)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"26199ed8c6e4d884248ce20b124aa44422bc8291","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"b70e8e07_9a9c3b31","updated":"2026-09-01 15:04:57.000000000","message":"oh sh*t the eventlet based functional job has a lot more leaking tasks :/","commit_id":"2cec50f11fe9efffdc013ed5a4a017f92e7df31b"},{"author":{"_account_id":8556,"name":"Ghanshyam Maan","display_name":"Ghanshyam Maan","email":"gmaan.os14@gmail.com","username":"ghanshyam"},"change_message_id":"29685a4e80fea73fbc123b1ccde60b2f323d6b0f","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":3,"id":"7f7c66c7_4f2a86a5","updated":"2026-09-04 05:11:45.000000000","message":"I debugged the failure, which is a valid test failure, not a production bug. This is not related to eventlet, and the test will fail in both modes; in threading mode, the job passes because this test is skipped there.\n\nAdded the detailed debugging/root cause in bug https://bugs.launchpad.net/nova/+bug/2166439","commit_id":"8a2315ff37c6e6c8c52b3016ae55233efd1e8d17"},{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"d2833ddb67713b5432a4e7a2313be87a8ca01711","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":3,"id":"3da692b3_f59252da","updated":"2026-09-03 08:54:19.000000000","message":"We still have to look at the eventlet based job\u0027s failures if they are test problems or different leaks.","commit_id":"8a2315ff37c6e6c8c52b3016ae55233efd1e8d17"},{"author":{"_account_id":8556,"name":"Ghanshyam Maan","display_name":"Ghanshyam Maan","email":"gmaan.os14@gmail.com","username":"ghanshyam"},"change_message_id":"681808fbc22ae7552b1ffcdf741320956fdc948c","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":4,"id":"782276f3_c4ff981e","updated":"2026-09-04 05:59:35.000000000","message":"@gibizer@gmail.com, as it is late for me now, the default value should also reproduce this here (I can reproduce it locally with manager_shutdown_timeout\u003d35 sec).\n\nI proposed the fix for the failure and tested localy- https://review.opendev.org/c/openstack/nova/+/1003910\n\nOnce it fails here with default manager_shutdown_timeout, can you rebase this on top of that fix to see if anything else fails (maybe stack other live migration task leak fix also 1003386 -\u003e 1003910 -\u003e this change )","commit_id":"40d1133235d9298aaac9cc9f1957d728c925ac0e"}],"nova/tests/fixtures/conf.py":[{"author":{"_account_id":8556,"name":"Ghanshyam Maan","display_name":"Ghanshyam Maan","email":"gmaan.os14@gmail.com","username":"ghanshyam"},"change_message_id":"29685a4e80fea73fbc123b1ccde60b2f323d6b0f","unresolved":true,"context_lines":[{"line_number":70,"context_line":"        # This does not slow down the good test cases that are finishing"},{"line_number":71,"context_line":"        # cleanly as the wait is conditional in service.stop() on the shutdown"},{"line_number":72,"context_line":"        # thread exit."},{"line_number":73,"context_line":"        self.conf.set_default(\u0027manager_shutdown_timeout\u0027, 300)"},{"line_number":74,"context_line":""},{"line_number":75,"context_line":"        # oslo.limit requires endpoint_id since 2.3.0"},{"line_number":76,"context_line":"        self.conf.set_default(\u0027endpoint_id\u0027, \u0027ENDPOINT_ID\u0027, group\u003d\u0027oslo_limit\u0027)"}],"source_content_type":"text/x-python","patch_set":3,"id":"e39b3cbb_09258088","line":73,"range":{"start_line":73,"start_character":57,"end_line":73,"end_character":60},"updated":"2026-09-04 05:11:45.000000000","message":"exact 300 will not be in effect until we increase the graceful_shutdown_timeout. If manager_shutdown_timeout is greater than  graceful_shutdown_timeout, then we reduce then manager_shutdown_timeout  to graceful_shutdown_timeout - 10 which means manager_shutdown_timeout will be 170 in this case (default value of graceful_shutdown_timeout  180 -10 ).\n\nThat is why you can see in log that compute manager wait for only 150 (170-20) sec for in-progress tasks to complete.\n\nanyways, the value here does not matter as any higher manager_shutdown_timeout will unhide the failure,  but just to highlight that this value is not matching with log so can be confusing.\n\nlet\u0027s not set this at all and run our tests with defaults (both graceful_shutdown_timeout  as well as manager_shutdown_timeout)\n\n\n    2026-09-02 19:14:21,910 INFO [nova.compute.manager] Compute manager waiting up to 150 seconds for in-progress tasks to complete (20 seconds reserved for cleanup_host).","commit_id":"8a2315ff37c6e6c8c52b3016ae55233efd1e8d17"},{"author":{"_account_id":8556,"name":"Ghanshyam Maan","display_name":"Ghanshyam Maan","email":"gmaan.os14@gmail.com","username":"ghanshyam"},"change_message_id":"fda5cb6ef5d26555df78d9684b60fc246c39f2b3","unresolved":true,"context_lines":[{"line_number":70,"context_line":"        # This does not slow down the good test cases that are finishing"},{"line_number":71,"context_line":"        # cleanly as the wait is conditional in service.stop() on the shutdown"},{"line_number":72,"context_line":"        # thread exit."},{"line_number":73,"context_line":"        self.conf.set_default(\u0027manager_shutdown_timeout\u0027, 300)"},{"line_number":74,"context_line":""},{"line_number":75,"context_line":"        # oslo.limit requires endpoint_id since 2.3.0"},{"line_number":76,"context_line":"        self.conf.set_default(\u0027endpoint_id\u0027, \u0027ENDPOINT_ID\u0027, group\u003d\u0027oslo_limit\u0027)"}],"source_content_type":"text/x-python","patch_set":3,"id":"894e70b9_ee2842a8","line":73,"range":{"start_line":73,"start_character":57,"end_line":73,"end_character":60},"in_reply_to":"b851aa3c_2abaf365","updated":"2026-09-04 05:13:56.000000000","message":"btw, let me reproduce this bug by default timeouts and increasing the test timeout to higher than the shutdown timeouts","commit_id":"8a2315ff37c6e6c8c52b3016ae55233efd1e8d17"},{"author":{"_account_id":8556,"name":"Ghanshyam Maan","display_name":"Ghanshyam Maan","email":"gmaan.os14@gmail.com","username":"ghanshyam"},"change_message_id":"3db34361edf8dde5069f54b6e869f9a32b46ddb5","unresolved":true,"context_lines":[{"line_number":70,"context_line":"        # This does not slow down the good test cases that are finishing"},{"line_number":71,"context_line":"        # cleanly as the wait is conditional in service.stop() on the shutdown"},{"line_number":72,"context_line":"        # thread exit."},{"line_number":73,"context_line":"        self.conf.set_default(\u0027manager_shutdown_timeout\u0027, 300)"},{"line_number":74,"context_line":""},{"line_number":75,"context_line":"        # oslo.limit requires endpoint_id since 2.3.0"},{"line_number":76,"context_line":"        self.conf.set_default(\u0027endpoint_id\u0027, \u0027ENDPOINT_ID\u0027, group\u003d\u0027oslo_limit\u0027)"}],"source_content_type":"text/x-python","patch_set":3,"id":"b851aa3c_2abaf365","line":73,"range":{"start_line":73,"start_character":57,"end_line":73,"end_character":60},"in_reply_to":"e39b3cbb_09258088","updated":"2026-09-04 05:12:55.000000000","message":"other constraint to the higher timeout is the test timeout set in tox.ini, we need to increase that timeout also if we want to run functional tests with higher shutdown timeout.","commit_id":"8a2315ff37c6e6c8c52b3016ae55233efd1e8d17"}],"nova/tests/functional/compute/test_live_migration.py":[{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"d2833ddb67713b5432a4e7a2313be87a8ca01711","unresolved":true,"context_lines":[{"line_number":127,"context_line":"        * Assert that the original source host volume attachment remains"},{"line_number":128,"context_line":"        * Assert that the original src connection_info is in the bdm"},{"line_number":129,"context_line":"        \"\"\""},{"line_number":130,"context_line":""},{"line_number":131,"context_line":"        # Mock out driver.live_migration so that we always rollback"},{"line_number":132,"context_line":"        def _fake_live_migration_with_rollback("},{"line_number":133,"context_line":"                context, instance, dest, post_method, recover_method,"}],"source_content_type":"text/x-python","patch_set":3,"id":"bfdc3314_5e72e282","line":130,"updated":"2026-09-03 08:54:19.000000000","message":"drop it.","commit_id":"8a2315ff37c6e6c8c52b3016ae55233efd1e8d17"}],"nova/tests/functional/regressions/test_bug_1888395.py":[{"author":{"_account_id":9708,"name":"Balazs Gibizer","display_name":"gibi","email":"gibizer@gmail.com","username":"gibi"},"change_message_id":"d2833ddb67713b5432a4e7a2313be87a8ca01711","unresolved":true,"context_lines":[{"line_number":140,"context_line":"        dom.fail_job()"},{"line_number":141,"context_line":""},{"line_number":142,"context_line":"    def test_live_migration_rollback(self):"},{"line_number":143,"context_line":""},{"line_number":144,"context_line":"        self.server \u003d self._create_server("},{"line_number":145,"context_line":"            host\u003d\u0027start_host\u0027,"},{"line_number":146,"context_line":"            networks\u003d[{\u0027port\u0027: self.neutron.port_1[\u0027id\u0027]}])"}],"source_content_type":"text/x-python","patch_set":3,"id":"5edc387a_5ab1c4ac","line":143,"updated":"2026-09-03 08:54:19.000000000","message":"drop it","commit_id":"8a2315ff37c6e6c8c52b3016ae55233efd1e8d17"}]}