)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"052855be60a256ccc516b289ca388e6f02fdc6d2","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"9f34cff8_0b6ef86a","updated":"2026-03-12 13:07:24.000000000","message":"i kind of liek the idea but not the approch\n\ni dont like the use of xml/csv files \nand i woudl not consider the current fixture to be fixutrs.\nthose are proxies or fakes.\n\nso if we were to take this direction then i woudl want this to be in a diffent repo\n\n\nmy prefence woudl instead to emulate the nova/aetos apis including compute nodes/hyrperviosrs\n and build the moduels based on that.\n \n so no static data.\n \n\nif we want 3 compute we would register 3 hyperviors with the nova fixture and return the approate repsonce in the hyperviors api.\nsame fo instance/falvors\n\nwe would then use somethign like wsgi_intercpet to hook the sdk connectoin or manually monkey patch the connecton used by our nova_helper to call the fiture functions ectra.","commit_id":"7ccf49d37d44f79f125c60d57f28d45ea8700293"},{"author":{"_account_id":16312,"name":"Alfredo Moralejo","email":"amoralej@redhat.com","username":"amoralej"},"change_message_id":"f568aa26df172591a509b095715a190712de1e5e","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"bd0e2a94_355b36d4","in_reply_to":"9f34cff8_0b6ef86a","updated":"2026-03-13 08:36:46.000000000","message":"\u003e i dont like the use of xml/csv files\n\nFor me it has a big advantadge of being able to use xml content from real cases taken from the decision-engine logs directly. I consider that capacity of replicating real-world cases a big advandge. Similar for the metrics and csv (that\u0027s the parser_watcher_logs.py that i added in contrib creates that csv from workload_stabilization logs.\n\nIIUC, your proposal is that the nova fixture should be more complete and instances and hypervisors should be created by emulating those nova API interfaces (server create, hypervisor register, etc...) and calling it from the tests? Is that your proposal?\n\n\n\u003e  I woudl not consider the current fixture to be fixture. those are proxies or fakes. \n\nI know, this was just an initial PoC. As i stated in the commit message \"This initial approach is only standalone but i think it can be easily\nadapted to be used as fixtures with wsgi_intercept\". I expect i will be able to use the same app in a wsgi_intercept interface for the functional tests so that i can use the same server both from main method as standalone and in functional tests wrapped by wsgi_intercept (didn\u0027t try yet).\n\n\u003e if we were to take this direction then i woudl want this to be in a diffent repo\n\nAt this point, I consider critical the availability of a tool to reproduce scalability (potentially also for non-scalability ones, but the existing bugs is what led me to this now) issues locally in a running environment to work on fixing some existing issues. I had the expectation that this work may be incorporated to the functional tests effort so that the same nova and prometheus fake interface can be used in both cases. If there is agreement in that those two efforts are not compatible, I will put this fake servers in contrib directory in watcher, in other watcher repo (if there is any better candidate) or in a personal repo in github.","commit_id":"7ccf49d37d44f79f125c60d57f28d45ea8700293"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"498ac14e14098c30f9c29102bcc4991ab0429268","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":2,"id":"e2c78a41_927c1dd4","in_reply_to":"bd0e2a94_355b36d4","updated":"2026-03-13 11:00:44.000000000","message":"so the xml and csv files are to large to sanely review\nto me that is a beine in this case secuirty issues \n\ni dont think we shoudl be commiting large data sets liek this into this repo even in a contirb directory.\n\nif you want to have partail templeats and generate fake data\nor have a ways to laod an external file but not commit them that is ok\nbut i dont think this is compatiable with functional testing. \n\nits somethign else. it is not unfueful in fact tis far form that but if we were to buidl this this woudl be a seperate perfoamce or intergatrion test suite not a funcitonla test suite.","commit_id":"7ccf49d37d44f79f125c60d57f28d45ea8700293"},{"author":{"_account_id":11604,"name":"sean mooney","email":"smooney@redhat.com","username":"sean-k-mooney"},"change_message_id":"396121933979e983fbe3d2088c3375d448481a7e","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":2,"id":"3d87af80_6a39fbe1","in_reply_to":"dd805289_171b3cf9","updated":"2026-04-10 17:12:35.000000000","message":"so i built a quick simulator for soemthign esle in nova \n\nhttps://review.opendev.org/c/openstack/nova/+/980641\ni would prefer if we did soemting liek that here.\n\nwe shoudl be able to build the xmls and metric data programmaticly which will make maintianing it and simulating diffent senarios simpelr over time.\n\nironic apprently also has speical helpers for thesting the api by create dbs with diffent contnet to test sclaing so we coudl look at that too.","commit_id":"7ccf49d37d44f79f125c60d57f28d45ea8700293"},{"author":{"_account_id":30002,"name":"Douglas Viroel","email":"viroel@gmail.com","username":"dviroel"},"change_message_id":"544cccd9f1e136858edc00eb5b21c553ce4b7287","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":2,"id":"dd805289_171b3cf9","in_reply_to":"e2c78a41_927c1dd4","updated":"2026-03-20 17:23:03.000000000","message":"yeah, the browser can\u0027t even load the larger xml, and is also impossible to review. In my understanding, most of the date in the xml are duplicated info, that can be easily generated (from a template perhaps as sean mentioned), so we would just need to provide the amount of nodes/instances. I already had to edit the current xmls in watcher repos, to add a new attribute and was a nightmare already.","commit_id":"7ccf49d37d44f79f125c60d57f28d45ea8700293"}],"watcher/tests/functional/fixtures/nova_api_emulator.py":[{"author":{"_account_id":30002,"name":"Douglas Viroel","email":"viroel@gmail.com","username":"dviroel"},"change_message_id":"544cccd9f1e136858edc00eb5b21c553ce4b7287","unresolved":true,"context_lines":[{"line_number":85,"context_line":"                    \"ram\": inst_ram,"},{"line_number":86,"context_line":"                    \"disk\": inst_disk,"},{"line_number":87,"context_line":"                },"},{"line_number":88,"context_line":"                \"created\": \"2025-08-27T09:49:58-05:00\","},{"line_number":89,"context_line":"                \"updated\": \"2025-08-27T09:49:58-05:00\","},{"line_number":90,"context_line":"                \"addresses\": {},"},{"line_number":91,"context_line":"                \"accessIPv4\": \"\","},{"line_number":92,"context_line":"                \"accessIPv6\": \"\","},{"line_number":93,"context_line":"                \"links\": [],"},{"line_number":94,"context_line":"                \"OS-DCF:diskConfig\": \"AUTO\","},{"line_number":95,"context_line":"                \"progress\": 0,"},{"line_number":96,"context_line":"                \"OS-EXT-STS:power_state\": 1,"},{"line_number":97,"context_line":"                \"OS-EXT-STS:vm_state\": state,"},{"line_number":98,"context_line":"                \"OS-EXT-STS:task_state\": None,"},{"line_number":99,"context_line":"                \"OS-EXT-SRV-ATTR:host\": hostname,"},{"line_number":100,"context_line":"                \"OS-EXT-SRV-ATTR:instance_name\": inst.attrib[\"name\"],"},{"line_number":101,"context_line":"                \"OS-EXT-SRV-ATTR:hypervisor_hostname\": hostname,"},{"line_number":102,"context_line":"                \"OS-EXT-AZ:availability_zone\": \"nova\","},{"line_number":103,"context_line":"                \"config_drive\": \"\","},{"line_number":104,"context_line":"                \"key_name\": None,"},{"line_number":105,"context_line":"                \"security_groups\": [{\"name\": \"default\"}],"},{"line_number":106,"context_line":"                \"os-extended-volumes:volumes_attached\": [],"},{"line_number":107,"context_line":"                \"locked\": inst.attrib.get(\"locked\", \"False\") \u003d\u003d \"True\","},{"line_number":108,"context_line":"            }"},{"line_number":109,"context_line":""},{"line_number":110,"context_line":"            SERVERS_BY_ID[server[\"id\"]] \u003d server"}],"source_content_type":"text/x-python","patch_set":2,"id":"0eee5a09_db501606","line":107,"range":{"start_line":88,"start_character":0,"end_line":107,"end_character":71},"updated":"2026-03-20 17:23:03.000000000","message":"i understand that the idea is not to test watcher-nova integration, but at the same time, it is harder to maintain these attributes in sync with current nova api. I believe that at some point they will diverge and will require maintenance","commit_id":"7ccf49d37d44f79f125c60d57f28d45ea8700293"}]}