Comment # 9
on bug 13238
from Eric Kotz
> I think the better solution would be to try to address the malformed portion
> of the packet/data, however I don't have access to the spec (it's not free)
> to confirm my findings.
It's important to fix that, but that's not actually a solution to this issue.
I believe this and Bug #13237 have become intertwined at this point, even
though there are actually two separate issues here:
1. Wireshark is not parsing descriptor tag 0xA2 properly
2. Wireshark should continue parsing other MPEG packets within the IP packet
even if it hits a corrupt MPEG packet.
Additionally, Jeremy brought up the suggestion (and I agree, but this is more
of a feature request) that even if Wireshark didn't understand a descriptor, it
should use the length field to continue after skipping it.
Without (2) Wireshark will log continuity errors where they do not exist simply
because it aborts parsing. While fixing (1) would fix this specific problem,
any other MPEG parser bug would still trigger (2), which is why I believe it's
a separate bug that should also be addressed.
You are receiving this mail because:
- You are watching all bug changes.