Bug 1047

Summary: Xntpd will mis-synchronize
Product: [Retired] Red Hat Linux Reporter: keybounce
Component: xntp3Assignee: Jeff Johnson <jbj>
Status: CLOSED WORKSFORME QA Contact:
Severity: medium Docs Contact:
Priority: medium    
Version: 5.1CC: aleksey
Target Milestone: ---   
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 1999-03-24 23:48:46 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description keybounce 1999-02-05 00:15:43 UTC 
Xntpd will improperly synchronize the clock.

The problem does not appear so much to be in the code, as in
one of the constants defined by the specifications -- it is
too loose.

A patch is coming.

Brief summary: Xntpd synchronzes when the "error" is less
than 1; this causes problems with all of the root clocks, as
they do not work that smoothly. Thus, they have a lot of
time steps; so do other sites that synch to them.

In addition, if you are doing a file transfer, xntpd's
concept of error will still go well below 1 during the file
transfer, and a 2-4 second error will be stepped into your
clock.

This patch lowers the treshold to .025.

I have been running this patch on xntpd-3.51 for over a year
on another machine; I have just applied it to our internal
time server today.

(yes, this really should be back-prop'd to the maintainers
of xntpd; I figured that it would be more likely to be
accepted if it had your backing as well)
Comment 1 Jeff Johnson 1999-03-24 23:48:59 UTC 
Please reopen this bug with the patch. The Right Thing To Do
is to send the patch to the xntp3 crowd -- we'll be happy
to include the patch with an update of xntp3.