)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"bfa48340039f071a6c3564201533eedd51a5598a","unresolved":true,"context_lines":[{"line_number":10,"context_line":"ResourceProviderCreationFailed exception when"},{"line_number":11,"context_line":"we enter pre_start_hook() we simply continue and let"},{"line_number":12,"context_line":"this issue loop forever and it\u0027s happily reporting"},{"line_number":13,"context_line":"it\u0027s state as up."},{"line_number":14,"context_line":""},{"line_number":15,"context_line":"This can happen if there is a naming conflict for"},{"line_number":16,"context_line":"resource providers and we should fail to start the"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"694fa77a_4515c401","line":13,"updated":"2024-01-09 13:05:53.000000000","message":"this commit message does not make sense\n\nthe nova-comptue agent is not interacting with teh nova-scheduler process.\n\ndid you mean the schduler client or placement cleint as those are very different things.","commit_id":"d8aab2a714c92970d409715bc7e78a2323372151"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"bfa48340039f071a6c3564201533eedd51a5598a","unresolved":true,"context_lines":[{"line_number":15,"context_line":"This can happen if there is a naming conflict for"},{"line_number":16,"context_line":"resource providers and we should fail to start the"},{"line_number":17,"context_line":"service so that the nova-compute service is correctly"},{"line_number":18,"context_line":"set to down to indicate an issue."},{"line_number":19,"context_line":""},{"line_number":20,"context_line":"Change-Id: I5c36dcf2bf88f1f19639e6fd946a4ae9086e0151"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"b382fb39_daa71597","line":18,"updated":"2024-01-09 13:05:53.000000000","message":"with the introduction of the compute stable uuid feature in antelope this should no longer happen in general\n\nhttps://specs.openstack.org/openstack/nova-specs/specs/2023.1/implemented/stable-compute-uuid.html\n\nif you deploy 2 services the first one will register in the db and be allocated a uuid which it will write to a file.\nthe second one will find that i t does not ahve a compute uuid file but have a compute service record\n\nthis will cause the second agent to exit if its antelope or newer.\n\n\n\nwe chatted about your specific usecase on irc and in your case you were doing a compute node repalcement following a scale in and scale out approch keeping the old host name.\n\nthe step that was missed was the compute service delete wich in turn delete all allocations on the associated RP and finally removes the rp.\n\nim guessing your useing a pre antelope openstack without the stable uuid feature?\ncan you file a bug for this?","commit_id":"d8aab2a714c92970d409715bc7e78a2323372151"}],"/PATCHSET_LEVEL":[{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"480d547a6372b3dd417b4eea226d1f5b5a37d265","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"cd7519d8_0dd99a36","updated":"2024-01-19 13:33:11.000000000","message":"so as noted in the commit message the stable uuid feature should prevet us getting to the placement case in the event that we have duplicate hosts\n\nbut i could forcee there beeing stale resouce providers in other case that woudl cause the rp create to fail so i think this is still useful espeically on older branches without the stable uuid changes.\n\nThis needs at least a unit test and or functional test to demonstrate its working as intended however.","commit_id":"c6c38b6018cd72736762c827a3ef52faa15ef760"},{"author":{"_account_id":16137,"name":"Tobias Urdin","email":"tobias.urdin@binero.com","username":"tobasco"},"change_message_id":"70feaf22c90a2534668b5a9220fda9e267c16494","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"1386438e_2085491c","updated":"2024-01-22 16:21:52.000000000","message":"Updated with testing per Sean\u0027s request","commit_id":"2ca3d3cf23f87d994c58e440e0a8a94ae8c0990a"}],"nova/compute/manager.py":[{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"bfa48340039f071a6c3564201533eedd51a5598a","unresolved":true,"context_lines":[{"line_number":10638,"context_line":"        except Exception:"},{"line_number":10639,"context_line":"            LOG.exception(\"Error updating resources for node %(node)s.\","},{"line_number":10640,"context_line":"                          {\u0027node\u0027: nodename})"},{"line_number":10641,"context_line":"            if startup:"},{"line_number":10642,"context_line":"                # If this happens during startup, we need to let it raise to"},{"line_number":10643,"context_line":"                # abort our service startup."},{"line_number":10644,"context_line":"                raise"}],"source_content_type":"text/x-python","patch_set":1,"id":"30649436_1cd8510c","line":10641,"updated":"2024-01-09 13:05:53.000000000","message":"so i think we will need to add a check to either skip this for ironic so that\nwe dont exit if any of the ironic nodes has an error as this could cause quite a large outage if just one RP got currpted.\n\nso you just need to check the compute driver in the config and not raise if ironic.\nAlternatively we  might need to make this configurable\n\nsometing liek placement_update_failure_is_faital.","commit_id":"d8aab2a714c92970d409715bc7e78a2323372151"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"1b5395edc0b25cc55ae91cb839f38a609e62b4e4","unresolved":false,"context_lines":[{"line_number":10638,"context_line":"        except Exception:"},{"line_number":10639,"context_line":"            LOG.exception(\"Error updating resources for node %(node)s.\","},{"line_number":10640,"context_line":"                          {\u0027node\u0027: nodename})"},{"line_number":10641,"context_line":"            if startup:"},{"line_number":10642,"context_line":"                # If this happens during startup, we need to let it raise to"},{"line_number":10643,"context_line":"                # abort our service startup."},{"line_number":10644,"context_line":"                raise"}],"source_content_type":"text/x-python","patch_set":1,"id":"5f29b6fe_199c14af","line":10641,"in_reply_to":"30649436_1cd8510c","updated":"2024-01-19 13:33:33.000000000","message":"Done","commit_id":"d8aab2a714c92970d409715bc7e78a2323372151"}]}