If an HTTP/2 client connecting to Apache Tomcat 10.0.0-M1 to 10.0.0-M7, 9.0.0.M1 to 9.0.37 or 8.5.0 to 8.5.57 exceeded the agreed maximum number of concurrent streams for a connection (in violation of the HTTP/2 protocol), it was possible that a subsequent request made on that connection could contain HTTP headers - including HTTP/2 pseudo headers - from a previous request rather than the intended headers. This could lead to users seeing responses for unexpected resources.

Security constraints defined by annotations of Servlets in Apache Tomcat 9.0.0.M1 to 9.0.4, 8.5.0 to 8.5.27, 8.0.0.RC1 to 8.0.49 and 7.0.0 to 7.0.84 were only applied once a Servlet had been loaded. Because security constraints defined in this way apply to the URL pattern and any URLs below that point, it was possible - depending on the order Servlets were loaded - for some security constraints not to be applied. This could have exposed resources to users who were not authorised to access them.

As part of the fix for bug 61201, the documentation for Apache Tomcat 9.0.0.M22 to 9.0.1, 8.5.16 to 8.5.23, 8.0.45 to 8.0.47 and 7.0.79 to 7.0.82 included an updated description of the search algorithm used by the CGI Servlet to identify which script to execute. The update was not correct. As a result, some scripts may have failed to execute as expected and other scripts may have been executed unexpectedly. Note that the behaviour of the CGI servlet has remained unchanged in this regard. It is only the documentation of the behaviour that was wrong and has been corrected.