[ntp:bugs] [Bug 538] 4.2 'ntptrace' perl script shows incorrect offsets

[Pranabendu Misra via the NTP Bugzilla]

http://bugs.ntp.org/538



----------------------------------------------------------------------------
Additional Comments From pranabendu.m at gmail.com (Pranabendu Misra)
Submitted on 2009-04-05 23:27

(I was asked to take a look at this bug as part of GSoC application evaluation
process. From my mint-fresh knowledge of ntp here's what I think)

The current ntptrace does behave differently than what is said in documentation.

The perl script uses ntpq -n -c rv <host>. The offset printed is the sys.offset
of the host.

While the documentation
(http://www.eecis.udel.edu/~mills/ntp/html/ntptrace.html) states,  "...the time
offset between that host and the local host..."

(I think) "local host" is the host ntptrace is running on.

So ntptrace perl script certainly behaves different from it's documentation.

I also talked to davehart on IRC who was of the view that the current behavior
is corrent and desirable, as it is supposed to trace the path to primary time
source, printing stats of time servers along the way, as it does now.

So either the program or the documentation is at fault.

As for the older behavior, I modified the perl script to 

I was also asked to take a look at sntp, relevent to this problem.

I did and modified the perl script, to produce the old behavior.

It's there at 
http://www.cmi.ac.in/~pranabendu/ntp/ntptrace2





-- 
Pranabendu Misra <pranabendu.m at gmail.com>



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