)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":4395,"name":"Aaron Rosen","email":"aaronorosen@gmail.com","username":"arosen"},"change_message_id":"2ea204636ec2ea259d8f1856d453468d43b790cf","unresolved":false,"context_lines":[{"line_number":6,"context_line":""},{"line_number":7,"context_line":"VMware: update the default \u0027task_poll_interval\u0027 time"},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"The original menas that each operation against the backend takes at"},{"line_number":10,"context_line":"least 5 seconds. The default is updated to 0.5 seconds."},{"line_number":11,"context_line":""},{"line_number":12,"context_line":"DocImpact"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":2,"id":"AAAAUH%2F9ogo%3D","line":9,"updated":"2014-02-05 14:13:24.000000000","message":"means?","commit_id":"0d50c7b3383330436b0cf3137f676ad8347e7b31"},{"author":{"_account_id":8027,"name":"Vui Lam","email":"vui@vmware.com","username":"vui"},"change_message_id":"f4f44c5ff19199b4f56cbee38b1497b840da0fdc","unresolved":false,"context_lines":[{"line_number":6,"context_line":""},{"line_number":7,"context_line":"VMware: update the default \u0027task_poll_interval\u0027 time"},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"The original means that each operation against the backend takes at"},{"line_number":10,"context_line":"least 5 seconds. The default is updated to 0.5 seconds."},{"line_number":11,"context_line":""},{"line_number":12,"context_line":"DocImpact"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":3,"id":"AAAAUH%2F9bLw%3D","line":9,"updated":"2014-02-07 00:41:01.000000000","message":"To be precise \"each operation against the backend\" should say \"each operation again the backend that returns a task\"","commit_id":"73c87a280e77e03d228d34ab4781ca2e3b02e40e"}],"nova/virt/vmwareapi/driver.py":[{"author":{"_account_id":8027,"name":"Vui Lam","email":"vui@vmware.com","username":"vui"},"change_message_id":"f4f44c5ff19199b4f56cbee38b1497b840da0fdc","unresolved":false,"context_lines":[{"line_number":59,"context_line":"                    \u0027Used only if compute_driver is \u0027"},{"line_number":60,"context_line":"                    \u0027vmwareapi.VMwareVCDriver.\u0027),"},{"line_number":61,"context_line":"    cfg.FloatOpt(\u0027task_poll_interval\u0027,"},{"line_number":62,"context_line":"                 default\u003d0.5,"},{"line_number":63,"context_line":"                 help\u003d\u0027The interval used for polling of remote tasks.\u0027),"},{"line_number":64,"context_line":"    cfg.IntOpt(\u0027api_retry_count\u0027,"},{"line_number":65,"context_line":"               default\u003d10,"}],"source_content_type":"text/x-python","patch_set":3,"id":"AAAAUH%2F9bG4%3D","line":62,"updated":"2014-02-07 00:41:01.000000000","message":"Some additional experimentation needed, something to\nconsider, but need not hold up this patch:\n\nMany operations finish way before 5 seconds, so the original value is too big in general. It may be worthwhile to consider the flip side of hitting the server for 120 times a minute for a single operation that already takes a long time (suspend comes to mind). If we stick with polling, one option would be to use some exponential backoff of the interval; another simpler one is to leverage the initial_delay in the FixedIntervalLoopingCall to give the very first call a more reasonable chance of seeing task completion, and use a slightly bigger interval (e.g. initial_delay\u003d0.5, task_poll_interval\u003d1.5).","commit_id":"73c87a280e77e03d228d34ab4781ca2e3b02e40e"},{"author":{"_account_id":1653,"name":"garyk","email":"gkotton@vmware.com","username":"garyk"},"change_message_id":"9f9094da13cb615b6f3ee7281de2e1cf46491f35","unresolved":false,"context_lines":[{"line_number":59,"context_line":"                    \u0027Used only if compute_driver is \u0027"},{"line_number":60,"context_line":"                    \u0027vmwareapi.VMwareVCDriver.\u0027),"},{"line_number":61,"context_line":"    cfg.FloatOpt(\u0027task_poll_interval\u0027,"},{"line_number":62,"context_line":"                 default\u003d0.5,"},{"line_number":63,"context_line":"                 help\u003d\u0027The interval used for polling of remote tasks.\u0027),"},{"line_number":64,"context_line":"    cfg.IntOpt(\u0027api_retry_count\u0027,"},{"line_number":65,"context_line":"               default\u003d10,"}],"source_content_type":"text/x-python","patch_set":3,"id":"AAAAUH%2F9XZs%3D","line":62,"in_reply_to":"AAAAUH%2F9bG4%3D","updated":"2014-02-07 12:31:28.000000000","message":"Nice idea.","commit_id":"73c87a280e77e03d228d34ab4781ca2e3b02e40e"}]}