Wireshark mailing list archives
Recursion depth limit for packet reassembly
From: Thomas Wiens <th.wiens () gmx de>
Date: Sat, 9 Nov 2019 13:17:46 +0100
Hi, I'm working on a protocol dissector where I'm using packet reassembly. Now I've got a capture with 1026 fragments, which fails to reassemble with the standard Wireshark (assert fails) due to the limitation to 500 fragments in epan/packet.c: #define PINFO_LAYER_MAX_RECURSION_DEPTH 500 For testing I just increased the limit up to 2000. With the change the reassembly takes some time when I select the last fragment where it's reassembled (about 15 seconds on a rather old 32 bit windows machine, but there are also hundreds of zlib blocks which are decompressed in the dissector). Other that it takes some time and memory, I didn't notice any problems. Is there any reason why it's limited to exactly 500? Would it be possible to make this limit configurable in the Wireshark settings dialog? -- Best Regards Thomas Wiens ___________________________________________________________________________ Sent via: Wireshark-dev mailing list <wireshark-dev () wireshark org> Archives: https://www.wireshark.org/lists/wireshark-dev Unsubscribe: https://www.wireshark.org/mailman/options/wireshark-dev mailto:wireshark-dev-request () wireshark org?subject=unsubscribe
Current thread:
- Recursion depth limit for packet reassembly Thomas Wiens (Nov 09)
- Re: Recursion depth limit for packet reassembly Maynard, Chris via Wireshark-dev (Nov 13)