We can see that many Tempest tests failed on CI labs because of strange issues. On our personal QA labs these tests passed.
We need to investigate the root of the issue and fix it.
Starting from ISO #188, the periodic-testing-deploy-openstack-8.0-ubuntu-vlan-ceph-swift and periodic-testing-deploy-openstack-8.0-ubuntu-vxlan-lvm-swift jobs (https://jenkins.mosi.mirantis.net/view/Periodic%20(deployment)/) fail with numerous identical failures in Tempest tests, especially for Compute and Volume components.
To reproduce it, the environment based on ISO #201 has been deployed and Tempest tests have been run on it. The Tempest test reports for Compute and Volume components are attached in comment below.
Tempest test report for periodic-testing-deploy-openstack-8.0-ubuntu-vxlan-lvm-swift job (ISO # 201): https://jenkins.mosi.mirantis.net/job/periodic-testing-deploy-openstack-8.0-ubuntu-vxlan-lvm-swift/46/artifact/mos-deploy/logs/tempest-report.html
It was infra issue. Host with Tempest containers was rebooted and now all works fine.