)]}'
{"neutron/wsgi.py":[{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"54c60b1750e816eede7815ae9944a27784ff795b","unresolved":false,"context_lines":[{"line_number":211,"context_line":"            # For the case where only one process is required."},{"line_number":212,"context_line":"            self._server \u003d self.pool.spawn(self._run, application,"},{"line_number":213,"context_line":"                                           self._socket)"},{"line_number":214,"context_line":"            systemd.notify_once()"},{"line_number":215,"context_line":"        else:"},{"line_number":216,"context_line":"            # Minimize the cost of checking for child exit by extending the"},{"line_number":217,"context_line":"            # wait interval past the default of 0.01s."}],"source_content_type":"text/x-python","patch_set":2,"id":"1ae5cdf2_6e7b88f4","line":214,"updated":"2014-06-17 17:50:35.000000000","message":"Maybe a dumb question but: why don\u0027t we just use ProcessLauncher to spawn a single worker and reuse any additional services provided to us by service module, like that systemd notification?","commit_id":"09856a0cf63052045b3d9f4c8d2a5aa920c7c935"},{"author":{"_account_id":2035,"name":"Maru Newby","email":"marun@redhat.com","username":"maru"},"change_message_id":"0673498310feae9b5300d4ad4978809eb2b41311","unresolved":false,"context_lines":[{"line_number":211,"context_line":"            # For the case where only one process is required."},{"line_number":212,"context_line":"            self._server \u003d self.pool.spawn(self._run, application,"},{"line_number":213,"context_line":"                                           self._socket)"},{"line_number":214,"context_line":"            systemd.notify_once()"},{"line_number":215,"context_line":"        else:"},{"line_number":216,"context_line":"            # Minimize the cost of checking for child exit by extending the"},{"line_number":217,"context_line":"            # wait interval past the default of 0.01s."}],"source_content_type":"text/x-python","patch_set":2,"id":"1ae5cdf2_c3be25ae","line":214,"in_reply_to":"1ae5cdf2_6e7b88f4","updated":"2014-06-17 22:32:52.000000000","message":"I\u0027m assuming that using ProcessLauncher is undesirable because it adds the memory overhead of an additional process rather than running the wsgi server in-process.  If the goal is not to use workers (i.e. be single-process), adding a call to systemd would seem a small price to pay.","commit_id":"09856a0cf63052045b3d9f4c8d2a5aa920c7c935"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"ec50496567d338047979cdc38d7d0978fc4780dd","unresolved":false,"context_lines":[{"line_number":211,"context_line":"            # For the case where only one process is required."},{"line_number":212,"context_line":"            self._server \u003d self.pool.spawn(self._run, application,"},{"line_number":213,"context_line":"                                           self._socket)"},{"line_number":214,"context_line":"            systemd.notify_once()"},{"line_number":215,"context_line":"        else:"},{"line_number":216,"context_line":"            # Minimize the cost of checking for child exit by extending the"},{"line_number":217,"context_line":"            # wait interval past the default of 0.01s."}],"source_content_type":"text/x-python","patch_set":2,"id":"1ae5cdf2_95be01ad","line":214,"in_reply_to":"1ae5cdf2_c3be25ae","updated":"2014-06-18 08:35:35.000000000","message":"On the bright side, reusing ProcessLauncher would also give us automatic respawn of the main process in case of failure.\n\nBut ok, I agree this is not really a question for this particular patch.","commit_id":"09856a0cf63052045b3d9f4c8d2a5aa920c7c935"}]}