Michal Zalewski wrote: > On Wed, 3 Jan 2007, William A. Rowe, Jr. wrote: > >> If you have an issue with this behavior, of HTTP, then you have an issue >> with the behavior under FTP or a host of other protocols. > > Not really; see above. These are typically well known, preventable by > configuring server-side limits, and require a much higher overhead. On the matter of your 1GB window (which is, again, the real issue), you have any examples of a kernel that permits that large a sliding window buffer by default, or any examples where this is commonly set quite that high? If not, what is your logic for negotiating a packet window larger than the OS will permit?
