)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":1131,"name":"Brian Haley","email":"haleyb.dev@gmail.com","username":"brian-haley"},"change_message_id":"8baab634d187b6eddc000d9bd766b3176bed3a81","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"b40432da_e8649bd8","updated":"2024-04-19 14:18:43.000000000","message":"At a minimum this will need a test.","commit_id":"d09c82661c252eefc7cfebc98c474c1d9b801ff9"},{"author":{"_account_id":1131,"name":"Brian Haley","email":"haleyb.dev@gmail.com","username":"brian-haley"},"change_message_id":"3d25f09d2789399aae7459d933163d4fda15ad5b","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"69f9a319_a7261ad6","updated":"2024-09-09 03:35:29.000000000","message":"recheck get recent logs","commit_id":"d09c82661c252eefc7cfebc98c474c1d9b801ff9"},{"author":{"_account_id":28056,"name":"Liushy","email":"liuxie_11@163.com","username":"liuxie_11"},"change_message_id":"3dde80c5c75a8f2e77d4b2c5ca9301321dfd4f7a","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"601e9d5c_4ba5ea17","in_reply_to":"b40432da_e8649bd8","updated":"2024-04-24 04:02:11.000000000","message":"I can see there is already a test:  MetadataAgentHealthEvent\nhttps://github.com/openstack/neutron/blob/484b04cf019e9e5fcc2b429ef916c54211c3e38c/neutron/tests/functional/agent/ovn/metadata/test_metadata_agent.py#L46","commit_id":"d09c82661c252eefc7cfebc98c474c1d9b801ff9"},{"author":{"_account_id":5756,"name":"Terry Wilson","email":"twilson@redhat.com","username":"otherwiseguy"},"change_message_id":"018516c95412d074ffd982f2b6bbb022d75b22fa","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"a13b14f3_fee433da","updated":"2024-12-06 22:28:16.000000000","message":"I\u0027m having some trouble understanding the situation this is trying to resolve. Is it that in some cases ovsdb-server is busy and is not responding, and in these cases you think that the agent shouldn\u0027t be reporting \"down\" because the update failed? In general when an ovsdb update transaction is made, it *should* continue through ovsdb-server hiccups, or even restarts. of ovsdb-server as long as the txn timeout has not been reached.\n\nIf this isn\u0027t happening, then it is a bug either in ovsdbapp or python-ovs that we need to investigate. I know that @froyo@redhat.com is currently looking at an issue where it seems like the o-api service is getting blocked/txns fail while a txn waits on the transaction timeout to expire.\n\nBut in general, python-ovs should loop through and handle any connection issues/delays while waiting for a transaction to commit. And if the transaction timeout is exceeded, only then should there be an error. And you should really make sure your transaction timeout and agent_down_time work well together.","commit_id":"a0ba5b2b74a715e236cb7f3da7e1509670b51039"}]}