[ntp:bugs] [Bug 431] Wrong auth state reported by "ntpq -c as"

[bugzilla at ntp.isc.org]

http://bugs.ntp.isc.org/show_bug.cgi?id=431





------- Additional Comments From mills at udel.edu  2005-05-23 17:58 -------
Subject: Wrong auth state reported by "ntpq -c as"

Martin,

The logic is indeed incorrect; however, the fix is not just inverting 
the bit. The semantics of the bit is to be lit if the packet has a MAC 
and the MAC is correct. Otherwise, your fix will light the bit when the 
packet has no MAC, which is not the intent.

I have fixed the logic in ntp_proto.c; however, the rebuild works on my 
FreeBSD machines, but breaks on my Solaris machines. On Solaris after 
recompiling and relinking, the build fails apparently with a format 
error in line 1 of ntp_control.c. What is it doing with ntp_control.c 
-after- linking the library? I have long since given up hope of 
understanding the build process.

Dave,

bugzilla at ntp.isc.org wrote:

>http://bugs.ntp.isc.org/show_bug.cgi?id=431
>
>
>martin.burnicki at meinberg.de changed:
>
>           What    |Removed                     |Added
>----------------------------------------------------------------------------
>             Status|RESOLVED                    |REOPENED
>         Resolution|FIXED                       |
>
>
>
>
>------- Additional Comments From martin.burnicki at meinberg.de  2005-05-23 11:21 -------
>Dave,  
>  
>Unfortunately your latest patch didn't fix the problem. However, it just looks 
>like the polarity of FLAG_AUTHENTIC just has to be reversed. 
> 
>In your patch, FLAG_AUTHENTIC is set when TEST5 is set, but TEST5 indicates an 
>auth error while FLAG_AUTHENTIC is supposed to indicate that authentication is 
>OK, so FLAG_AUTHENTIC should be cleared if TEST5 is set. Also FLAG_AUTHENTIC 
>should be set instead to be cleared if there's no auth error. 
> 
>After I've reverted the handling of the FLAG_AUTHENTIC flag it seems to work 
>correctly. Older and recent versions reflect the auth status correctly. 
> 
>I'll append a BK patch here and send you the diff output by private mail. 
> 
>Best regards, 
> 
>Martin 
> 
>
>
>
>------- You are receiving this mail because: -------
>You are on the CC list for the bug, or are watching someone who is.
>





------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.