Thanks, Tero. I ran the test cases, but apparently I missed that. Do you know if there's any procedure for reporting errata in RFCs?
Sheila -----Original Message----- From: Tero Kivinen [mailto:[email protected]] Sent: Wednesday, May 13, 2009 9:13 AM To: [email protected] Cc: [email protected]; [email protected] Subject: Error on RFC 4868 section 2.6.2.3 I was adding the RFC4868 test vectors to our code and noticed there is an error in the section 2.6.2.3 test vectors. The last SHA512 test vector has one extra line for the key. The key in the RFC is: Test Case AUTH512-4: Key = 0a0b0c0d0e0f10111213141516171819 0102030405060708090a0b0c0d0e0f10 1112131415161718191a1b1c1d1e1f20 2122232425262728292a2b2c2d2e2f30 3132333435363738393a3b3c3d3e3f40 (64 bytes) As you can clearly see the key is not 64 bytes as is said in the end, but is 80 bytes. On the other hand you can also see that the "0a0b0c0d0e0f10111213141516171819" line in the key does not match the other similar test vectors. If the test vector is changed so that key is 64 bytes from 01 to 40 then the end results given in the RFC are correct. I.e. the correct key for the AUTH512-4 test is: Test Case AUTH512-4: Key = 0102030405060708090a0b0c0d0e0f10 1112131415161718191a1b1c1d1e1f20 2122232425262728292a2b2c2d2e2f30 3132333435363738393a3b3c3d3e3f40 (64 bytes) -- [email protected] _______________________________________________ IPsec mailing list [email protected] https://www.ietf.org/mailman/listinfo/ipsec
