+20040308
+ - (dtucker) [sshd.c] Back out rev 1.270 as it caused problems on some
+ platforms (eg SCO, HP-UX) with logging in the wrong TZ. ok djm@
+ - (dtucker) [configure.ac sshd.c openbsd-compat/bsd-misc.h
+ openbsd-compat/setenv.c] Unset KRB5CCNAME on AIX to prevent it from being
+ inherited by the child. ok djm@
+ - (dtucker) [auth-pam.c auth-pam.h auth1.c auth2.c monitor.c monitor_wrap.c
+ monitor_wrap.h] Bug #808: Ensure force_pwchange is correctly initialized
+ even if keyboard-interactive is not used by the client. Prevents
+ segfaults in some cases where the user's password is expired (note this
+ is not considered a security exposure). ok djm@
+ - (djm) OpenBSD CVS Sync
+ - markus@cvs.openbsd.org 2004/03/03 06:47:52
+ [sshd.c]
+ change proctiltle after accept(2); ok henning, deraadt, djm
+ - djm@cvs.openbsd.org 2004/03/03 09:30:42
+ [sftp-client.c]
+ Don't print duplicate messages when progressmeter is off
+ Spotted by job317 AT mailvault.com; ok markus@
+ - djm@cvs.openbsd.org 2004/03/03 09:31:20
+ [sftp.c]
+ Fix initialisation of progress meter; ok markus@
+
+20040307
+ - (tim) [regress/login-timeout.sh] fix building outside of source tree.
+
+20040304
+ - (dtucker) [auth-pam.c] Don't try to export PAM when compiled with
+ -DUSE_POSIX_THREADS. From antoine.verheijen at ualbert ca. ok djm@
+ - (dtucker) [auth-pam.c] Reset signal status when starting pam auth thread,
+ prevent hanging during PAM keyboard-interactive authentications. ok djm@
+ - (dtucker) [auth-passwd.c auth-sia.c auth-sia.h defines.h
+ openbsd-compat/xcrypt.c] Bug #802: Fix build error on Tru64 when
+ configured --with-osfsia. ok djm@
+
+20040303
+ - (djm) [configure.ac ssh-agent.c] Use prctl to prevent ptrace on ssh-agent
+ ok dtucker
+
+20040229
+ - (tim) [configure.ac] Put back bits mistakenly removed from Rev 1.188
+
20040229
- (dtucker) OpenBSD CVS Sync
- djm@cvs.openbsd.org 2004/02/25 00:22:45
[sshd.c]
typo in comment
+ - dtucker@cvs.openbsd.org 2004/02/27 22:42:47
+ [dh.c]
+ Prevent sshd from sending DH groups with a primitive generator of zero or
+ one, even if they are listed in /etc/moduli. ok markus@
+ - dtucker@cvs.openbsd.org 2004/02/27 22:44:56
+ [dh.c]
+ Make /etc/moduli line buffer big enough for 8kbit primes, in case anyone
+ ever uses one. ok markus@
+ - dtucker@cvs.openbsd.org 2004/02/27 22:49:27
+ [dh.c]
+ Reset bit counter at the right time, fixes debug output in the case where
+ the DH group is rejected. ok markus@
+ - dtucker@cvs.openbsd.org 2004/02/17 08:23:20
+ [regress/Makefile regress/login-timeout.sh]
+ Add regression test for LoginGraceTime; ok markus@
+ - markus@cvs.openbsd.org 2004/02/24 16:56:30
+ [regress/test-exec.sh]
+ allow arguments in ${TEST_SSH_XXX}
+ - markus@cvs.openbsd.org 2004/02/24 17:06:52
+ [regress/ssh-com-client.sh regress/ssh-com-keygen.sh
+ regress/ssh-com-sftp.sh regress/ssh-com.sh]
+ test against recent ssh.com releases
+ - dtucker@cvs.openbsd.org 2004/02/28 12:16:57
+ [regress/dynamic-forward.sh]
+ Make dynamic-forward understand nc's new output. ok markus@
+ - dtucker@cvs.openbsd.org 2004/02/28 13:44:45
+ [regress/try-ciphers.sh]
+ Test acss too; ok markus@
+ - (dtucker) [regress/try-ciphers.sh] Skip acss if not compiled in (eg if we
+ built with openssl < 0.9.7)
20040226
- (bal) KNF our sshlogin.c even if the code looks nothing like upstream