Wireshark-bugs: [Wireshark-bugs] [Bug 3173] changeback messages not detected inside MTP3 over RUDP (Cisco proprietary)

From: bugzilla-daemon@xxxxxxxxxxxxx
Date: Mon, 5 Jan 2009 12:53:37 -0800 (PST)

https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=3173


Jeff Morriss <jeff.morriss.ws@xxxxxxxxx> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |INVALID




--- Comment #4 from Jeff Morriss <jeff.morriss.ws@xxxxxxxxx>  2009-01-05 12:53:36 PDT ---
(I should state that my earlier comment was a bit bogus: if you're seeing COOs
then you must have 2 links so you "should" also see CBDs.)

(In reply to comment #3)
> Obviously Solaris application running on receiving endpoint is receiving
> COO&COA (step 3) and CBD&CBA (step 5). Receiving endpoint has only one Ethernet

Do you know that from the application itself or just by the fact that the link
comes back into service normally?

If the latter then:

> interface. Common sense is giving us just two options: either snoop is somehow
> filtering CBD&CBA or Wireshark is not decoding CBD&CBA. Of course, it is also

Or Cisco is not using CBDs to do changeback on this (proprietary) interface. 
Changeback is basically needed to ensure message ordering when rerouting SLS
from 1 link back to the restored link.  An alternative to CBDs is to do a Time
Controlled Diversion.  Or maybe some other message is used: I noticed that
there are some Cisco SM messages exchanged between the endpoints prior to the
link coming back up.  They are not MTP3 messages (at least I can't make sense
of them) but Cisco could use them for the same purpose.

Anyway I don't think Wireshark is doing anything wrong here.


-- 
Configure bugmail: https://bugs.wireshark.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.