)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":748,"name":"Armando Migliaccio","email":"armamig@gmail.com","username":"armando-migliaccio"},"change_message_id":"4ed5be215d91354e9328fb10368e7c2d790d8392","unresolved":false,"context_lines":[{"line_number":14,"context_line":"and raising."},{"line_number":15,"context_line":""},{"line_number":16,"context_line":"Closes-Bug: #1618987"},{"line_number":17,"context_line":"Closes-Bug: #1623732"},{"line_number":18,"context_line":"Change-Id: Ia73a9be047093c02f61e3e9ce13d98dccd49dfeb"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":2,"id":"7a8ec9b2_dfd014f9","line":17,"updated":"2016-09-15 20:29:53.000000000","message":"Technically this is a partial for bug 1623732. We discovered there\u0027s a hole in the exception handling logic for router failures.","commit_id":"481d5b6dfa09dca383467872131ca782fb3a03cf"}],"neutron/agent/linux/interface.py":[{"author":{"_account_id":748,"name":"Armando Migliaccio","email":"armamig@gmail.com","username":"armando-migliaccio"},"change_message_id":"35c2d33e361e6870dfba9505307fe1f46e00b820","unresolved":false,"context_lines":[{"line_number":320,"context_line":"            ns_dev \u003d ip.device(device_name)"},{"line_number":321,"context_line":""},{"line_number":322,"context_line":"        internal \u003d not self.conf.ovs_use_veth"},{"line_number":323,"context_line":"        self._ovs_add_port(bridge, tap_name, port_id, mac_address,"},{"line_number":324,"context_line":"                           internal\u003dinternal)"},{"line_number":325,"context_line":"        for i in range(9):"},{"line_number":326,"context_line":"            # workaround for the OVS shy port syndrome. ports sometimes"}],"source_content_type":"text/x-python","patch_set":3,"id":"7a8ec9b2_8ba3a7ff","line":323,"updated":"2016-09-15 23:19:19.000000000","message":"I wonder if a better fix would be to restore the calls that got removed in [1], so that we are more defensive about this.\n\n[1] https://review.openstack.org/#/c/344859/13/neutron/agent/common/ovs_lib.py","commit_id":"a75ce6850f3954edafbb0c128750e39b57875743"},{"author":{"_account_id":7448,"name":"Carl Baldwin","email":"carl@ecbaldwin.net","username":"carl-baldwin"},"change_message_id":"b9da5d76d14403c159e303266a1064ab4ab0ab16","unresolved":false,"context_lines":[{"line_number":322,"context_line":"        internal \u003d not self.conf.ovs_use_veth"},{"line_number":323,"context_line":"        self._ovs_add_port(bridge, tap_name, port_id, mac_address,"},{"line_number":324,"context_line":"                           internal\u003dinternal)"},{"line_number":325,"context_line":"        for i in range(9):"},{"line_number":326,"context_line":"            # workaround for the OVS shy port syndrome. ports sometimes"},{"line_number":327,"context_line":"            # hide for a bit right after they are first created."},{"line_number":328,"context_line":"            # see bug/1618987"}],"source_content_type":"text/x-python","patch_set":3,"id":"7a8ec9b2_dfb1f148","line":325,"updated":"2016-09-15 23:33:27.000000000","message":"Don\u0027t we have something for this kind of retry logic? (retrying.retry maybe?) I wouldn\u0027t insist on changing this now. But, it might be nice to pull the retry logic apart from the actual call to set_address.","commit_id":"a75ce6850f3954edafbb0c128750e39b57875743"},{"author":{"_account_id":748,"name":"Armando Migliaccio","email":"armamig@gmail.com","username":"armando-migliaccio"},"change_message_id":"133b934b047eb9209433ab0d793a6a09d1e52deb","unresolved":false,"context_lines":[{"line_number":322,"context_line":"        internal \u003d not self.conf.ovs_use_veth"},{"line_number":323,"context_line":"        self._ovs_add_port(bridge, tap_name, port_id, mac_address,"},{"line_number":324,"context_line":"                           internal\u003dinternal)"},{"line_number":325,"context_line":"        for i in range(9):"},{"line_number":326,"context_line":"            # workaround for the OVS shy port syndrome. ports sometimes"},{"line_number":327,"context_line":"            # hide for a bit right after they are first created."},{"line_number":328,"context_line":"            # see bug/1618987"}],"source_content_type":"text/x-python","patch_set":3,"id":"7a8ec9b2_56b866c8","line":325,"in_reply_to":"7a8ec9b2_ab9ce35d","updated":"2016-09-16 00:10:50.000000000","message":"Talked to Kevin. Otherwiseguy managed to reproduce the underlying ovs issue so a real fix is coming.","commit_id":"a75ce6850f3954edafbb0c128750e39b57875743"},{"author":{"_account_id":748,"name":"Armando Migliaccio","email":"armamig@gmail.com","username":"armando-migliaccio"},"change_message_id":"08a7ea6459f6d4f15607cfcf3a81f618721548ec","unresolved":false,"context_lines":[{"line_number":322,"context_line":"        internal \u003d not self.conf.ovs_use_veth"},{"line_number":323,"context_line":"        self._ovs_add_port(bridge, tap_name, port_id, mac_address,"},{"line_number":324,"context_line":"                           internal\u003dinternal)"},{"line_number":325,"context_line":"        for i in range(9):"},{"line_number":326,"context_line":"            # workaround for the OVS shy port syndrome. ports sometimes"},{"line_number":327,"context_line":"            # hide for a bit right after they are first created."},{"line_number":328,"context_line":"            # see bug/1618987"}],"source_content_type":"text/x-python","patch_set":3,"id":"7a8ec9b2_ab9ce35d","line":325,"in_reply_to":"7a8ec9b2_dfb1f148","updated":"2016-09-15 23:35:59.000000000","message":"I think this is what the existing logic, linked above in my comment used to have.","commit_id":"a75ce6850f3954edafbb0c128750e39b57875743"}]}
