This feature determines what happens when a request results in a partial cache miss on an edge server. A partial cache miss describes the cache status for an asset that was not completely downloaded to an edge server. If an asset is only partially cached on an edge server, then the next request for that asset will be forwarded again to the origin server.
Important: This feature is not available for the HTTP Small or the ADN platform. The typical traffic on both of these platforms consists of relatively small assets. The size of the assets served through these platforms helps mitigate the effects of partial cache misses, since the next request will typically result in the asset being cached on that POP.
A partial cache miss typically occurs after a user aborts a download or for assets that are solely requested using HTTP range requests. This feature is most useful for large assets where users will not typically download them from start to finish (e.g., videos). As a result, this feature is enabled by default on the HTTP Large platform. It is disabled on all other platforms.
It is recommended to leave the default configuration for the HTTP Large platform, since it will reduce the load on your customer origin server and increase the speed at which your customers download your content.
Important: Due to the manner in which cache settings are tracked, this feature cannot be associated with the following match options: Edge Cname, Request Header Literal, Request Header Wildcard, URL Query Literal, and URL Query Wildcard.
Value | Result |
---|---|
Enabled | Restores the default behavior. The default behavior is to force the edge server to initiate a background fetch of the asset from the origin server. After which, the asset will be in the edge server's local cache. |
Disabled | Prevents an edge server from performing a background fetch for the asset. This means that the next request for that asset from that region will cause an edge server to request it from the customer origin server. |
Default Behavior: Enabled.