Bug 875865

Summary: NTP service stopped after stopping network service
Product: [Fedora] Fedora Reporter: Igor Lvovsky <ilvovsky>
Component: ntpAssignee: Miroslav Lichvar <mlichvar>
Status: CLOSED NOTABUG QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: high Docs Contact:
Priority: unspecified    
Version: 18CC: dkenigsb, iheim, johannbg, lnykryn, lpeer, metherid, mlichvar, mschmidt, msekleta, notting, pertusus, plautrba, systemd-maint, vpavlin
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2012-11-20 08:06:51 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:
Attachments:
Description Flags
ntpd failure log none

Description Igor Lvovsky 2012-11-12 17:23:43 UTC 
Description of problem:
 NTP service crashed after stopping network service

Version-Release number of selected component (if applicable):
ntp-4.2.6p5-5.fc18.x86_64

How reproducible:
100%

Steps to Reproduce:
1. run ntp service
2. stop network service 
3. ntp service will fail
  
Actual results:
ntp service will fail

Expected results:
ntp service shouldn't fail

Additional info:
Comment 1 Igor Lvovsky 2012-11-12 17:29:29 UTC 
Created attachment 643642 [details]
ntpd failure log

--- snip from the log ----

Nov 11 16:35:20 dhcp-1-165 ntpd[14399]: 0.0.0.0 c615 05 clock_sync
Nov 11 16:37:23 dhcp-1-165 systemd[1]: Stopping LSB: Bring up/down networking...
Nov 11 16:37:23 dhcp-1-165 systemd[1]: Stopping Network Time Service...
Nov 11 16:37:23 dhcp-1-165 ntpd[14399]: ntpd exiting on signal 15
Nov 11 16:37:23 dhcp-1-165 systemd[1]: Starting Network Time Service...
Nov 11 16:37:23 dhcp-1-165 ntpd[15124]: ntpd 4.2.6p5 Wed Oct 24 16:46:54 UTC 2012 (1)
Nov 11 16:37:23 dhcp-1-165 ntpd[15125]: proto: precision = 0.069 usec

---- snip from the log ----

In this particular log you can see that NTP raise up again because it explicitly required by vdsmd service.
Comment 2 Miroslav Lichvar 2012-11-13 11:46:07 UTC 
It seems the ntpd service didn't crash, but was restarted for some reason. It happens without vdsmd service too.

Looking at the systemctl dot output I don't see anything suspicious, let's see if systemd maintainers know what is going on.
Comment 3 Michal Schmidt 2012-11-13 12:45:42 UTC 
Both chrony and ntp ship hooks in /etc/dhcp/dhclient.d/ and /etc/NetworkManager/dispatcher.d/. Is it possible that some of the hooks is the cause?
Comment 4 Miroslav Lichvar 2012-11-13 13:12:19 UTC 
Good point, Michal. That would explain it.

Igor, do you see the restart of the ntp service when PEERNTP=no is added to /etc/sysconfig/network? That would be expected behavior and not a bug.
Comment 5 Igor Lvovsky 2012-11-19 21:14:52 UTC 
(In reply to comment #4)
> Good point, Michal. That would explain it.
> 
> Igor, do you see the restart of the ntp service when PEERNTP=no is added to
> /etc/sysconfig/network? That would be expected behavior and not a bug.

No, with PEERNTP=no ntpd doesn't restart
Comment 6 Miroslav Lichvar 2012-11-20 08:06:51 UTC 
Ok, thanks. Let's close this as NOTABUG.