As a User, when an error occurs I want an easy way to diagnose and remedy.  The current process of viewing multiple log files (perhaps on different clusters) and sorting through which statements are relevant to my specific migration is difficult and consumes more time than I want to spend.

Thoughts to consider:

• Would presenting a 'virtual view' of logs help? what if we could
  • parse all of the various logs of interest and correlate with timestamps
  • allow realtime adjustments to filter log levels:  Warn, Info, Debug, etc
  • Filter on other characteristics such as:
    • Specific migration
    • Cluster (source, host, destination, etc)
    • Component (Velero, DVM, mig-controller, operator)
    • Regex
• Perhaps we could implement a structured logging kind of approach in our components to add key/value pairs to help aid the parsing/virtual view above
• Would emitting 'events' help to aid debugging?