Ethereal-users: Re: [Ethereal-users] "file proto.c: line 1651: should not be reached" in v0.9.9.

Note: This archive is from the project's previous web site, ethereal.com. This list is no longer active.

From: "Martin Regner" <martin.regner@xxxxxxxxx>
Date: Mon, 27 Jan 2003 22:14:33 +0100
Ketil Kintel:

<Bug report:
<After upgrading the Windows Version from 0.9.8 to 0.9.9 i constantly get the following error:
<"file proto.c: line 1651: should not be reached"
<"Error: Unspecified fatal error encountered, aborting."
<
<Version: "ethereal 0.9.9, compiled with GLib 1.3.2, with libpcap (version unknown), with libz 1.1.4, with Net-SNMP 5.0.6" and <both WinPCap v 2.3 and v3.0a4 on WinXP sp3.

Hi,

If there is some fault in some dissector that calls proto_tree_add_int  you could get that assert if there is a packet
that triggers this fault to occur when it is dissected, I think.

I assume that you get this only when Ethereal is dissecting packets (after you stop capturing if "Update list of packets in real time" is disabled, maybe when you are capturing if "Update list of packets in real time" is enabled, when reading in a capture file or similar). 

Do you still get the problem if you disable all protocols (Edit/Protocols.../Disable All) before capturing or reading capture files?

If you disable all protocols and capture a file you could then try to enable all protocols and see if you get the assert then.

It would be interesting to know more about what protocol dissector that seems to cause this problem, and I think
that maybe you need to send some sample capture that causes this problem - or at least figure out 
more details about for exactly what protocol you get this problem (e.g. by trying to find what packets that causes this problem and trying to disable/enable specific protocols depending on what protocols you think there is in your capture file). 
Be sure not to include any sensitive data in the capture if you send a capture.