx += 1
         self.run_results[idx]["log"].append((severity, msg, self._log_idx))
+        if "log-to-syslog" in self.module.params["__debug_flags"]:
+            self.module.log("%s: %s: %s" % (LogLevel.fmt(severity), idx, msg))
         if severity == LogLevel.ERROR:
             if force_fail or not ignore_errors:
                 self.fail_json(
@@ -2019,13 +2021,17 @@ def log_fatal(self, idx, msg, warn_traceback=False):
         )
 
     def log(self, idx, severity, msg, warn_traceback=False, force_fail=False):
+        if len(self.connections) > idx:
+            ignore_errors = self.connection_ignore_errors(self.connections[idx])
+        else:
+            ignore_errors = self._ignore_errors
         self.run_env.log(
             self.connections,
             idx,
             severity,
             msg,
             is_changed=self.is_changed_modified_system,
-            ignore_errors=self.connection_ignore_errors(self.connections[idx]),
+            ignore_errors=ignore_errors,
             warn_traceback=warn_traceback,
             force_fail=force_fail,
         )
@@ -2125,7 +2131,9 @@ def run(self):
         self.run_prepare()
         while self.check_mode_next() != CheckMode.DONE:
             if self.check_mode == CheckMode.REAL_RUN:
+                self.log(0, LogLevel.INFO, "Starting transaction")
                 self.start_transaction()
+                self.log(0, LogLevel.INFO, "Started transaction")
 
             # Reasoning for this order:
             # For down/up profiles might need to be present, so do this first
@@ -2146,11 +2154,15 @@ def run(self):
                                 self.run_action_down(idx)
                     except Exception as error:
                         if self.check_mode == CheckMode.REAL_RUN:
+                            self.log(idx, LogLevel.INFO, "Rolling back transaction")
                             self.rollback_transaction(idx, action, error)
+                            self.log(0, LogLevel.INFO, "Rolled back transaction")
                         raise
 
             if self.check_mode == CheckMode.REAL_RUN:
+                self.log(0, LogLevel.INFO, "Finishing transaction")
                 self.finish_transaction()
+                self.log(0, LogLevel.INFO, "Finished transaction")
 
     def run_prepare(self):
         for idx, connection in enumerate(self.connections):
@@ -2912,6 +2924,7 @@ def main():
             changed=(cmd is not None and cmd.is_changed_modified_system),
             warn_traceback=not isinstance(e, MyError),
         )
+    cmd.log(0, LogLevel.INFO, "Exiting")
     run_env_ansible.exit_json(
         connections, changed=(cmd is not None and cmd.is_changed_modified_system)
     )

tests/playbooks/tests_route_device.yml

@@ -131,8 +131,8 @@
             the output device is logged for initscripts provider
           assert:
             that:
-              - __network_connections_result.stderr is search("\[003\]
-                <warn>  .0, state.None persistent_state.present,
+              - __network_connections_result.stderr is search("<warn>  .0,
+                state.None persistent_state.present,
                 '{{ interface1 }}'. The connection {{ interface1 }} does not
                 specify an interface name. Therefore, the route to
                 198.58.10.64/26 will be configured without the output device