While the folks who work with our software, from a deployment management perspective, have reported a significant reduction in the amount of time they spend doing so (i.e., deploying new servers, upgrading existing servers, etc.), we haven’t attempted to actually quantify the amount of time and resources saved.

The Config Server handles remote deployment. Agents are remotely deployed via SSH. Once they’re in place, we remotely deploy servers to managed computers using the functionality provided by STAF.

With respect to your question about a watchdog / recovery service… For our servers, we leverage the Tanuki Java Service Wrapper. Used in combination with our Monitoring Server, we’re able to get a pretty good sense of what the servers in a deployment. We have not yet implemented similar functionality for our agents (i.e., an agent watchdog and a monitoring component for the Config Server Agent Manager service). Long-term, that’s something we’d definitely like to do.

For persistence, the Config Server currently uses XML. For now, given the amount of data it currently needs to deal with, XML works fine.

Thanks for the questions!