TCP sessions cleaned up prematurely
If one TCP endpoint closes the connection, but there are still TCP segments on the network that have not been received yet by suricata, these segments might not be processed.
The following patch delays the cleanup until the FIN form both side has been seen:
This applies both to the master and the master-1.3.x branches.
Updated by Victor Julien almost 9 years ago
- Status changed from New to Closed
- Assignee set to Victor Julien
- Target version set to 1.4
I've applied this to the master with a small addition: send eof on pseudo packets as well. Saw some missing logs in the http.log otherwise.
This has led me to the conclusion that it's a pretty major change that needs some more testing. So holding off on the 1.3 branch for now.