When it works, nobody cares

About two weeks ago, we got an alert from NoaNet that a portion of our primary circuit out of Kitsap was logging physical layer coding errors, only on the outbound path. The errors went away. A few days later they started accumulating again. We are talking a few thousand error packets per day. During peak hours we move over 500,000 packets per second! These few thousand errors are truly specks of dirt on the eye of the bug hitting the windshield types of numbers. However, if that one dropped packet was the “click” that some gamer lost his tournament over, or someone was in the middle of a difficult conversation and it got hiccupped, it isn’t OK with me.

Today (6/5) around 1PM KPUD, NoaNet and Net253 did a coordinated maintenance. At 2:05PM we switched back to our primary link (which is a dedicated wavelength of light that goes under Puget Sound directly to the Westin Data Center in Seattle). Because we have extra paths, we were able to take our primary link to Seattle out of service, live, without dropping any active internet sessions (zoom/teams/streaming/gaming)

We are able to make planned service events to get back to perfection without large disruption to the subscribers. I find that pretty darn awesome. This is part of why we provide the best value on the KPUD open access network.