One of our engineers, Nalin, did a quick review of diffs between 3.0.0 and 3.0.1 and noticed this. I agree with Nalin, it looks like an unintentional mistake, was it?

I did a quick(ish?) skim of a diff between the two ...

Also from skimming that diff, I can randomly offer that in rlm_krb5.c, the switch statement that follows a call to krb5_verify_user_opt() falls through after the KRB5KDC_ERR_C_PRINCIPAL_UNKNOWN case, but the similarly-structured switch statement after the krb5_get_init_creds_password() call includes a "break;". The omission in the first doesn't look intentional to me.

-- John