Bug 788935

Summary: yet another "chrony does not start via systemd"
Product: [Fedora] Fedora Reporter: udo.rader
Component: chronyAssignee: Miroslav Lichvar <mlichvar>
Status: CLOSED DUPLICATE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 16CC: mlichvar
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: 2012-02-09 10:53:17 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:

Description udo.rader 2012-02-09 10:31:12 UTC 
-------CUT--------
[root@artio ~]# systemctl status chronyd.service
chronyd.service - NTP client/server
	  Loaded: loaded (/lib/systemd/system/chronyd.service; enabled)
	  Active: failed since Thu, 09 Feb 2012 11:22:12 +0100; 15s ago
	 Process: 3539 ExecStartPost=/usr/libexec/chrony-helper add-dhclient-servers (code=exited, status=1/FAILURE)
	 Process: 3536 ExecStart=/usr/sbin/chronyd -u chrony $OPTIONS (code=exited, status=0/SUCCESS)
	 Process: 3530 ExecStartPre=/usr/libexec/chrony-helper generate-commandkey (code=exited, status=0/SUCCESS)
	Main PID: 3538 (code=exited, status=0/SUCCESS)
	  CGroup: name=systemd:/system/chronyd.service
-------CUT--------

starting the server manually succeeds:

-------CUT--------
[root@artio ~]# chronyd -u chrony
[root@artio ~]# echo $?
0
-------CUT--------

I neither get a SELinux exception as in bug 787477 nor do I have the same servers in /etc/chrony.conf as in /var/lib/dhclient/chrony.servers.*, as described in bug 787042.

My chrony.conf is as it was shipped and our DHCP server sends NTP information pointing to our internal time server:

-------CUT--------
[root@artio ~]# ls -l /var/lib/dhclient/chrony.servers.*
-rw-r--r--. 1 root root 19 Feb  9  2012 /var/lib/dhclient/chrony.servers.eth0
-rw-r--r--. 1 root root 19 Feb  9  2012 /var/lib/dhclient/chrony.servers.wlan0
[root@artio ~]# cat /var/lib/dhclient/chrony.servers.*
172.20.10.4 iburst
172.20.10.4 iburst
-------CUT--------

I've also tried with only one interface active (either wlan0 or eth0), but that did not make a difference.
Comment 1 udo.rader 2012-02-09 10:38:36 UTC 
Ok, I found the problem, against my previous assumption it is indeed related to bug 787042. 

The problem is that my two interfaces eth0 and wlan0 point to the same NTP server. Deleting one file, eg /var/lib/dhclient/chrony.servers.wlan0 allows me to start chronyd via systemctl:

------CUT------
root@artio ~]# rm /var/lib/dhclient/chrony.servers.wlan0
[root@artio ~]# systemctl start chronyd.servic
------CUT------

So the workaround from bug 787042 will probably also work for me, OTOH this is quite a major issue because it it probably "not too uncommon" having more than one interface, at least with laptops.
Comment 2 Miroslav Lichvar 2012-02-09 10:53:17 UTC 
Yes, it's the same problem as in bug 787042. Adding a server which was already added (either from chrony.conf or DHCP) fails, the script exits with a non-zero exit code and the service is halted.

*** This bug has been marked as a duplicate of bug 787042 ***