Closed cdumez closed 6 years ago
Tracing back, it looks like that was introduced in original implementation in Blink. Testing in FF, I don't see max-age or any other cache-related request headers. @nolanlawson @toddreifsteck Edge?
Seems to me that sendBeacon() should bypass the cache.
Curious, what makes you say that? It's not obvious to me why we should special case sendBeacon; seems like the decision should rest with the endpoint. Also, if we special case sendBeacon, then we should also think about the "keepalive" flag in Fetch.. and there I think it makes even less sense?
/cc @tyoshino
nudge
@cdumez any followup thoughts or questions on this one? As stated above, I don't believe we should be special casing this. As next steps, I propose we (a) open a Chrome bug to remove max-age=0 and (b) close this with no action on the spec. Does that sound reasonable?
Ok.
Thanks! Filed Chrome bug: https://crbug.com/755725
Should sendBeacon set the FetchRequest's cacheMode to "no-cache"?
Based on code introspection, Blink seems to bypass the cache by adding "max-age=0" Cache-control header on the request. I don't know about Gecko.
Seems to me that sendBeacon() should bypass the cache.