Hello all,
I have a tcp/ip server who's sole purpose is to spit back weather
imagery on the fly. It uses HTTP protocol so browsers can directly
access the server. We also have a javascript "player" page that will
get a bunch of images from my image server and loop through them with
the usual javascript "rollover" type code (preloading, swapping etc).
A typical image request looks like this:
http//someplace.blah:1234/map_19239230239_RADAR_frame0.jpg
The thing to note is the 19239230239 in the request. This is just a
javascript Date.GetTime() timestamp. I generate one of these for each
page load. This will fool Netscape 4.x and Internet Explorer x.x into
caching the imagery so that the rollover animation loop will go
forever. My problem comes with Netscape 6.x and 7.x... These browsers
insist on constantly hitting my server and checking to see if the
image has changed. As far as I'm concerned, after the first time the
loop is loaded in the page, it WILL NOT change. After awhile,
Netscape/Mozilla begins wildly querying my server and things get out
of hand. I've been experimenting with different things to put in my
HTTP header that gets sent back to the browser. Things like max-age,
and such. These seem to have no impact. Is there anything I can place
in the response header to prevent this from happening? My server
already traps "If-Modified-Since" requests and always replies with the
proper "not-modified" response. However, sometimes things go nuts and
a single user can cause my server to grind to a halt. What do I do?
I'd hate to deny Netscape users access to the play button, but its
getting to that point!
Sincerely,
Chris Sobotkiewicz
