Wireshark-dev: Re: [Wireshark-dev] help needed in tracking down a bug in SSL dissector
From: Brian Vandenberg <phantal@xxxxxxxxx>
Date: Wed, 30 May 2007 17:23:35 -0600
As an addendum to this, one thing I tried for tracking this down was
to set a conditional breakpoint to watch for each instance of an
execution of add_new_data_source where the name of the data source
contains the text "Reassembled SSL", then set a watch for
tvb->real_data+c, for some c that jumps to a particular part of the
decrypted packet that contains a unique sessionid to have a quick means
of looking for duplicate calls to add_new_data_source for the same
chunk, but each sessionid receives only one call.
Is there another function other than add_new_data_source that can add a tab to the bottom when dissecting something?
-Brian Brian Vandenberg wrote:
I found a bug about 9-10 months ago in the SSL dissector that was never fixed, so I'm trying to find/fix the problem myself, but I'm having a hard time tracking it down.Essentially this is what happens: * POST to server occurs, the SSL chunk isn't fragmented.* When SSL dissector decrypts it a single new data source is added and displayed entitled "Decrypted SSL Data"* The new data source is then passed to the http dissector. * ... * A response from the server is sent, the SSL chunk *IS* fragmented.* The SSL dissector decrypts each chunk then reconstructs it; four new data sources are added:** Decrypted SSL data (X bytes) ** Reassembled SSL (Y bytes) ** Decrypted SSL data (Z bytes) ** Reassembled SSL (Y bytes)** note that both 'Reassembled' data sources have the same size, and the exact same data.* The reassembled chunks are *both* passed to the http dissector.The problem is that the SSL dissector is adding the same reassembled chunk as a new data source twice (possibly more, if it is broken up into more than 2 chunks, but so far I've only seen it broken in two chunks). I've spent a couple of days on this, and though I think I'm close, I just don't know enough about the Wireshark API, so I'm making very slow progress.I expected that it would be as simple as going to line 920 of packet-ssl.c, set a breakpoint, observe that the line is executed twice for each packet where this occurs, then figure out why it executes more than once ... but it only executes that line once for each packet where this occurs, and I don't see another line of code that tries to add a new data source with the text "Reassembled SSL", . I haven't managed to find where the 2nd call to add_new_data_source takes place.I'm going to continue working on this, however, if anyone has a suggestion, or see's something I'm missing, I'd appreciate the help.-Brian Anders Broman wrote:Hi, ETR 091 (ETSI ETR 091 ed.1 (1993-07)) downloadable from ETSI. http://www.etsi.org/services_products/freestandard/home.htm Regards Anders ________________________________________ Från: wireshark-dev-bounces@xxxxxxxxxxxxx [mailto:wireshark-dev-bounces@xxxxxxxxxxxxx] För Kukosa, Tomas Skickat: den 30 maj 2007 23:19 Till: wireshark-dev@xxxxxxxxxxxxx Ämne: Re: [Wireshark-dev] [Wireshark-commits] rev 22008:/trunk/asn1/gsmmap/ /trunk/asn1/gsmmap/:MAP-ApplicationContexts.asn MAP-BS-Code.asnMAP-CH-DataTypes.asn MAP-CommonDataTypes.asnMAP-ER-DataTypes.asn MAP-GR-DataTypes.asn ... Hi,which document the MobileDomainDefinitions.asn comes from? I have tries do find any reference in the 3GPP TS 29.002 but without any success. Regards,Tomas ________________________________________ Od: wireshark-commits-bounces@xxxxxxxxxxxxx za uživatele etxrab@xxxxxxxxxxxxx Odesláno: st 30.5.2007 21:03 Komu: wireshark-commits@xxxxxxxxxxxxx Předmět: [Wireshark-commits] rev 22008: /trunk/asn1/gsmmap/ /trunk/asn1/gsmmap/: MAP-ApplicationContexts.asn MAP-BS-Code.asn MAP-CH-DataTypes.asn MAP-CommonDataTypes.asn MAP-ER-DataTypes.asn MAP-GR-DataTypes.asn ... http://anonsvn.wireshark.org/viewvc/viewvc.cgi?view=rev&revision=22008 User: etxrab Date: 2007/05/30 07:03 PM Log: Add separate asn1 files that may be used later. Directory: /trunk/asn1/gsmmap/ Changes Path Action +196 -0 MAP-ApplicationContexts.asn Added +130 -0 MAP-BS-Code.asn Added +463 -0 MAP-CH-DataTypes.asn Added +627 -0 MAP-CommonDataTypes.asn Added +406 -0 MAP-ER-DataTypes.asn Added +197 -0 MAP-GR-DataTypes.asn Added +2596 -0 MAP-MS-DataTypes.asn Added +214 -0 MAP-OM-DataTypes.asn Added +246 -0 MAP-SM-DataTypes.asn Added +186 -0 MAP-SS-Code.asn Added +341 -0 MAP-SS-DataTypes.asn Added (3 files not shown) _______________________________________________ Wireshark-commits mailing list Wireshark-commits@xxxxxxxxxxxxx http://www.wireshark.org/mailman/listinfo/wireshark-commits _______________________________________________ Wireshark-dev mailing list Wireshark-dev@xxxxxxxxxxxxx http://www.wireshark.org/mailman/listinfo/wireshark-dev_______________________________________________ Wireshark-dev mailing list Wireshark-dev@xxxxxxxxxxxxx http://www.wireshark.org/mailman/listinfo/wireshark-dev
- References:
- Re: [Wireshark-dev] [Wireshark-commits] rev 22008:/trunk/asn1/gsmmap/ /trunk/asn1/gsmmap/:MAP-ApplicationContexts.asn MAP-BS-Code.asnMAP-CH-DataTypes.asn MAP-CommonDataTypes.asnMAP-ER-DataTypes.asn MAP-GR-DataTypes.asn ...
- From: Anders Broman
- [Wireshark-dev] help needed in tracking down a bug in SSL dissector
- From: Brian Vandenberg
- Re: [Wireshark-dev] [Wireshark-commits] rev 22008:/trunk/asn1/gsmmap/ /trunk/asn1/gsmmap/:MAP-ApplicationContexts.asn MAP-BS-Code.asnMAP-CH-DataTypes.asn MAP-CommonDataTypes.asnMAP-ER-DataTypes.asn MAP-GR-DataTypes.asn ...
- Prev by Date: [Wireshark-dev] help needed in tracking down a bug in SSL dissector
- Next by Date: Re: [Wireshark-dev] File:Import?
- Previous by thread: [Wireshark-dev] help needed in tracking down a bug in SSL dissector
- Next by thread: [Wireshark-dev] rev 22007: /trunk/epan/dissectors/ /trunk/epan/dissectors/: packet-gtp.c broke the build
- Index(es):