Bug 1196635

Summary: ntpd -x steps clock on leap second
Product: [Fedora] Fedora Reporter: Miroslav Lichvar <mlichvar>
Component: ntpAssignee: Miroslav Lichvar <mlichvar>
Status: CLOSED RAWHIDE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: rawhideCC: aschorr, mlichvar
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: ntp-4.2.6p5-28.fc23 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: 1190619 Environment:
Last Closed: 2015-02-26 14:57:26 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description Miroslav Lichvar 2015-02-26 12:29:49 UTC
Description of problem:
With older ntp versions, the -x option was sometimes used as a workaround to avoid kernel inserting/deleting leap seconds by stepping the clock and possibly upsetting running applications. That no longer works with 4.2.6 as ntpd steps the clock itself when a leap second occurs.

The fix is to treat the one second offset gained during leap second as a normal offset and check the stepping threshold (set by -x or tinker step) to decide if a step should be applied.

Version-Release number of selected component (if applicable):
ntp-4.2.6p5

How reproducible:
Always

Steps to Reproduce:
1. add -x to /etc/sysconfig/ntpd
2. wait for leap second
3. observe how was clock corrected

Actual results:
step

Expected results:
slew

Comment 1 Fedora Update System 2015-02-26 15:12:39 UTC
ntp-4.2.6p5-28.fc22 has been submitted as an update for Fedora 22.
https://admin.fedoraproject.org/updates/ntp-4.2.6p5-28.fc22

Comment 2 Fedora Update System 2015-03-13 17:03:19 UTC
ntp-4.2.6p5-28.fc22 has been pushed to the Fedora 22 stable repository.  If problems still persist, please make note of it in this bug report.

Comment 3 Andrew J. Schorr 2015-07-01 15:15:48 UTC
Shouldn't this behavior be documented in the man page?