Description of problem ======================* Originally reported here: https://bugs.launchpad.net/nova/+bug/1808951 The issue is around using of eventlet.monkey_patch even if running under wsgi[1]. Eventlet Best practices suggest to not run(skip) monkey_patch when running behind wsgi[3] but with nova it's not the case. Expected result =============== Nova api should response successfully. Actual result ============= Nova api returns Recursion Error. [1] https://github.com/openstack/nova/blob/a674241d0f026b546b38ec77ab472e916805c679/nova/api/openstack/wsgi_app.py#L32 [3] https://review.openstack.org/#/c/154642/2/specs/eventlet-best-practices.rst@54 Other useful links ================== https://github.com/eventlet/eventlet/issues/371
The eventlet folks seems to be treating this as an eventlet bug against python 3.7: https://github.com/eventlet/eventlet/pull/531
Please can you try to reproduce this against a fully update Fedora? I haven't been able to reproduce this using the only test which was demonstrating it, and the eventlet bug suggests that it was being fixed in eventlet. Note that Nova uses eventlet explicitly in a number of places which can be called from services running both under and not under eventlet, so we must (until that changes, at least) always monkey patch. For example, disabling monkey patching would most likely break multi-cell instance listing.
The fix for this has merged upstream, and a backport to stein proposed.
*** Bug 1686817 has been marked as a duplicate of this bug. ***
This was triggered upstream by: GET /v2.1/servers/detail?all_tenants=True&deleted=True So I guess to reproduce we should execute a multi-tenant instance list returning at least 1 instance, and verify that it doesn't return an error.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHEA-2019:2811
The needinfo request[s] on this closed bug have been removed as they have been unresolved for 1000 days