I bind the node IP of CF and request the same resource, but hit and miss will occur alternately. In the case of hit, the age returned is not increasing, indicating that it is not the resource returned by the same machine. What are the rules for requesting to reach the node of CF and a specific machine, random? Can the request be hashed to a fixed machine according to the URL to improve the hit rate?
Thank you very much
I suspect it’s a catch-22. Requests are distributed across the various servers in a POP. But I suspect if you set a cookie, it will route through the same server. But the presence of a cookie often disables caching. It’d certainly be worth an experiment to see how it behaves.
Realistically, anything that can be a HIT should be cached by the browser, so it shouldn’t be making another request for the same resource.
Ultimately, your cache hit ratio is determined by the amount of traffic you get through that node. If you don’t get much traffic, those HITs will become MISSes within a few hours no matter what you do for your individual requests.
This topic was automatically closed 15 days after the last reply. New replies are no longer allowed.