Hello Juergen, > I saw the draft-almesberger-wajhak-diffserv-linux-00.txt file. > But in this file say that AF21 is 0x18 and two paragraphs before that they state that AF21 is 0x12 - so maybe its a typing error. Or the AF-DSCP values were not defined clearly at that time - both, rfc2597 and the draft were written in june99. Or its pure stubbornness - imagine Werner realy _wanted_ his AF22 to be 0x1a. The values in rfc2597 are RECOMMENDED values after all. ;) > If the rfc2597 is right(list below) > The RECOMMENDED values of the AF codepoints are as follows: AF11 = ' > 001010', AF12 = '001100', AF13 = '001110', AF21 = '010010', AF22 = ' > 010100', AF23 = '010110', AF31 = '011010', AF32 = '011100', AF33 = ' > 011110', AF41 = '100010', AF42 = '100100', and AF43 = '100110'. > > AF21 and AF22 is 010010 and 010100 > 010010 = 0x12 > 010100 = 0x14 > is that right?????? Yes. > Why many papers and documents arent' use this table????? Stubbornness, ignorance, inertia, eccentricsm, typos... hmm, probably mostly ignorance - that's my guess. Kurt
