DaanHoogland commented on pull request #5552: URL: https://github.com/apache/cloudstack/pull/5552#issuecomment-937792230
> > nice idea @mlsorensen , I understand this are hard coded hooks/callbacks , not configurable!? > > This will probably solve a lot of old standing issues. I wonder if this could be generalised to solve some of the vmware storage issues that occur on disconnect from the vcenters as well. > > Thanks @DaanHoogland - yes the hooks are intended for a way for CommandWrappers to rollback or change the work they're doing during a disconnect, it's tied to the code that is operating in the CommandWrapper so it's tightly coupled. They are configurable in the sense that the CommandWrapper is free to pass variables to the hooks as it creates them - in the cancel migration example the hook passes the name of the VM to use for migration cancellation in the event the hook is called. > > I'm less familiar with the VMware case, but I recall that the hypervisor plugin is server-side. I imagine something along these lines could be replicated, but only to some degree if we can't have code on the vCenter side to react to outages. Just a guess. hm, maybe the architecture is too different. It would be comparible to the agent no longer being able to access libvirt/qemu after it had issued a migrate command. On vmware the disconnect would be between the agent and vcenter. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
