https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=3916
Kovarththanan Rajaratnam <kovarththanan.rajaratnam@xxxxxxxxx> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |kovarththanan.rajaratnam@gma
| |il.com
--- Comment #1 from Kovarththanan Rajaratnam <kovarththanan.rajaratnam@xxxxxxxxx> 2009-08-19 06:59:46 PDT ---
(In reply to comment #0)
> Build Information:
> svn HEAD
> --
> Hi,
>
> For two reasons:
> 1) tap_queue_init() also set tapping_is_active to TRUE if there's tap
> listeners and there's always tap listeners... cf gsm code (surely a
> bug).
You're right. Fixed in r29475.
> 2) there's no guarantee it's a constant, some dissectors (smb and smb2 do)
> can register a tap listener when decoding a packet and they may have a
> filter.
>
I think this is a dissector bug. The smb dissector registers a tap listener
when it detects it needs it. However, this only take effect on the _next_
packet. I think the correct solution is to move the registration to the
proto_reg_handoff routine (this is similar to what MATE does).
--
Configure bugmail: https://bugs.wireshark.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.