[ntp:questions] Re: Correcting my time servers clock drift on Alpha ES40s / Tru64

Danny Mayer mayer at ntp.isc.org
Mon Oct 24 03:15:17 UTC 2005

Brad Knowles wrote:
> At 3:53 PM +0000 2005-10-23, Tom Smith wrote:
>>  Brad Knowles wrote:
>>>      Right, as Dave said -- it has been changed.  Starting with 4.0,
>>>  the source code was renamed back to "ntpd", not "xntpd".  So, anyone
>>>  shipping something called "xntpd" which claims a version of 4.0 or
>>>  greater means that they must have modified the source code or made
>>>  other changes.
>>  ntpd V4.2.0 built from source and started via a link named "xntpd":
>>  Oct 23 00:34:02 seeaxp xntpd[56499]: ntpd 4.2.0 at 1.1161-r Tue Feb  8
>> 16:04:28 EST 2005 (1)
>     Right, let's go back to my statement.  I said that they either
> modified the source code (i.e., the old NTPv3/xntpd source code to claim
> a version 4.x), or they made other changes (i.e., they took NTPv4 source
> code and changed the name to be xntpd instead of ntpd).
>     That statement is still accurate.  In this case, they appear to have
> done the grave disservice to their customer of taking an NTPv4 codebase
> and calling the resulting executable "xntpd".

Brad, I agree with Tom here. It doesn't matter what the binary image is
called or whether there is a link to it from another name. As long as it
is identifiable for what it really is it is maintainable by us. That's
not the important part. Yes it is misleading to call it by another name
and does everyone a disservice because of the confusing names, but
that's not unusual in any system.

Let's not confuse names with what they really are.


More information about the questions mailing list