search for: kdfname

...signed-bugs at mindrot.org Reporter: arthurmesh at gmail.com This defect was found on OpenSSH 6.5; however, it appears that code in question has not changed between 6.5 and openssh-SNAP-20140204.tar.gz. Thoughts: while reading the code -- this caught attention: authfile.c: 279 kdfname = buffer_get_cstring_ret(&copy, NULL); 280 if (kdfname == NULL || 281 (!strcmp(kdfname, "none") && !strcmp(kdfname, "bcrypt"))) { 282 error("%s: unknown kdf name", __func__); 283 goto out; 284 } &...

...L_VERBOSE) { +?? ??? ?printf("Key protection details:\n"); +?? ??? ?printf("File format: %s\n", sshkey_format_name(vault_info->format)); +?? ??? ?if ( (vault_info->ciphername == NULL || strcmp(vault_info->ciphername, "none") == 0) +?? ??? ?? || (vault_info->kdfname == NULL || strcmp(vault_info->kdfname, "none") == 0)) { +?? ??? ??? ?printf("no passphrase\n"); +?? ??? ?} else { +?? ??? ??? ?printf("cipher: %s\n", vault_info->ciphername); +?? ??? ??? ?printf("kdf: %s\n", vault_info->kdfname); +?? ??? ??? ?printf(...

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, I got a proposal for a slight different default private key encryption protocol. Here is my understanding what ssh-keygen currently does. According to this article: http://martin.kleppmann.com/2013/05/24/improving-security-of-ssh-private-keys.html when you create a new key with a passphrase to protect it, ssh-keygen uses a hard-coded openssl