)]}'
{"/PATCHSET_LEVEL":[{"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":"aab6ff84bda2c0e70cdd2096edea8ccadb0be097","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"c2c29e1f_a21542fb","updated":"2022-05-31 12:39:16.000000000","message":"I am +1 on adding pre-ping but we have to attend to the existing pre-ping infrastructure here (which is what led to the creation of SQLAlchemy\u0027s pre-ping feature in the first place)","commit_id":"7d50883e998f89fe75aee95127d8e858b8295093"},{"author":{"_account_id":28522,"name":"Hervé Beraud","email":"herveberaud.pro@gmail.com","username":"hberaud"},"change_message_id":"23c23550861b25da01ea1b86c777099e1af00b69","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"a2b5e827_6e78a405","updated":"2022-05-31 09:15:14.000000000","message":"LGTM.\n\nIMO this is a good idea of feature for oslo.db.","commit_id":"7d50883e998f89fe75aee95127d8e858b8295093"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"f546291471d2911c0a1390207b750bbb97f5bb05","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"5683cee0_c20a5e84","updated":"2023-04-17 09:43:24.000000000","message":"We now use the native pre-ping handler when we have SQLAlchemy 2.x, so I don\u0027t think we need this anymore? I\u0027m going to close this on that assumption, but please reopen if this is not the case","commit_id":"7d50883e998f89fe75aee95127d8e858b8295093"},{"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":"89da77cea441118870af8eba5f4aa61df9f91b8c","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"6ebe7938_0b2de055","updated":"2023-04-17 12:57:54.000000000","message":"agree superseded by https://review.opendev.org/c/openstack/oslo.db/+/878146","commit_id":"7d50883e998f89fe75aee95127d8e858b8295093"}],"oslo_db/sqlalchemy/engines.py":[{"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":"aab6ff84bda2c0e70cdd2096edea8ccadb0be097","unresolved":true,"context_lines":[{"line_number":203,"context_line":"    exc_filters.register_engine(engine)"},{"line_number":204,"context_line":""},{"line_number":205,"context_line":"    # register engine connect handler"},{"line_number":206,"context_line":"    event.listen(engine, \"engine_connect\", _connect_ping_listener)"},{"line_number":207,"context_line":""},{"line_number":208,"context_line":"    # initial connect + test"},{"line_number":209,"context_line":"    # NOTE(viktors): the current implementation of _test_connection()"}],"source_content_type":"text/x-python","patch_set":1,"id":"b94171ae_07ed3f61","line":206,"updated":"2022-05-31 12:39:16.000000000","message":"this is already a pre-ping handler here, which is the first version of the \"legacy\" pre ping handler you see at https://docs.sqlalchemy.org/en/14/core/pooling.html#custom-legacy-pessimistic-ping , that eventually became a built in feature.\n\nSo two/three things:\n\n1. if we have pre-ping as an option, it should disable this event listener here and \n\n2. oslo.db sets \"pre-ping\" on unconditionally for all projects via this event handler.   do we think it\u0027s time to use the new \"pre-ping\" exclusively instead, remove the old handler, and therefore we don\u0027t need this option?\n\n3. note that the MySQL pre-ping handler uses connection.ping(), which is a driver-specific feature, and doesn\u0027t do all the \"on disconnect error\" stuff.  this is a good thing, but it\u0027s a change in how this would work.","commit_id":"7d50883e998f89fe75aee95127d8e858b8295093"}]}