)]}'
{"nova/cmd/manage.py":[{"author":{"_account_id":6873,"name":"Matt Riedemann","email":"mriedem.os@gmail.com","username":"mriedem"},"change_message_id":"dddf7cd283bb981ac8e465761c5d13c6a2deccc6","unresolved":false,"context_lines":[{"line_number":1118,"context_line":"            # That said, we need to notify the operator that the command"},{"line_number":1119,"context_line":"            # partially completed and potentially need a further run."},{"line_number":1120,"context_line":"            print(_(\u0027No hosts found to map to cell.\u0027))"},{"line_number":1121,"context_line":"            print(_(\u0027Please run the command again once you start new compute \u0027"},{"line_number":1122,"context_line":"                    \u0027nodes.\u0027))"},{"line_number":1123,"context_line":"            return 0"},{"line_number":1124,"context_line":"        self.map_instances(cell_uuid)"}],"source_content_type":"text/x-python","patch_set":1,"id":"7a3c09a3_060d403e","line":1121,"updated":"2017-01-13 18:52:15.000000000","message":"Couldn\u0027t this also be telling them to run map_cell_and_hosts, or discover_hosts, as alternatives to this? I\u0027m not sure what\u0027s best, I really need to map out what each of these CLIs does, when it\u0027s expected to be run and under what circumstances.","commit_id":"a31ac0fa0ba25a5a4bb8811c3528f3669848277d"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"e46769abcf85e1156c29d07d4c4a08bcaef42174","unresolved":false,"context_lines":[{"line_number":1118,"context_line":"            # That said, we need to notify the operator that the command"},{"line_number":1119,"context_line":"            # partially completed and potentially need a further run."},{"line_number":1120,"context_line":"            print(_(\u0027No hosts found to map to cell.\u0027))"},{"line_number":1121,"context_line":"            print(_(\u0027Please run the command again once you start new compute \u0027"},{"line_number":1122,"context_line":"                    \u0027nodes.\u0027))"},{"line_number":1123,"context_line":"            return 0"},{"line_number":1124,"context_line":"        self.map_instances(cell_uuid)"}],"source_content_type":"text/x-python","patch_set":1,"id":"7a3c09a3_89c78311","line":1121,"in_reply_to":"7a3c09a3_060d403e","updated":"2017-01-13 19:12:01.000000000","message":"Yeah. The simple command is really supposed to be for people that don\u0027t need to run the others. Having it too varied in its behavior defeats the purpose I think.","commit_id":"a31ac0fa0ba25a5a4bb8811c3528f3669848277d"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"309d135275ba0f9fe4e3e6f07be07365394dcc8c","unresolved":false,"context_lines":[{"line_number":1120,"context_line":"            print(_(\u0027No hosts found to map to cell.\u0027))"},{"line_number":1121,"context_line":"            print(_(\u0027Please run the command again once you start new compute \u0027"},{"line_number":1122,"context_line":"                    \u0027nodes.\u0027))"},{"line_number":1123,"context_line":"            return 0"},{"line_number":1124,"context_line":"        self.map_instances(cell_uuid)"},{"line_number":1125,"context_line":"        return 0"},{"line_number":1126,"context_line":""}],"source_content_type":"text/x-python","patch_set":1,"id":"7a3c09a3_be655ae2","line":1123,"updated":"2017-01-13 17:24:32.000000000","message":"Personally, I think that this is wrong. We should return zero from this command if everything is setup. If there is more to do, then we should return 1.\n\nIn this case, we should return 1, and print a message that the op needs to run \"discover_hosts\" once they have computes, not jut run this simple command again.\n\nIMHO, the above says \"Well, your deployment doesn\u0027t work but I\u0027m returning zero anyway\" and that\u0027s not what we should do, IMHO.","commit_id":"a31ac0fa0ba25a5a4bb8811c3528f3669848277d"},{"author":{"_account_id":6873,"name":"Matt Riedemann","email":"mriedem.os@gmail.com","username":"mriedem"},"change_message_id":"e70b6476838454fc992e2105d56f534b65dd612d","unresolved":false,"context_lines":[{"line_number":1120,"context_line":"            print(_(\u0027No hosts found to map to cell.\u0027))"},{"line_number":1121,"context_line":"            print(_(\u0027Please run the command again once you start new compute \u0027"},{"line_number":1122,"context_line":"                    \u0027nodes.\u0027))"},{"line_number":1123,"context_line":"            return 0"},{"line_number":1124,"context_line":"        self.map_instances(cell_uuid)"},{"line_number":1125,"context_line":"        return 0"},{"line_number":1126,"context_line":""}],"source_content_type":"text/x-python","patch_set":1,"id":"7a3c09a3_c6156de5","line":1123,"in_reply_to":"7a3c09a3_2f1041a8","updated":"2017-01-17 18:45:21.000000000","message":"nova-manage cell_v2 list_cells is in now:\n\nhttps://review.openstack.org/#/c/420440/","commit_id":"a31ac0fa0ba25a5a4bb8811c3528f3669848277d"},{"author":{"_account_id":16312,"name":"Alfredo Moralejo","email":"amoralej@redhat.com","username":"amoralej"},"change_message_id":"99cb708792a0bed7c52d6f78d046aeab462b0ac7","unresolved":false,"context_lines":[{"line_number":1120,"context_line":"            print(_(\u0027No hosts found to map to cell.\u0027))"},{"line_number":1121,"context_line":"            print(_(\u0027Please run the command again once you start new compute \u0027"},{"line_number":1122,"context_line":"                    \u0027nodes.\u0027))"},{"line_number":1123,"context_line":"            return 0"},{"line_number":1124,"context_line":"        self.map_instances(cell_uuid)"},{"line_number":1125,"context_line":"        return 0"},{"line_number":1126,"context_line":""}],"source_content_type":"text/x-python","patch_set":1,"id":"7a3c09a3_2f1041a8","line":1123,"in_reply_to":"7a3c09a3_31918df1","updated":"2017-01-14 11:42:00.000000000","message":"Please, in order to make implementation easier, map_cell0 and create_cell should be idempotent. BTW, is there any command to list existing cells?","commit_id":"a31ac0fa0ba25a5a4bb8811c3528f3669848277d"},{"author":{"_account_id":6873,"name":"Matt Riedemann","email":"mriedem.os@gmail.com","username":"mriedem"},"change_message_id":"dddf7cd283bb981ac8e465761c5d13c6a2deccc6","unresolved":false,"context_lines":[{"line_number":1120,"context_line":"            print(_(\u0027No hosts found to map to cell.\u0027))"},{"line_number":1121,"context_line":"            print(_(\u0027Please run the command again once you start new compute \u0027"},{"line_number":1122,"context_line":"                    \u0027nodes.\u0027))"},{"line_number":1123,"context_line":"            return 0"},{"line_number":1124,"context_line":"        self.map_instances(cell_uuid)"},{"line_number":1125,"context_line":"        return 0"},{"line_number":1126,"context_line":""}],"source_content_type":"text/x-python","patch_set":1,"id":"7a3c09a3_66a8c420","line":1123,"in_reply_to":"7a3c09a3_3e7ecae4","updated":"2017-01-13 18:52:15.000000000","message":"I also like Dan\u0027s idea of a --no-computes argument to give TripleO (or any deployer) an outlet to run this when setting up the control plane before there are computes so that it will do it\u0027s thing up until this point, so you can run it early with --no-computes, then setup computes, and then run w/o the --no-computes option and it\u0027s a failure if there are no computes found yet.","commit_id":"a31ac0fa0ba25a5a4bb8811c3528f3669848277d"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"e46769abcf85e1156c29d07d4c4a08bcaef42174","unresolved":false,"context_lines":[{"line_number":1120,"context_line":"            print(_(\u0027No hosts found to map to cell.\u0027))"},{"line_number":1121,"context_line":"            print(_(\u0027Please run the command again once you start new compute \u0027"},{"line_number":1122,"context_line":"                    \u0027nodes.\u0027))"},{"line_number":1123,"context_line":"            return 0"},{"line_number":1124,"context_line":"        self.map_instances(cell_uuid)"},{"line_number":1125,"context_line":"        return 0"},{"line_number":1126,"context_line":""}],"source_content_type":"text/x-python","patch_set":1,"id":"7a3c09a3_89eea389","line":1123,"in_reply_to":"7a3c09a3_66a8c420","updated":"2017-01-13 19:12:01.000000000","message":"After talking with melanie and sylvain, I think I\u0027ve convinced myself that we shouldn\u0027t do that. This simple command should do the simple thing or not. If you (tripleo) need to do the pieces in multiple piaces/times, then you should use the manual commands.\n\nI think we\u0027re going to move forward with the other patch Andrey already has up, which gives us the ability to do the cell create manually as well, and then just make tripleo save the discovery bit until the end when computes are up.","commit_id":"a31ac0fa0ba25a5a4bb8811c3528f3669848277d"},{"author":{"_account_id":6873,"name":"Matt Riedemann","email":"mriedem.os@gmail.com","username":"mriedem"},"change_message_id":"51f5f8495d318a3380a1cb095e8e06ffbab49649","unresolved":false,"context_lines":[{"line_number":1120,"context_line":"            print(_(\u0027No hosts found to map to cell.\u0027))"},{"line_number":1121,"context_line":"            print(_(\u0027Please run the command again once you start new compute \u0027"},{"line_number":1122,"context_line":"                    \u0027nodes.\u0027))"},{"line_number":1123,"context_line":"            return 0"},{"line_number":1124,"context_line":"        self.map_instances(cell_uuid)"},{"line_number":1125,"context_line":"        return 0"},{"line_number":1126,"context_line":""}],"source_content_type":"text/x-python","patch_set":1,"id":"7a3c09a3_ef89e7cb","line":1123,"in_reply_to":"7a3c09a3_89eea389","updated":"2017-01-13 20:07:29.000000000","message":"For my reference, that\u0027s this:\n\nhttps://review.openstack.org/#/c/332713/","commit_id":"a31ac0fa0ba25a5a4bb8811c3528f3669848277d"},{"author":{"_account_id":4690,"name":"melanie witt","display_name":"melwitt","email":"melwittt@gmail.com","username":"melwitt"},"change_message_id":"0d935c7be29979f36ae4747a2ae57b240f97a34b","unresolved":false,"context_lines":[{"line_number":1120,"context_line":"            print(_(\u0027No hosts found to map to cell.\u0027))"},{"line_number":1121,"context_line":"            print(_(\u0027Please run the command again once you start new compute \u0027"},{"line_number":1122,"context_line":"                    \u0027nodes.\u0027))"},{"line_number":1123,"context_line":"            return 0"},{"line_number":1124,"context_line":"        self.map_instances(cell_uuid)"},{"line_number":1125,"context_line":"        return 0"},{"line_number":1126,"context_line":""}],"source_content_type":"text/x-python","patch_set":1,"id":"7a3c09a3_be79da31","line":1123,"in_reply_to":"7a3c09a3_be655ae2","updated":"2017-01-13 17:32:49.000000000","message":"It wouldn\u0027t necessarily be \u0027discover_hosts\u0027 though, because \u0027discover_hosts\u0027 assumes the existence of a cell mapping already. If there isn\u0027t a cell_mapping for hosts to go to yet, they would need \u0027map_cell_and_hosts\u0027.\n\nDo we have anything against the idea of creating an empty cell? Andrey has a change up [1] to allow creating an empty cell. That would mean the flow could be, create an empty cell, then later use \u0027discover_hosts\u0027 to map hosts to that cell. Instead of the confusing difference between \u0027discover_hosts\u0027 and \u0027map_cell_and_hosts\u0027.\n\n[1] https://review.openstack.org/#/c/332713/","commit_id":"a31ac0fa0ba25a5a4bb8811c3528f3669848277d"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"3cbd6cf055304acc5c4109ba5a78e02c638a6dfd","unresolved":false,"context_lines":[{"line_number":1120,"context_line":"            print(_(\u0027No hosts found to map to cell.\u0027))"},{"line_number":1121,"context_line":"            print(_(\u0027Please run the command again once you start new compute \u0027"},{"line_number":1122,"context_line":"                    \u0027nodes.\u0027))"},{"line_number":1123,"context_line":"            return 0"},{"line_number":1124,"context_line":"        self.map_instances(cell_uuid)"},{"line_number":1125,"context_line":"        return 0"},{"line_number":1126,"context_line":""}],"source_content_type":"text/x-python","patch_set":1,"id":"7a3c09a3_3e7ecae4","line":1123,"in_reply_to":"7a3c09a3_be79da31","updated":"2017-01-13 17:37:57.000000000","message":"Okay, I thought it would still create the cell before it looked for hosts. But no, I\u0027m fine with a create cell command and I think doing that, then a discover later would be ideal.","commit_id":"a31ac0fa0ba25a5a4bb8811c3528f3669848277d"},{"author":{"_account_id":6873,"name":"Matt Riedemann","email":"mriedem.os@gmail.com","username":"mriedem"},"change_message_id":"53b7e74da9f40a04b4e1c0eb899d8ac3e743f057","unresolved":false,"context_lines":[{"line_number":1120,"context_line":"            print(_(\u0027No hosts found to map to cell.\u0027))"},{"line_number":1121,"context_line":"            print(_(\u0027Please run the command again once you start new compute \u0027"},{"line_number":1122,"context_line":"                    \u0027nodes.\u0027))"},{"line_number":1123,"context_line":"            return 0"},{"line_number":1124,"context_line":"        self.map_instances(cell_uuid)"},{"line_number":1125,"context_line":"        return 0"},{"line_number":1126,"context_line":""}],"source_content_type":"text/x-python","patch_set":1,"id":"7a3c09a3_31918df1","line":1123,"in_reply_to":"7a3c09a3_ef89e7cb","updated":"2017-01-14 02:23:58.000000000","message":"We also talked in IRC today about using simple_cell_setup for upgrade scenarios, so leave this in use in grenade, at least for now (upgrading from newton), but don\u0027t use it for devstack (fresh installs) and use the map_cell0/create_cell/discover_hosts commands in devstack for initial setup. So I think we could change this patch to not modify simple_cell_setup, but I think the change below for map_cell_and_hosts is still valid.","commit_id":"a31ac0fa0ba25a5a4bb8811c3528f3669848277d"}]}