Any information available from the crash might help isolate the problem.
But if this is repeatable it represents a major security issue for TFIM,
since it shows an easy denial-of-service attack.
- Dennis
Brandon Richins wrote:
When I send a RST to IBM TFIM, my STS, I crash it to the point where it has to
be rebooted. If I send the same request (captured) via SoapUI or TCPMon, it
doesn't crash it.
I was wondering what kind of connection parameters are added to http(s)
connections, specifically when communicating with an STS. I saw Dan's recent
post indicating that CXF uses Java's HttpConnection and not Apache commons-http
component. I even created a simple java client that matched the headers from
CXF 100% but I still can't replicate taking it down. I have looked at the
HttpConduit and didn't see anything out of the ordinary that could cause this.
Any ideas as I try to develop a test case or debug this?
I tried to look at increasing the log level but
http://cxf.apache.org/docs/debugging-and-logging.html seems to be missing the
example content and the linked config
(http://cwiki.apache.org/CXF20DOC/configuration.html) returns a 404. Any tips
here would be great too.
Thanks in advance,
Brandon
