- From: Patrick McManus <mcmanus@appliedtheory.com>
- Date: Wed, 27 Nov 1996 12:08:43 -0500 (EST)
- To: Jeffrey Mogul <mogul@pa.dec.com>
- Cc: mcmanus@appliedtheory.com, http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
In a previous episode Jeffrey Mogul said... :: :: In the absence of a specific server request for any minimum return :: count, the proxy cannot know if the report is worth sending. Some :: servers may not believe that a return connection is worth the :: overhead of receiving a 1/0 report, just as some servers may insist :: on seeing a 0/0 report to know that the proxy's 'best-effort' is :: succeeding. [..] :: :: It would not complicate things too much to add a Meter response-directive :: along the lines of :: Meter: want-report= MIN/MAX :: as long as the default was to NOT send this, and that the default :: MIN is equal to 1 and the default MAX is equal to infinity. :: This satisfies my interest in the matter. The only point I might question in the fact that we have lost the use/re-use granularity in this response-directive that is maintained elsewhere throughout the proposal. This breaks that pattern relying on aggregates instead, but I can't see an alternative short of 3 different directives (report-use-count, report-reuse-count, report-aggregate-count) and I don't really like that as an option. :: I'm really reluctant to add something from the proxy to the server :: to indicate the proxy's "willingness to report" limits, since it's :: hard to imagine that a proxy cache with non-infinite disk space :: could actually guarantee a minimum. I'm convinced on this issue. The above mechanism should be sufficient anyhow. -Patrick -- Patrick R. McManus - Applied Theory Communications - Software Engineering http://pat.appliedtheory.com/~mcmanus Programmer Analyst mcmanus@AppliedTheory.com 'Prince of Pollywood' Standards, today! *** - You Kill Nostalgia, Xenophobic Fears. It's Now or Neverland. - ***
Received on Wednesday, 27 November 1996 09:31:57 UTC