[Bug 68901] Coyote is hardcoded to drop connections on 400|408|411|414|500|503|501 which should be configurable for application level errors to prevent expensive TLS handshake/resumption on reconnect

2024-04-30 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=68901

--- Comment #3 from Christopher Schultz  ---
I think it's reasonable for Tomcat to drop the connection if it *knows* that
the connection is broken e.g. Tomcat has detected malformed headers, trailers,
etc.

But if Tomcat hands-off control to the application and the application
generates a 4xx response, etc. then I think it makes sense to allow the
connection to remain open for subsequent pipelined requests.

A case I'm not so sure about is when the application will be consuming the
request and so Tomcat doesn't "know" until later in the request-processing that
the request body is malformed. Think "chunked encoding gone wrong". Tomcat
needs a way to ensure that it can tell the difference between this kind of
situation and one where the application itself has decided to return a 400
response.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org



[Bug 68901] Coyote is hardcoded to drop connections on 400|408|411|414|500|503|501 which should be configurable for application level errors to prevent expensive TLS handshake/resumption on reconnect

2024-04-30 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=68901

Mark Thomas  changed:

   What|Removed |Added

 CC||adwsi...@gmail.com

--- Comment #2 from Mark Thomas  ---
*** Bug 68958 has been marked as a duplicate of this bug. ***

-- 
You are receiving this mail because:
You are the assignee for the bug.
-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org



[Bug 68901] Coyote is hardcoded to drop connections on 400|408|411|414|500|503|501 which should be configurable for application level errors to prevent expensive TLS handshake/resumption on reconnect

2024-04-16 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=68901

Michael Osipov  changed:

   What|Removed |Added

 CC||micha...@apache.org

-- 
You are receiving this mail because:
You are the assignee for the bug.
-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org



[Bug 68901] Coyote is hardcoded to drop connections on 400|408|411|414|500|503|501 which should be configurable for application level errors to prevent expensive TLS handshake/resumption on reconnect

2024-04-16 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=68901

--- Comment #1 from Michael Osipov  ---
I'd like to see this backed by the current RFC if it is a requirement or just
the way Tomcat handles it. Also maybe it might be worth to raise the same issue
with HTTPd as well.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org



[Bug 68901] Coyote is hardcoded to drop connections on 400|408|411|414|500|503|501 which should be configurable for application level errors to prevent expensive TLS handshake/resumption on reconnect

2024-04-16 Thread bugzilla
https://bz.apache.org/bugzilla/show_bug.cgi?id=68901

Tiago do Couto  changed:

   What|Removed |Added

 CC||ti...@docouto.dev

-- 
You are receiving this mail because:
You are the assignee for the bug.
-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org