)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":30491,"name":"Radosław Piliszek","display_name":"Radek","email":"radek@piliszek.it","username":"yoctozepto","status":"self-employed techologist, collaborating mostly with 7bulls.com"},"change_message_id":"518285b2dcb83ef3b6b89d58bb3bf5d13e298300","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":3,"id":"7582b1bc_f723a859","updated":"2021-12-30 08:34:35.000000000","message":"(loose comment)\n\nalternative solutions proposed before:\n- https://review.opendev.org/c/openstack/kolla-ansible/+/569219/\n- https://review.opendev.org/c/openstack/kolla-ansible/+/619754/\n","commit_id":"0c20e714db3f3a4e6918ac85c6bb5f53ee87d8b3"},{"author":{"_account_id":30491,"name":"Radosław Piliszek","display_name":"Radek","email":"radek@piliszek.it","username":"yoctozepto","status":"self-employed techologist, collaborating mostly with 7bulls.com"},"change_message_id":"be86436b929103a6ea7a3dc71b521c2792db56c4","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":3,"id":"5203ecab_0395b2e3","in_reply_to":"464d4323_baa82a19","updated":"2022-07-21 12:42:40.000000000","message":"If you meant \"analogously to mariadb shards\", then it makes sense.","commit_id":"0c20e714db3f3a4e6918ac85c6bb5f53ee87d8b3"},{"author":{"_account_id":31506,"name":"Lin PeiWen","email":"591171850@qq.com","username":"LinPeiWen"},"change_message_id":"2ae4aea638a8ad887217719306db6436b2ee4d24","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":3,"id":"78546b61_abae54f8","in_reply_to":"5203ecab_0395b2e3","updated":"2022-07-22 15:18:19.000000000","message":"Our production environment has 1,000 to several thousand computing nodes. We considered the load of CPU and memory, and chose to deploy a separate rabbitmq cluster in each cell controller, so that the load of the rabbitmq cluster would not be too high. The architecture is similar to what I submitted. this patch","commit_id":"0c20e714db3f3a4e6918ac85c6bb5f53ee87d8b3"},{"author":{"_account_id":27339,"name":"Michal Arbet","email":"michal.arbet@ultimum.io","username":"michalarbet"},"change_message_id":"035d00c10eaa2db9e187fdf33d23f2b20047a938","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":3,"id":"464d4323_baa82a19","in_reply_to":"7582b1bc_f723a859","updated":"2022-07-21 12:32:34.000000000","message":"What about implement rabbitmq for cells as mariadb shards ? Group by id of cluster and instal...","commit_id":"0c20e714db3f3a4e6918ac85c6bb5f53ee87d8b3"},{"author":{"_account_id":31506,"name":"Lin PeiWen","email":"591171850@qq.com","username":"LinPeiWen"},"change_message_id":"c45a9d5ecf3cbf7258a3c36bfc3dce41b75c3f92","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":3,"id":"79e5144c_58883e38","in_reply_to":"78546b61_abae54f8","updated":"2022-07-22 15:20:59.000000000","message":"rabbitmq sharding is a good idea, I don\u0027t think the two conflict","commit_id":"0c20e714db3f3a4e6918ac85c6bb5f53ee87d8b3"},{"author":{"_account_id":32553,"name":"Sven Kieske","email":"sven_oss@posteo.de","username":"skieske"},"change_message_id":"c649d94b1a7e390f68f2c9b1812b54035b57a08e","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":7,"id":"37cadf9b_d1d9ee0c","updated":"2024-06-25 07:28:57.000000000","message":"I think such a large change - from an architecture point of view - needs at a minimum some basic tests, please add some, thank you!\n\nalso the merge conflict obviously needs to get resolved.","commit_id":"07d73221e68436d975db8c603fd1fe84edc0e896"},{"author":{"_account_id":34579,"name":"Juan Pablo Suazo","email":"jsuazo@whitestack.com","username":"jsuazo"},"change_message_id":"824965de8b863c0183b30a02e3730d73baae2c7e","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":7,"id":"ff9e3a83_f605b153","updated":"2023-12-18 19:17:52.000000000","message":"It would be nice to add a test for these new configurations, but that may be left for a followup review if its considered too time intensive.","commit_id":"07d73221e68436d975db8c603fd1fe84edc0e896"},{"author":{"_account_id":36670,"name":"Jay Jahns","email":"jayjahns@gmail.com","username":"jayjahns"},"change_message_id":"4e1595ee06c68e9e8b7e5bfb79be36178b6c875c","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":7,"id":"ec793156_ef9659c9","updated":"2024-06-10 20:37:37.000000000","message":"The service-rabbitmq change is affecting all deployments that run rabbitmq on separate nodes from control. This needs to be prioritized as it is a deployment breaking issue.","commit_id":"07d73221e68436d975db8c603fd1fe84edc0e896"}],"ansible/inventory/all-in-one":[{"author":{"_account_id":32553,"name":"Sven Kieske","email":"sven_oss@posteo.de","username":"skieske"},"change_message_id":"c649d94b1a7e390f68f2c9b1812b54035b57a08e","unresolved":true,"context_lines":[{"line_number":270,"context_line":"nova"},{"line_number":271,"context_line":""},{"line_number":272,"context_line":"[cells-rabbitmq-clusters:children]"},{"line_number":273,"context_line":"nova-conductor"},{"line_number":274,"context_line":""},{"line_number":275,"context_line":"# Neutron"},{"line_number":276,"context_line":"[neutron-server:children]"}],"source_content_type":"application/octet-stream","patch_set":7,"id":"b61abc7d_6c4f510c","line":273,"range":{"start_line":273,"start_character":0,"end_line":273,"end_character":2},"updated":"2024-06-25 07:28:57.000000000","message":"why is this installed alongside nova-conductor?","commit_id":"07d73221e68436d975db8c603fd1fe84edc0e896"}],"ansible/inventory/multinode":[{"author":{"_account_id":32553,"name":"Sven Kieske","email":"sven_oss@posteo.de","username":"skieske"},"change_message_id":"c649d94b1a7e390f68f2c9b1812b54035b57a08e","unresolved":true,"context_lines":[{"line_number":288,"context_line":"nova"},{"line_number":289,"context_line":""},{"line_number":290,"context_line":"[cells-rabbitmq-clusters:children]"},{"line_number":291,"context_line":"nova-conductor"},{"line_number":292,"context_line":""},{"line_number":293,"context_line":"# Neutron"},{"line_number":294,"context_line":"[neutron-server:children]"}],"source_content_type":"application/octet-stream","patch_set":7,"id":"2a47c127_d5acb72f","line":291,"range":{"start_line":291,"start_character":0,"end_line":291,"end_character":2},"updated":"2024-06-25 07:28:57.000000000","message":"what is the reasoning here for the service placement? please explain it.","commit_id":"07d73221e68436d975db8c603fd1fe84edc0e896"}],"doc/source/reference/compute/nova-cells-guide.rst":[{"author":{"_account_id":32553,"name":"Sven Kieske","email":"sven_oss@posteo.de","username":"skieske"},"change_message_id":"c649d94b1a7e390f68f2c9b1812b54035b57a08e","unresolved":true,"context_lines":[{"line_number":152,"context_line":"and message queue instance. These could be separate database and message queue"},{"line_number":153,"context_line":"clusters, or shared database and message queue clusters partitioned via"},{"line_number":154,"context_line":"database names and virtual hosts. Currently Kolla Ansible supports deployment"},{"line_number":155,"context_line":"of shared database cluster."},{"line_number":156,"context_line":""},{"line_number":157,"context_line":"Configuration"},{"line_number":158,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":7,"id":"a63314f9_41ac9ac5","line":155,"range":{"start_line":155,"start_character":0,"end_line":155,"end_character":2},"updated":"2024-06-25 07:28:57.000000000","message":"we don\u0027t remove the ability to run shared message queue clusters, do we? so why remove this sentence?","commit_id":"07d73221e68436d975db8c603fd1fe84edc0e896"},{"author":{"_account_id":34579,"name":"Juan Pablo Suazo","email":"jsuazo@whitestack.com","username":"jsuazo"},"change_message_id":"824965de8b863c0183b30a02e3730d73baae2c7e","unresolved":true,"context_lines":[{"line_number":343,"context_line":"Globals.yml:"},{"line_number":344,"context_line":""},{"line_number":345,"context_line":".. code-block:: yaml"},{"line_number":346,"context_line":""},{"line_number":347,"context_line":"   enable_cells_rabbitmq: yes"},{"line_number":348,"context_line":"   rabbitmq_clusters:"},{"line_number":349,"context_line":"     - group: \"cell1_rabbitmq\""}],"source_content_type":"text/x-rst","patch_set":7,"id":"35eeb028_d4120802","line":346,"updated":"2023-12-18 19:17:52.000000000","message":"I think it would be beneficial to exemplify configurations for 2 cells with dedicated message queue clusters and one reusing the global message queue cluster. This way we can see what the differences should be between the configurations of 2 rabbitmq clusters, if any.","commit_id":"07d73221e68436d975db8c603fd1fe84edc0e896"},{"author":{"_account_id":32553,"name":"Sven Kieske","email":"sven_oss@posteo.de","username":"skieske"},"change_message_id":"c649d94b1a7e390f68f2c9b1812b54035b57a08e","unresolved":true,"context_lines":[{"line_number":343,"context_line":"Globals.yml:"},{"line_number":344,"context_line":""},{"line_number":345,"context_line":".. code-block:: yaml"},{"line_number":346,"context_line":""},{"line_number":347,"context_line":"   enable_cells_rabbitmq: yes"},{"line_number":348,"context_line":"   rabbitmq_clusters:"},{"line_number":349,"context_line":"     - group: \"cell1_rabbitmq\""}],"source_content_type":"text/x-rst","patch_set":7,"id":"0a273612_cf4cbd61","line":346,"in_reply_to":"35eeb028_d4120802","updated":"2024-06-25 07:28:57.000000000","message":"agreed, but the default of one global message queue cluster should also be explained imho. start with the simplest setup - that is one global cluster - and start getting more complicated from there.","commit_id":"07d73221e68436d975db8c603fd1fe84edc0e896"},{"author":{"_account_id":34579,"name":"Juan Pablo Suazo","email":"jsuazo@whitestack.com","username":"jsuazo"},"change_message_id":"824965de8b863c0183b30a02e3730d73baae2c7e","unresolved":true,"context_lines":[{"line_number":356,"context_line":""},{"line_number":357,"context_line":"Inventory:"},{"line_number":358,"context_line":""},{"line_number":359,"context_line":".. code-block:: INI"},{"line_number":360,"context_line":""},{"line_number":361,"context_line":"   [cell1_rabbitmq]"},{"line_number":362,"context_line":"   cell-control01"}],"source_content_type":"text/x-rst","patch_set":7,"id":"82c2940f_48bd3bae","line":359,"updated":"2023-12-18 19:17:52.000000000","message":"If `cell-control01` was declared like this, but additionally setup as a conductor node, could this result in applying the general rabbitmq task AND the specific cell rabbitmq tasks on the same host ?\n\nI know this config is not correct, but im thinking if the case were this config may be already set:\n```\n[nova-conductor]\ncell-control01\n```\nAnd then the multi-cell configs are added without removing the previous one by mistake.\n\nIm not certain this would be a problem, but in the case it is, we should either add documentation clarifying this or add a task that checks that hosts do not have the `cells-rabbitmq-clusters` before applying the `rabbitmq` tasks.","commit_id":"07d73221e68436d975db8c603fd1fe84edc0e896"}]}