Hi CloudShare team,
I like the point where infrastructure has to work for somebody on the other end of it. In your case, a customer expects a lab to be ready, the network to work and the environment to behave enough like the real thing to teach them something. When any of those pieces breaks, tracking it down is the work I enjoy.
At Front Range Internet I ran NOC operations and handled engineering escalations for hosting and colocation customers. I built diagnostics operators could use, led SOC 1 and SOC 2 audits, and migrated workloads from Hyper-V and VMware ESXi into Docker and Kubernetes. At Technomancy IT, I still handle the issues that make it past the first round of troubleshooting. For one client in Wray I modernized aging systems and set up Windows Server, Active Directory and VPN access. I also built a backup check that restores a PostgreSQL database in a disposable environment, because a completed backup job doesn't mean much if you can't restore it.
I'd bring that same approach to CloudShare: find the failing layer, get the customer's environment working again, and leave behind a check or a runbook for the next person.
Thanks,
Chad