A tracker for Command infrastructures related tasks
Moving from 4.0 alpha to 4.0 beta since 4.0 alpha has been already released and bug is not ON_QA.
Can you propose some verification steps for this one please?
Basically, the changes of the command infrastructure should be verified by testing the flows which use command-coordinator: 1. Upgrade host manager, when the host is in 'Up' status. 2. Live storage migration 3. V2V Liron, Arik, any other flows that should be covered as part of verifying the series of patches which modified CoCo ? Any guidelines to combine tests with compensation ?
From the storage team prespective I'd test the LSM flow on different scenarios (postive, migration of few vm disks and fail one migration, engine restart) and test flows the combine async task operation with no async task operation (for example, creation of a snapshot combining image and cinder disks). As regards to compensation, I'd fail command that using the command-coordinator infrastructure by performing engine restart and by causing failure to the executed vdsm operation. Generally speaking, I'd suggest to coordinate with the other QE teams to see what postivie/negative flows they are testing and decide what (if needed) we are going to test here.
Tested upgrade host manager and it's working as expected apart from small logging issue BZ#1361511. As this Infra part of this was already verified rest is on storage, moving qa_contact to storage team.
So only v2v left.
Import from VMware, during import "initialize" progress, restart ovirt-engine and verify paddle lock icon is still appears on VM and verify import progress continues and completed successfully. Version: 4.0.2.2-0.1.el7ev