On Tue, Jan 01, 2002 at 10:08:02PM +1100, Hamish Moffatt wrote: > 0000 00 40 95 e1 16 40 00 40 05 69 a2 40 08 00 45 00 .@...@.@ .i.@..E. > 0010 05 dc 68 3f 40 00 e5 06 3f 06 c2 49 49 71 cb 0e ..h?@... ?..IIq.. > 0020 12 0d 00 50 08 db f3 58 ca 5e 02 b6 08 57 80 10 ...P...X .^...W.. > 0030 fe 88 44 af 00 00 01 01 08 0a 07 79 a6 a0 00 06 ..D..... ...y.... > 0040 55 32 0e 0a 30 07 1f a0 8c a5 c0 9a 43 fd a6 89 U2..0... ....C... > 0050 d0 07 65 30 04 37 50 0c 32 f0 01 1a 80 2f e8 37 ..e0.7P. 2..../.7 > [...] > Ethereal (0.9.0) says this second frame has an incorrect TCP checksum. > I can't figure out why changing the link layer headers causes the > TCP checksum to change? Or is there a bug in the calculation > somewhere? Stranger yet; I fed the exact same packet as captured back in (text2pcap with no -l option) and got the same checksum error in the decode. Hamish -- Hamish Moffatt VK3SB <hamish@xxxxxxxxxx> <hamish@xxxxxxxxxxxx>
Powered by MHonArc 2.6.10