This is somewhat difficult to describe. If there is a delay when processing
a GET submission, Ubiquity appears to issue many more identical requests
more or less immediately. The scenario is this. I'm on page 1 of the
attached IWD application and press "page 2" in order to submit the current
instance data and transition to the next page. This operation is composed
of a series of submissions one of which is a GET (replace-all).
On the server to which the GET request is submitted, the web application
that will receive the request is running under debug and a breakpoint is
set in code at the beginning of the method that handles the GET.
When the GET is issued, the breakpoint is hit. Then almost immediately
(easily within a second) many many more request threads are created with
the same request (i.e., they're all broken at the same breakpoint.) It
appears that the request threads are created until the server configuration
limit is reached.
It's not clear if this is limited to GETs although it appears to be - I
don't, for example, observe it for PUT requests.
Please contact me if you need any more information.
Original issue reported on code.google.com by michael.junkin@gmail.com on 24 Apr 2009 at 6:23
Original issue reported on code.google.com by
michael.junkin@gmail.com
on 24 Apr 2009 at 6:23Attachments: