On 2023-03-31 in the late evening, the UPS at LEB turned off. This had the following impacts:
DNS (recursive and authoritative) was unavailable, because the only live servers on network are at LEB
Blueiris went down, because this is not a HA application
All of the monitoring went down (prometheus+grafana+alertmanager)
The net was split into two areas
atl, azo, leb, ftn, and ret; this net had internet access
mno, hil, crw; this net had no uplink available
This is obvious when you look at the following diagram and consider what happens if LEB is unavailable:
On 2023-03-31 in the late evening, the UPS at LEB turned off. This had the following impacts:
... work in progress ...