in the last couple of days, there was a wide spread network outage.

how big? a large portion of the south island was offline for about 12 hours

the annoying thing is that it tookout my network internally, thinking about it somewhat after the event, it’s obvious that that would happen.

my DNS resolver was failing because upstream dns was not there.

How to Fix it

well since dns was the problem, First i thought i’d set up a dns server that i can put in place when theres a network failure.

there were some down sides to this..

1. it requires manual intervention.
2. it needs to be kept up to date.

well that’s no way to live, how about something a little more automatic.

right, no maintenance how can i handle this?

how about i replicate the feature of upstreaam dns. put something on the wan port of my router to just provide name resolution. that could work, sure i’d get garbage back for external sites, but there’s nothing there anyway.

what’s the point in that, well making upstream dns work lets my internal dns operate (theoretically) and i keep my internal network working.

Fully Automating it

how can i drop the device in automatically? wellll i have a WAN2 port that can be enabled as a failover link..

so i’ll feed that back into my managed switch, wait what? surely it goes to the uplink device, eventually yes.. but here’s the cunning bit, on the other end of my switch, there is a port configured with POE, going to a raspberry pi. It just keeps the cables tidy

Did it work?

we’ll find out, stay tuned

so, yeah