)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":1653,"name":"garyk","email":"gkotton@vmware.com","username":"garyk"},"change_message_id":"e62463bbe071f1e76c568ae48561301ad0d74de0","unresolved":false,"context_lines":[{"line_number":12,"context_line":"router is reconfigured. However, it tries to reconfigure the"},{"line_number":13,"context_line":"cached router. This is a change of behavior from the fullsync"},{"line_number":14,"context_line":"days. The retry is more likely to succeed if the"},{"line_number":15,"context_line":"router is retrieved from the server, instead of using"},{"line_number":16,"context_line":"the locally cached version, in case the user or operator"},{"line_number":17,"context_line":"fixed bad input, or if the router was retrieved in a bad"},{"line_number":18,"context_line":"state due to a server-side race condition."}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"3a7e1126_b61dc163","line":15,"updated":"2015-12-20 09:50:43.000000000","message":"why is this more likely to succeed? Is there missing information? Could the cached version be out of sync?","commit_id":"822ad5f06bcef8f95f36032d4fd4709975cecc31"},{"author":{"_account_id":8873,"name":"Assaf Muller","email":"amuller@redhat.com","username":"amuller"},"change_message_id":"bcacae91b004c6fcd8e3f3bf4361684599e2d3e0","unresolved":false,"context_lines":[{"line_number":12,"context_line":"router is reconfigured. However, it tries to reconfigure the"},{"line_number":13,"context_line":"cached router. This is a change of behavior from the fullsync"},{"line_number":14,"context_line":"days. The retry is more likely to succeed if the"},{"line_number":15,"context_line":"router is retrieved from the server, instead of using"},{"line_number":16,"context_line":"the locally cached version, in case the user or operator"},{"line_number":17,"context_line":"fixed bad input, or if the router was retrieved in a bad"},{"line_number":18,"context_line":"state due to a server-side race condition."}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"3a7e1126_f2959286","line":15,"in_reply_to":"3a7e1126_b61dc163","updated":"2015-12-20 16:30:21.000000000","message":"Well I named two possible reasons in lines 16-18 \u003dp\n\nThe server shouldn\u0027t be sending bad data, but it has happened in the past, and the agent should know how to recover. For example if the user created a router with an IP address outside of the subnet, he sees exceptions in the L3 agent and corrects his mistake - Without this patch the agent won\u0027t retrieve the updated info from the server. Another example is that we\u0027re currently in the middle of fixing a bug with HA routers where sometimes due to a race condition the agent asks the server for info about a router which hasn\u0027t yet created its HA port, and so the agent retrieves it without, which fails to configure. The agent should be able to recover from that.","commit_id":"822ad5f06bcef8f95f36032d4fd4709975cecc31"}]}
