)]}'
{"nova/compute/manager.py":[{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"47f847cc0caea08f6ee51c5f88c572e2795020c2","unresolved":false,"context_lines":[{"line_number":7620,"context_line":"        except TypeError:"},{"line_number":7621,"context_line":"            LOG.warning(\"Your virt driver appears to not support the \""},{"line_number":7622,"context_line":"                        \"\u0027use_cache\u0027 parameter to the \u0027get_info\u0027 method; \""},{"line_number":7623,"context_line":"                        \"please update your virt driver.\")"},{"line_number":7624,"context_line":""},{"line_number":7625,"context_line":"        # if it still looks off, go ahead and call stop()"},{"line_number":7626,"context_line":"        if vm_power_state in (power_state.SHUTDOWN,"}],"source_content_type":"text/x-python","patch_set":1,"id":"5fc1f717_8a6103ac","line":7623,"updated":"2019-03-13 15:41:32.000000000","message":"I don\u0027t really like this because we generally don\u0027t make provisions for out of tree drivers, and this kind of thing is easy to just not go back and clean up. I know it\u0027s because we\u0027re worried about backporting a virt driver api change, but I don\u0027t really think we should be (worried about it).","commit_id":"81b0e5a0680883d68d12253af64f8e10200df746"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"9b2b2c0b35fb8a9af368688e9d1a92771527a3b7","unresolved":false,"context_lines":[{"line_number":7620,"context_line":"        except TypeError:"},{"line_number":7621,"context_line":"            LOG.warning(\"Your virt driver appears to not support the \""},{"line_number":7622,"context_line":"                        \"\u0027use_cache\u0027 parameter to the \u0027get_info\u0027 method; \""},{"line_number":7623,"context_line":"                        \"please update your virt driver.\")"},{"line_number":7624,"context_line":""},{"line_number":7625,"context_line":"        # if it still looks off, go ahead and call stop()"},{"line_number":7626,"context_line":"        if vm_power_state in (power_state.SHUTDOWN,"}],"source_content_type":"text/x-python","patch_set":1,"id":"5fc1f717_ca74eb38","line":7623,"in_reply_to":"5fc1f717_8a6103ac","updated":"2019-03-13 15:46:13.000000000","message":"Okay, I\u0027m glad to see it\u0027s already ripped out of master.\n\nNote that there appears to be no test coverage for this path...","commit_id":"81b0e5a0680883d68d12253af64f8e10200df746"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"edf7abc385ba01a8d3f048183bbed9d181527027","unresolved":false,"context_lines":[{"line_number":7620,"context_line":"        except TypeError:"},{"line_number":7621,"context_line":"            LOG.warning(\"Your virt driver appears to not support the \""},{"line_number":7622,"context_line":"                        \"\u0027use_cache\u0027 parameter to the \u0027get_info\u0027 method; \""},{"line_number":7623,"context_line":"                        \"please update your virt driver.\")"},{"line_number":7624,"context_line":""},{"line_number":7625,"context_line":"        # if it still looks off, go ahead and call stop()"},{"line_number":7626,"context_line":"        if vm_power_state in (power_state.SHUTDOWN,"}],"source_content_type":"text/x-python","patch_set":1,"id":"5fc1f717_0a6e732e","line":7623,"in_reply_to":"5fc1f717_ca74eb38","updated":"2019-03-13 15:52:20.000000000","message":"Confirmed locally:\n\nhttps://pastebin.com/RhWtiVJs","commit_id":"81b0e5a0680883d68d12253af64f8e10200df746"}],"nova/virt/ironic/driver.py":[{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"47f847cc0caea08f6ee51c5f88c572e2795020c2","unresolved":false,"context_lines":[{"line_number":907,"context_line":"                    state\u003dmap_power_state(node.power_state))"},{"line_number":908,"context_line":"            except exception.InstanceNotFound:"},{"line_number":909,"context_line":"                return hardware.InstanceInfo("},{"line_number":910,"context_line":"                    state\u003dmap_power_state(ironic_states.NOSTATE))"},{"line_number":911,"context_line":""},{"line_number":912,"context_line":"        if not use_cache:"},{"line_number":913,"context_line":"            return _fetch_from_ironic(self, instance)"}],"source_content_type":"text/x-python","patch_set":1,"id":"5fc1f717_cabceb07","line":910,"updated":"2019-03-13 15:41:32.000000000","message":"It doesn\u0027t look like we save this in the cache anywhere (nor do we at L925). I guess the thinking is that we\u0027ll just wait and update it at the next loop anyway.However, the problem is that this makes two subsequent calls to get_info() return different data if use_cache isn\u0027t consistent. I dunno if that happens anywhere, but it\u0027d be something I would try to avoid for the future when someone puts another call somewhere in a cleanup routine or something. Because you might\u0027ve headed down a path based on non-cached info, and then fail to clean up (or do anything else) based on the cached info.","commit_id":"81b0e5a0680883d68d12253af64f8e10200df746"}]}