Issue in stop_start_servers VM dynamic workload

[Sanjay Chari]

There is an edge case in the stop_start_servers dynamic workload where the workload has outdated information about the floating IP of a VM. The workload takes information about each server from a list of servers that is fetched at the beginning of the workload. During the workload, some of the servers in the list could swap their floating IPs with other servers in another iteration. In this case, the workload still tries to ping the old floating IP, and the workload fails as the server to which the floating IP was reassigned is not switched off.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.opendev.org/c/x/browbeat/+/865258

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/c/x/browbeat/+/865258
Committed: https://opendev.org/x/browbeat/commit/78353343bed07a20cd4e9c5f4341127545d5a681
Submitter: "Zuul (22348)"
Branch: master

commit 78353343bed07a20cd4e9c5f4341127545d5a681
Author: Sanjay Chari <email address hidden>
Date: Tue Nov 22 15:35:30 2022 +0530

    Dynamic Workloads: Fix stop_start_servers

    Closes-Bug: #1997353
    Change-Id: Iea8da816e1b9c34627248f8e6f26cc3aaf044036