)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":33987,"name":"Pedro Monteiro Azevedo de Moura Almeida","display_name":"Pedro Almeida","email":"pedro.monteiroazevedodemouraalmeida@windriver.com","username":"pedroalmd"},"change_message_id":"2adbec5b77d159cc20a6d2a3f5ee2b3952c305f3","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":9,"id":"55c7a93b_4020cb89","updated":"2022-04-07 12:50:23.000000000","message":"LGTM","commit_id":"1c408c602900549042b3b2a7c43405148690fcc6"},{"author":{"_account_id":33594,"name":"Thales Elero Cervi","display_name":"Thales Cervi","email":"thaleselero.cervi@windriver.com","username":"tcervi"},"change_message_id":"77d06f1de351c092be41fa10656126314390780f","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":9,"id":"88c9c448_62f8240d","updated":"2022-04-07 13:04:56.000000000","message":"Since the commit that introduced this glance-api extended grace period has no additional info about why it should be so high and OpenStack Helm already merged a change that decreases it (https://review.opendev.org/c/openstack/openstack-helm/+/836329) I think it is a safe change.","commit_id":"1c408c602900549042b3b2a7c43405148690fcc6"},{"author":{"_account_id":27498,"name":"Lucas Cavalcante","email":"lucasmedeiros.cavalcante@windriver.com","username":"lcavalca"},"change_message_id":"3f294cf1f080a3f9cd98ece103b2ed0263b24286","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":9,"id":"e0595caa_9ce7bf1e","updated":"2022-04-08 14:33:32.000000000","message":"This really seems like someone was trying to run away from a problem as a 10m SIGTERM timeout for a simple service like Glance is very weird. So I really dont see a problem in reducing this timeout.\n\nHowever if 10m are passing and Glance is receiving a SIGKILL I would like to know why,do you know why glance cant actually gracefully terminate?\n\nbasically, how do we know if we would not run into a new problem SIGKILLing glance everytime?","commit_id":"1c408c602900549042b3b2a7c43405148690fcc6"},{"author":{"_account_id":33618,"name":"Arthur Luz de Ávila","display_name":"Arthur Luz de Avila","email":"Arthur.LuzdeAvila@windriver.com","username":"luzdeav"},"change_message_id":"d6ff85ac817c74843abf869f9ac25bf6cd455e5b","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":9,"id":"f2db8802_b745e39c","in_reply_to":"e0595caa_9ce7bf1e","updated":"2022-04-08 14:49:00.000000000","message":"Well, as I investigated it, the Glance don\u0027t gracefully terminate due to a authenticate with kube-apiserver  with the error: \"Unable to authenticate the request\" err\u003d\"[invalid bearer token, Token has been invalidated]\"\n\nThe logs doesn\u0027t have much information about it, but when I was researching it I saw that it can be a timing error between the communication of the kube-apiserver and the glance-api when is in terminating state. I ran the test that test the gracefully terminate for 25x and couldn\u0027t see the problem. Normally, the glance api terminate gracefully before 30sec, but if the error of Unable to authenticate appear, it kill faster to keep the things working like the other services does.","commit_id":"1c408c602900549042b3b2a7c43405148690fcc6"}]}