This is a static archive of our old Q&A Site. Please post any new questions and answers at ask.wireshark.org.

Is it possible to do regression test of dissected packets

0

if we have developed our own protocol and if we want to perform regression test for that for dissection is it possible? if it is, then how?

asked 26 Mar '15, 04:41

ankit's gravatar image

ankit
65232328
accept rate: 25%


2 Answers:

0

If you want to test your own dissector you can use the ./tools/fuzz-test.sh utility, see the page on fuzztesting. It depends on having capture files of your protocol in action and makes subtle changes (bit errors) in the packets. This capture file is then handed to the dissectors.

answered 26 Mar '15, 05:15

Jaap's gravatar image

Jaap ♦
11.7k16101
accept rate: 14%

0

You could try the approach that's used in the Wireshark test suite, run tshark against a known capture file and compare the output against the expected values.

answered 26 Mar '15, 05:17

grahamb's gravatar image

grahamb ♦
19.8k330206
accept rate: 22%