Opeyemi Adeniji

Thank you Abhinav. "Time-to-page, not severity" is a much better way to put it than the way I had described it. You are right that the current report is a snapshot so it can see that a disk is at 92 percent but it cannot tell how fast that disk is filling. The tool already saves a timestamped report on every run so I can use those earlier runs to estimate how long each disk has until it is full and rank hosts by that estimate. I will share what I land on in the next version.

Thanks Abhinav, glad those two lines landed. The report ranked db-02 (UNREACHABLE) first, then the three CRIT hosts, then the WARN, then OK. So nothing hid below something worse across levels. Within a level, though, the order is only alphabetical, so web-07 (disk 92% full) sits below app-11 (app server down) because “app” sorts first, not because the report judged it less severe. That’s a limitation. A better version would rank inside each level by number of issues or by weight (service down above a nearly full disk). Good question, it’s going on my list for the next release.