I know there are more posts about this - but I’ve never encountered this before. I have a paid (pro) account and I suddenly noticed last week or so that my site was slower than usual. So checking the trace, I found that I was connecting to the LHR colo. This is not a normal situation - I ALWAYS used to connect to AMS (which really is a lot faster for my location).
So the question is; why am I now suddenly connecting to LHR? It has been like that for at least a week now… I checked the status page but AMS seems healthy.
The site in question is a CF Worker Site that gets it’s resources from a KV store. Not sure if that information is useful, but it might?
Is your site really much slower? What latency are you getting to LHR compared to AMS? If you ping 1.1.1.1 you should reach AMS, compare that with pinging your domain.
I realise that - however, we always (and I’m talking years here) where connecting to AMS. I also did a test (webpagetest.org) from a datacenter I know actually resides in an AMS DC - and that one still connects to LHR.
The difference between LHR and AMS is about 100 - 200ms (total load time homepage)… Ping time difference is 20ms (for 1.1.1.1) and 30ms (to the site). Slower is relative, however, we’ve spend a huge amount of effort in performance and it’s a shame that (suddenly?) CF decides to route to a less than ideal colo costing us precious ms…
The real issue here is - why is it suddenly doing this…?
As I said, it could be for a bunch of reasons, maybe AMS is under heavy load right now, for whatever reason. It will likely prioritize traffic for Enterprise and Business customers first
