)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":8788,"name":"Miguel Angel Ajo","email":"mangelajo@redhat.com","username":"mangelajo"},"change_message_id":"003387a7c8ac3de044443fc396b201e2e057d016","unresolved":false,"context_lines":[{"line_number":30,"context_line":"Note that this only affects initial synchronization time, not failover"},{"line_number":31,"context_line":"times. Failover times will continue to be fast because the ports"},{"line_number":32,"context_line":"are wired ahead of time, the initial GARP after the state transition"},{"line_number":33,"context_line":"to MASTER will be sent properly."},{"line_number":34,"context_line":""},{"line_number":35,"context_line":"Change-Id: I7a086472b8742828dae08ffd915c45e94fb4b94e"},{"line_number":36,"context_line":"Closes-Bug: #1520517"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"da6ed579_f5c41160","line":33,"range":{"start_line":33,"start_character":9,"end_line":33,"end_character":10},"updated":"2016-01-18 10:30:22.000000000","message":"Ok, it seems reasonable to me, still we\u0027re open to 1453855 in some situations when the l2 agent is very slow due to load for some reason.\n\nWould it make sense to introduce a second patch, reintroducing the current behaviour, but based on a config flag on the l3-agent.ini?,\n\nThis way it\u0027s up to the deployer to pick what suits him better, even to configure the delays, and repeat times if they will.","commit_id":"303cbc6b5b745aadf026627c4cf2e90e3862b586"},{"author":{"_account_id":8788,"name":"Miguel Angel Ajo","email":"mangelajo@redhat.com","username":"mangelajo"},"change_message_id":"22bd425085f76267faa3da596bc726a28acbe476","unresolved":false,"context_lines":[{"line_number":30,"context_line":"Note that this only affects initial synchronization time, not failover"},{"line_number":31,"context_line":"times. Failover times will continue to be fast because the ports"},{"line_number":32,"context_line":"are wired ahead of time, the initial GARP after the state transition"},{"line_number":33,"context_line":"to MASTER will be sent properly."},{"line_number":34,"context_line":""},{"line_number":35,"context_line":"Change-Id: I7a086472b8742828dae08ffd915c45e94fb4b94e"},{"line_number":36,"context_line":"Closes-Bug: #1520517"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"ba63e191_0e820801","line":33,"range":{"start_line":33,"start_character":9,"end_line":33,"end_character":10},"in_reply_to":"ba63e191_c7e29c9f","updated":"2016-01-19 10:14:27.000000000","message":"A follow patch seems reasonable, what you said makes sense.\n\nBtw, aren\u0027t config values defaulting to current behaviour backportable?","commit_id":"303cbc6b5b745aadf026627c4cf2e90e3862b586"},{"author":{"_account_id":6579,"name":"Nir Magnezi","email":"nmagnezi@redhat.com","username":"nmagnezi"},"change_message_id":"080c02e52b824335304ed0500c7aa1ba04deeaa1","unresolved":false,"context_lines":[{"line_number":30,"context_line":"Note that this only affects initial synchronization time, not failover"},{"line_number":31,"context_line":"times. Failover times will continue to be fast because the ports"},{"line_number":32,"context_line":"are wired ahead of time, the initial GARP after the state transition"},{"line_number":33,"context_line":"to MASTER will be sent properly."},{"line_number":34,"context_line":""},{"line_number":35,"context_line":"Change-Id: I7a086472b8742828dae08ffd915c45e94fb4b94e"},{"line_number":36,"context_line":"Closes-Bug: #1520517"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"ba63e191_c7e29c9f","line":33,"range":{"start_line":33,"start_character":9,"end_line":33,"end_character":10},"in_reply_to":"ba63e191_d4b499d2","updated":"2016-01-19 07:51:09.000000000","message":"valid point.","commit_id":"303cbc6b5b745aadf026627c4cf2e90e3862b586"},{"author":{"_account_id":8873,"name":"Assaf Muller","email":"amuller@redhat.com","username":"amuller"},"change_message_id":"bc31675fcadfe18b52af330f81220d832ee77fdd","unresolved":false,"context_lines":[{"line_number":30,"context_line":"Note that this only affects initial synchronization time, not failover"},{"line_number":31,"context_line":"times. Failover times will continue to be fast because the ports"},{"line_number":32,"context_line":"are wired ahead of time, the initial GARP after the state transition"},{"line_number":33,"context_line":"to MASTER will be sent properly."},{"line_number":34,"context_line":""},{"line_number":35,"context_line":"Change-Id: I7a086472b8742828dae08ffd915c45e94fb4b94e"},{"line_number":36,"context_line":"Closes-Bug: #1520517"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"ba63e191_d4b499d2","line":33,"range":{"start_line":33,"start_character":9,"end_line":33,"end_character":10},"in_reply_to":"da6ed579_3d04c27c","updated":"2016-01-18 22:28:23.000000000","message":"This patch needs to be backportable, so we can introduce new configuration options in a follow up patch.","commit_id":"303cbc6b5b745aadf026627c4cf2e90e3862b586"},{"author":{"_account_id":6579,"name":"Nir Magnezi","email":"nmagnezi@redhat.com","username":"nmagnezi"},"change_message_id":"fa323473e9618894396520868d2850aeaab6ebf0","unresolved":false,"context_lines":[{"line_number":30,"context_line":"Note that this only affects initial synchronization time, not failover"},{"line_number":31,"context_line":"times. Failover times will continue to be fast because the ports"},{"line_number":32,"context_line":"are wired ahead of time, the initial GARP after the state transition"},{"line_number":33,"context_line":"to MASTER will be sent properly."},{"line_number":34,"context_line":""},{"line_number":35,"context_line":"Change-Id: I7a086472b8742828dae08ffd915c45e94fb4b94e"},{"line_number":36,"context_line":"Closes-Bug: #1520517"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"da6ed579_3d04c27c","line":33,"range":{"start_line":33,"start_character":9,"end_line":33,"end_character":10},"in_reply_to":"da6ed579_f5c41160","updated":"2016-01-18 14:30:42.000000000","message":"+1. While I\u0027m not a fond of yet another configurable parameter, I tend to think we cannot find a single value to match everyone.","commit_id":"303cbc6b5b745aadf026627c4cf2e90e3862b586"}],"neutron/agent/linux/keepalived.py":[{"author":{"_account_id":8788,"name":"Miguel Angel Ajo","email":"mangelajo@redhat.com","username":"mangelajo"},"change_message_id":"003387a7c8ac3de044443fc396b201e2e057d016","unresolved":false,"context_lines":[{"line_number":32,"context_line":"# TODO(amuller): Use L3 agent constant when new constants module is introduced."},{"line_number":33,"context_line":"FIP_LL_SUBNET \u003d \u0027169.254.30.0/23\u0027"},{"line_number":34,"context_line":"KEEPALIVED_SERVICE_NAME \u003d \u0027keepalived\u0027"},{"line_number":35,"context_line":"GARP_MASTER_DELAY \u003d 60"},{"line_number":36,"context_line":""},{"line_number":37,"context_line":"LOG \u003d logging.getLogger(__name__)"},{"line_number":38,"context_line":""}],"source_content_type":"text/x-python","patch_set":1,"id":"da6ed579_b5d45947","line":35,"updated":"2016-01-18 10:30:22.000000000","message":"Isn\u0027t it big enough to possibly/eventually introduce SSH timeout errors in tempest tests?","commit_id":"303cbc6b5b745aadf026627c4cf2e90e3862b586"},{"author":{"_account_id":8788,"name":"Miguel Angel Ajo","email":"mangelajo@redhat.com","username":"mangelajo"},"change_message_id":"22bd425085f76267faa3da596bc726a28acbe476","unresolved":false,"context_lines":[{"line_number":32,"context_line":"# TODO(amuller): Use L3 agent constant when new constants module is introduced."},{"line_number":33,"context_line":"FIP_LL_SUBNET \u003d \u0027169.254.30.0/23\u0027"},{"line_number":34,"context_line":"KEEPALIVED_SERVICE_NAME \u003d \u0027keepalived\u0027"},{"line_number":35,"context_line":"GARP_MASTER_DELAY \u003d 60"},{"line_number":36,"context_line":""},{"line_number":37,"context_line":"LOG \u003d logging.getLogger(__name__)"},{"line_number":38,"context_line":""}],"source_content_type":"text/x-python","patch_set":1,"id":"ba63e191_6ee2a4d7","line":35,"in_reply_to":"ba63e191_54aae971","updated":"2016-01-19 10:14:27.000000000","message":"Right. It\u0027s not the VM port, it\u0027s the router port. But I wonder how it plays with floating ips.","commit_id":"303cbc6b5b745aadf026627c4cf2e90e3862b586"},{"author":{"_account_id":8873,"name":"Assaf Muller","email":"amuller@redhat.com","username":"amuller"},"change_message_id":"bc31675fcadfe18b52af330f81220d832ee77fdd","unresolved":false,"context_lines":[{"line_number":32,"context_line":"# TODO(amuller): Use L3 agent constant when new constants module is introduced."},{"line_number":33,"context_line":"FIP_LL_SUBNET \u003d \u0027169.254.30.0/23\u0027"},{"line_number":34,"context_line":"KEEPALIVED_SERVICE_NAME \u003d \u0027keepalived\u0027"},{"line_number":35,"context_line":"GARP_MASTER_DELAY \u003d 60"},{"line_number":36,"context_line":""},{"line_number":37,"context_line":"LOG \u003d logging.getLogger(__name__)"},{"line_number":38,"context_line":""}],"source_content_type":"text/x-python","patch_set":1,"id":"ba63e191_54aae971","line":35,"in_reply_to":"da6ed579_b5d45947","updated":"2016-01-18 22:28:23.000000000","message":"It shouldn\u0027t as I expect under normal operation (Not booting the node) the OVS agent to wire the ports before keepalived sends out the initial GARP.","commit_id":"303cbc6b5b745aadf026627c4cf2e90e3862b586"}]}
