Hi, I am working with the bug 1829925 "instance creating via horizon failed Edit" [0] and I need someone's help to confirm an induced behavior can be replicated in another StarlingX deployment, can you please run the script provided in any of your StarlingX deployments? Steps to Reproduce 1. StarlingX: Two Node System Deployed, Provisioned, Healthy 2. Instance Image: CentOS-7-x86_64-GenericCloud-1905.qcow2 3. Execute script openstack_instance.sh [1] In summary these are the high level errors I am seeing after more than 200 cycles of a single instance being created and removed: - Unable to establish connection to http://nova-api-proxy.openstack.svc.cluster.local:8774 - Failed to discover available identity versions when contacting http://keystone.openstack.svc.cluster.local/v3 - controller-1 was set to active, controller-0 was set to Standby - Platform Memory threshold exceeded ; threshold 80.00%, actual 89.23% - Instance centos owned by admin has failed to schedule For the complete history without a debug strategy yet, please see [2]. [0] https://bugs.launchpad.net/starlingx/+bug/1829925 [1] https://gist.github.com/xe1gyq/0757b9220051aa92569292de7252e0cb [2] https://github.com/xe1gyq/starlingx/blob/master/bugs/1829925_a.md