One of my many domains is unstable with 404’s, 200’s, etc. The other domains are fine. The normal domain and this unstable domain are all built on the same infrastructure and there is no difference in infrastructure. I have enabled Bypass Cache to determine the cause, but the symptoms have not improved. Perhaps Bypass Caching is not working properly. Does anyone else have the same symptoms?
Digging deeper all domains (WordPress containers) are deployed on the same node, but only one of them (domain) is unstable.
From Consul’s point of view, the unstable container is HEALTHY.
Also, accessing the IP of the container directly is normal.
Without the proxy it’s hard to test, also if intermittent. 404s would only come from your origin so if that only happens when Cloudflare is enabled you’d have to see what was happening in the Cloudflare pipeline for those requests to make your origin return 404 for them.
No “node” is “assigned” to your domain, you will use the Cloudflare PoP your ISP routes you to which hopefully is a near one. You can check which one for your own requests by going to https://pendulum.golf/cdn-cgi/trace when it is proxied.