From 0ae3f3366ffdd3a3121e99ebefdef1f995315818 Mon Sep 17 00:00:00 2001 From: Geoffrey King Date: Sat, 29 Aug 1998 00:10:24 +0000 Subject: [PATCH] Of course, the following sentence actually refers to the login function, not the pass function: Don't reply 230 "User logged in" if the user didn't send a PASS command; this causes the client to get a bit confused. ChangeLog: git-svn-id: svn://anonsvn.mit.edu/krb5/trunk@10900 dc483132-0cff-0310-8789-dd5450dbe970 --- src/appl/gssftp/ftpd/ChangeLog | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/src/appl/gssftp/ftpd/ChangeLog b/src/appl/gssftp/ftpd/ChangeLog index 6aa6a2323..a535839f9 100644 --- a/src/appl/gssftp/ftpd/ChangeLog +++ b/src/appl/gssftp/ftpd/ChangeLog @@ -2,11 +2,10 @@ * ftpd.c (login): New function. Essentially, the old pass function has been split into its two logical components, pass and - login. + login. Don't reply 230 "User logged in" if the user didn't + send a PASS command; this causes the client to get a bit confused. (pass): If auth_ok is true, reply with code 202 to tell the - user that a PASS command is not necessary. Also, don't reply - 230 "User logged in" if the user didn't send a PASS command; - this causes the client to get a bit confused. + user that a PASS command is not necessary. (auth_ok): New function that returns true if either gss_ok or kerb_ok is true (all the #ifdefs were beginning to clutter things, and it's a good abstraction in case other auth types are ever -- 2.26.2