)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":8864,"name":"Artom Lifshitz","email":"notartom@gmail.com","username":"artom"},"change_message_id":"5a4c3830acde095f65462cdcfcc4a334f60a1bbe","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"5258145c_fa5034ac","updated":"2023-03-13 21:12:25.000000000","message":"@Amit - I\u0027ve solved the \"we\u0027re not printing any error message\" question, I think - downstream, we\u0027ll need this backport: https://review.opendev.org/c/openstack/nova/+/807358","commit_id":"c28dceff2138f6a0e77a14486bf4ea1d7866c591"},{"author":{"_account_id":8864,"name":"Artom Lifshitz","email":"notartom@gmail.com","username":"artom"},"change_message_id":"8b1aa85bde1466fe9cc06e8c2ebc240d25e8f5c8","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"808fc66c_eca95feb","updated":"2023-03-13 21:03:25.000000000","message":"James and Jason (Red Hat QE) played around with this, and the results is that the logging configuration that\u0027s currently in the code (see my other comment) results in a log file being created (inside the container, since this is a TripleO deployment) called /var/log/nova/nova-manage.log. When the refresh fails, the error is logged in that log file, but nothing is printed on standard output. I think the fix here should be to always pring an error if we encounter it (we can keep the log file as an extra audit trail, but relying on users to figure out that they need to `echo $?` to see our return code and know whether we failed, then `tail \u003clog file\u003e` to see the error message, is bad UX).","commit_id":"c28dceff2138f6a0e77a14486bf4ea1d7866c591"},{"author":{"_account_id":8864,"name":"Artom Lifshitz","email":"notartom@gmail.com","username":"artom"},"change_message_id":"cf6d3628285419f33ddfb3ab8db4f51efabfb224","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"de61749a_5392d1b8","updated":"2023-03-13 21:14:24.000000000","message":"So - tl;dr, because this is getting confusing even to me - I think we\u0027re good to keep the \"please see log for more information\" (IOW, we don\u0027t need this proposed patch), but we need the patch that I pasted in my previous comment in our downstream backport of this.","commit_id":"c28dceff2138f6a0e77a14486bf4ea1d7866c591"},{"author":{"_account_id":34860,"name":"Amit Uniyal","email":"auniyal@redhat.com","username":"auniyal"},"change_message_id":"488cb27f601a89d064071e3ecfb33eb6f8aa7401","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"ec2a8081_b546d6a6","updated":"2023-02-20 06:25:48.000000000","message":"recheck create multi nic server failed (unrelated)","commit_id":"c28dceff2138f6a0e77a14486bf4ea1d7866c591"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"7621dda37fc07a7454c18390567d4293f36d76a2","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"e87ef6bf_253472e7","in_reply_to":"03407198_53230e04","updated":"2023-03-15 13:58:29.000000000","message":"I don\u0027t think this is a python logging thing, but more likely an oslo.log thing. I just tried and couldn\u0027t get any discernible difference with a combination of non-tty stdio, so I think there must be some flag somewhere. That makes it even more important to only report \"check the log\" if that flag or config is set, IMHO.","commit_id":"c28dceff2138f6a0e77a14486bf4ea1d7866c591"},{"author":{"_account_id":34860,"name":"Amit Uniyal","email":"auniyal@redhat.com","username":"auniyal"},"change_message_id":"ef361ddd24fbcce115d90e4b34b23ddc061b212f","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"1dc22045_605abc4c","in_reply_to":"5258145c_fa5034ac","updated":"2023-03-14 05:17:02.000000000","message":"Ack Artom, will create backport patches for same.","commit_id":"c28dceff2138f6a0e77a14486bf4ea1d7866c591"},{"author":{"_account_id":34860,"name":"Amit Uniyal","email":"auniyal@redhat.com","username":"auniyal"},"change_message_id":"ef361ddd24fbcce115d90e4b34b23ddc061b212f","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"77b3f5d3_bd08a9ea","in_reply_to":"5ec60b06_345e445a","updated":"2023-03-14 05:17:02.000000000","message":"Ack","commit_id":"c28dceff2138f6a0e77a14486bf4ea1d7866c591"},{"author":{"_account_id":4393,"name":"Dan Smith","email":"dms@danplanet.com","username":"danms"},"change_message_id":"19eb8388ed6b2eeea2ac7826a7186c5aedd33c29","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"a21572a0_aded8a56","in_reply_to":"77b3f5d3_bd08a9ea","updated":"2023-03-14 13:33:49.000000000","message":"Is this based on logging config, or is it triggered when stdio is not a tty? Personally I think the error message pointing to a log always for an interactive tool that usually doesn\u0027t have that log is a terrible idea. If we can tell whether or not we\u0027re generating the log, then we could adjust the output. If we can\u0027t I\u0027d much prefer that we take the mainstream usage of nova-manage as an interactive tool as the preferred and expected behavior.","commit_id":"c28dceff2138f6a0e77a14486bf4ea1d7866c591"},{"author":{"_account_id":4690,"name":"melanie witt","display_name":"melwitt","email":"melwittt@gmail.com","username":"melwitt"},"change_message_id":"b6e5c444b6b421f0d8e79b31f8dff778e144c869","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"03407198_53230e04","in_reply_to":"a21572a0_aded8a56","updated":"2023-03-15 00:34:50.000000000","message":"I would like to know as well because python logging makes no sense to me, so I haven\u0027t been able to find why or how sometimes there\u0027s nova-manage.log and sometimes not.","commit_id":"c28dceff2138f6a0e77a14486bf4ea1d7866c591"},{"author":{"_account_id":4690,"name":"melanie witt","display_name":"melwitt","email":"melwittt@gmail.com","username":"melwitt"},"change_message_id":"010ab0f69af62fca9fae8cc54f9d849d1bea5f32","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"5ec60b06_345e445a","in_reply_to":"de61749a_5392d1b8","updated":"2023-03-14 00:12:57.000000000","message":"I agree with you ... logging is configured in nova/cmd/manage.py as you noted and it looks like it is intended for there to be a nova-manage.log target, for example: https://docs.openstack.org/nova/latest/admin/configuration/logs.html.\n\nAlso agree that the linked patch is needed to always also print an error to the console.\n\nI don\u0027t have experience looking at nova-manage.log in a devstack environment because it isn\u0027t common for me to hit errors the rare times I run nova-manage. So I couldn\u0027t tell you if it appears there or not. I have however seen nova-manage.log in tripleo deployments, as you have already mentioned.","commit_id":"c28dceff2138f6a0e77a14486bf4ea1d7866c591"}],"nova/cmd/manage.py":[{"author":{"_account_id":8864,"name":"Artom Lifshitz","email":"notartom@gmail.com","username":"artom"},"change_message_id":"ddf2ed99e4d2e07614a72ba33e9c259dfc34fa74","unresolved":true,"context_lines":[{"line_number":3395,"context_line":"    logging.set_defaults("},{"line_number":3396,"context_line":"        default_log_levels\u003dlogging.get_default_log_levels() +"},{"line_number":3397,"context_line":"        _EXTRA_DEFAULT_LOG_LEVELS)"},{"line_number":3398,"context_line":"    logging.setup(CONF, \"nova\")"},{"line_number":3399,"context_line":"    objects.register_all()"},{"line_number":3400,"context_line":""},{"line_number":3401,"context_line":"    if CONF.category.name \u003d\u003d \"version\":"}],"source_content_type":"text/x-python","patch_set":1,"id":"57827536_ab96d9d1","line":3398,"updated":"2023-03-13 21:11:56.000000000","message":"\u003e This logging config \u003c","commit_id":"c28dceff2138f6a0e77a14486bf4ea1d7866c591"}]}