)]}'
{"oslo_db/sqlalchemy/models.py":[{"author":{"_account_id":28522,"name":"Hervé Beraud","email":"herveberaud.pro@gmail.com","username":"hberaud"},"change_message_id":"48e24c867ded7c08547a1fbef3a6462f7d1bea09","unresolved":true,"context_lines":[{"line_number":35,"context_line":"    def save(self, session):"},{"line_number":36,"context_line":"        \"\"\"Save this object.\"\"\""},{"line_number":37,"context_line":""},{"line_number":38,"context_line":"        # NOTE(boris-42): This part of code should be look like:"},{"line_number":39,"context_line":"        #                       session.add(self)"},{"line_number":40,"context_line":"        #                       session.flush()"},{"line_number":41,"context_line":"        #                 But there is a bug in sqlalchemy and eventlet that"},{"line_number":42,"context_line":"        #                 raises NoneType exception if there is no running"},{"line_number":43,"context_line":"        #                 transaction and rollback is called. As long as"},{"line_number":44,"context_line":"        #                 sqlalchemy has this bug we have to create transaction"},{"line_number":45,"context_line":"        #                 explicitly."},{"line_number":46,"context_line":"        with session.begin(subtransactions\u003dTrue):"},{"line_number":47,"context_line":"            session.add(self)"},{"line_number":48,"context_line":"            session.flush()"}],"source_content_type":"text/x-python","patch_set":5,"id":"d6cac2d2_98602060","side":"PARENT","line":45,"range":{"start_line":38,"start_character":8,"end_line":45,"end_character":37},"updated":"2021-08-11 12:57:44.000000000","message":"This message is more than 7 years old, it was added before oslo incubator.\nI would argue that it is now fixed but I\u0027ll let Mike confirm that point.","commit_id":"b3a56b35640b0cfceeeb97c885318a3aae7f7b22"},{"author":{"_account_id":11816,"name":"mike_mp@zzzcomputing.com","display_name":"Mike Bayer","email":"mike_mp@zzzcomputing.com","username":"zzzeek","status":"Red Hat"},"change_message_id":"e92859bf5e6427a229f3f1092be54ea88a033f2b","unresolved":false,"context_lines":[{"line_number":35,"context_line":"    def save(self, session):"},{"line_number":36,"context_line":"        \"\"\"Save this object.\"\"\""},{"line_number":37,"context_line":""},{"line_number":38,"context_line":"        # NOTE(boris-42): This part of code should be look like:"},{"line_number":39,"context_line":"        #                       session.add(self)"},{"line_number":40,"context_line":"        #                       session.flush()"},{"line_number":41,"context_line":"        #                 But there is a bug in sqlalchemy and eventlet that"},{"line_number":42,"context_line":"        #                 raises NoneType exception if there is no running"},{"line_number":43,"context_line":"        #                 transaction and rollback is called. As long as"},{"line_number":44,"context_line":"        #                 sqlalchemy has this bug we have to create transaction"},{"line_number":45,"context_line":"        #                 explicitly."},{"line_number":46,"context_line":"        with session.begin(subtransactions\u003dTrue):"},{"line_number":47,"context_line":"            session.add(self)"},{"line_number":48,"context_line":"            session.flush()"}],"source_content_type":"text/x-python","patch_set":5,"id":"42d939c7_811a67a8","side":"PARENT","line":45,"range":{"start_line":38,"start_character":8,"end_line":45,"end_character":37},"in_reply_to":"d6cac2d2_98602060","updated":"2022-02-02 19:35:59.000000000","message":"well the issue is surely related to the fact that they were using autocommit\u003dTrue.   we have certainly fixed a lot of eventlet failure mode bugs and i havent seen any new ones in years.  So to that degree i dont know what bug they were referring towards.  for now I\u0027d be fine with having this go out into tempest and the world and if we see a new such issue, we\u0027ll know what it is.","commit_id":"b3a56b35640b0cfceeeb97c885318a3aae7f7b22"}]}
