)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":9555,"name":"Matthew Booth","email":"mbooth@redhat.com","username":"MatthewBooth"},"change_message_id":"c2d4d1b859a6f8bc7ff132d2217a07981ae2de33","unresolved":false,"context_lines":[{"line_number":14,"context_line":"This has now changed so the BDM is only destroyed after a successful"},{"line_number":15,"context_line":"detach call has been made to cinder. In the event of a failure during"},{"line_number":16,"context_line":"this detach call to Cinder an operator can now make another attempt to"},{"line_number":17,"context_line":"detach the volume via Nova as the BDM is no longer destroyed."},{"line_number":18,"context_line":""},{"line_number":19,"context_line":"Change-Id: I2581ff9f9c0e7cfc14a25acf45eb1860df69eacf"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":7,"id":"3a1ff146_28166e55","line":17,"updated":"2017-03-16 10:09:20.000000000","message":"nit: This reads like a bug. It could have a bug associated with it.","commit_id":"31664579c607494c8755ccffde39809d4c216b17"}],"nova/compute/manager.py":[{"author":{"_account_id":9555,"name":"Matthew Booth","email":"mbooth@redhat.com","username":"MatthewBooth"},"change_message_id":"c2d4d1b859a6f8bc7ff132d2217a07981ae2de33","unresolved":false,"context_lines":[{"line_number":4937,"context_line":""},{"line_number":4938,"context_line":"        info \u003d dict(volume_id\u003dvolume_id)"},{"line_number":4939,"context_line":"        self._notify_about_instance_usage("},{"line_number":4940,"context_line":"            context, instance, \"volume.detach\", extra_usage_info\u003dinfo)"},{"line_number":4941,"context_line":""},{"line_number":4942,"context_line":"    @wrap_exception()"},{"line_number":4943,"context_line":"    @wrap_instance_fault"}],"source_content_type":"text/x-python","patch_set":7,"id":"3a1ff146_fb0c1e96","line":4940,"updated":"2017-03-16 10:09:20.000000000","message":"This is a change in behaviour: Previously we notified in advance, now we notify after. TBH, notifying after sounds more sane to me, but if you\u0027re going to change it can you please call it out and justify it in the commit message?\n\nEither way the notification should really live with the detach, i.e. up a few lines.","commit_id":"31664579c607494c8755ccffde39809d4c216b17"},{"author":{"_account_id":782,"name":"John Garbutt","email":"john@johngarbutt.com","username":"johngarbutt"},"change_message_id":"6c8909af5fb5d005cacfc4326f9594afce76af5a","unresolved":false,"context_lines":[{"line_number":4933,"context_line":"                               attachment_id)"},{"line_number":4934,"context_line":"        info \u003d dict(volume_id\u003dvolume_id)"},{"line_number":4935,"context_line":"        self._notify_about_instance_usage("},{"line_number":4936,"context_line":"            context, instance, \"volume.detach\", extra_usage_info\u003dinfo)"},{"line_number":4937,"context_line":""},{"line_number":4938,"context_line":"        if destroy_bdm:"},{"line_number":4939,"context_line":"            bdm.destroy()"}],"source_content_type":"text/x-python","patch_set":9,"id":"1a1ced50_32938ea5","line":4936,"updated":"2017-03-16 17:55:00.000000000","message":"This seems good, in no small part because terminate_connection and detach turn into a single delete attachment call in the future.\n\nIf cinder id down, terminate would fail and notify would never get called before. This keeps things as close as we can in the long run.","commit_id":"0ca8b5e23ecde0a4bc19d6aabc745e1171b9b72a"},{"author":{"_account_id":6873,"name":"Matt Riedemann","email":"mriedem.os@gmail.com","username":"mriedem"},"change_message_id":"6588656068b32c3f296dbe3fb004b741f3ed9d7e","unresolved":false,"context_lines":[{"line_number":4936,"context_line":"            context, instance, \"volume.detach\", extra_usage_info\u003dinfo)"},{"line_number":4937,"context_line":""},{"line_number":4938,"context_line":"        if destroy_bdm:"},{"line_number":4939,"context_line":"            bdm.destroy()"},{"line_number":4940,"context_line":""},{"line_number":4941,"context_line":"    @wrap_exception()"},{"line_number":4942,"context_line":"    @wrap_instance_fault"}],"source_content_type":"text/x-python","patch_set":11,"id":"5f4e5783_32cc7b88","line":4939,"updated":"2017-10-10 16:33:02.000000000","message":"Moving this results in a weird timing bug:\n\nhttps://bugs.launchpad.net/nova/+bug/1722577\n\nI think Tempest, as the client, is making an invalid request and should be handling the 400 case, so we can fix there, but it just goes to show how seemingly unrelated changes in the compute code changes API behavior.","commit_id":"bea6ff62afa4ebe464cc22a8ffd68d848b1ca078"}]}