Description of problem: systemctl start chronyd.service fails with: [root@daviesmachine ~]# systemctl start chronyd.service Job failed. See system logs and 'systemctl status' for details. [root@daviesmachine ~]# systemctl status chronyd.service chronyd.service - NTP client/server Loaded: loaded (/lib/systemd/system/chronyd.service; disabled) Active: failed since Sun, 05 Feb 2012 15:03:44 +0000; 5s ago Process: 1676 ExecStartPost=/usr/libexec/chrony-helper add-dhclient-servers (code=exited, status=1/FAILURE) Process: 1673 ExecStart=/usr/sbin/chronyd -u chrony $OPTIONS (code=exited, status=0/SUCCESS) Process: 1667 ExecStartPre=/usr/libexec/chrony-helper generate-commandkey (code=exited, status=0/SUCCESS) Main PID: 1675 (code=exited, status=0/SUCCESS) CGroup: name=systemd:/system/chronyd.service Version-Release number of selected component (if applicable): systemd-37-11.fc16.i686 How reproducible: Every time Steps to Reproduce: 1. systemctl start chronyd.service 2. chronyd does not start 3. Actual results: Job failed. See system logs and 'systemctl status' for details. Expected results: Expect chronyd.service to start normally Additional info: F16 fully up to date - Will attach chrony.conf shortly Interestingly running chronyd manually from a terminal window does start the daemon but there is an selinux avc. SELinux is preventing /usr/sbin/chronyd from getsession access on the None If you believe that chronyd should be allowed getsession access on the <Unknown> by default. Then you should report this as a bug. You can generate a local policy module to allow this access. Do allow this access for now by executing: # grep chronyd /var/log/audit/audit.log | audit2allow -M mypol # semodule -i mypol.pp Additional Information: Source Context system_u:system_r:chronyd_t:s0 Target Context unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1 023 Target Objects [ None ] Source chronyd Source Path /usr/sbin/chronyd Port <Unknown> Host daviesmachine Source RPM Packages chrony-1.26-3.20110831gitb088b7.fc16.i686 Target RPM Packages Policy RPM selinux-policy-3.10.0-75.fc16.noarch Selinux Enabled True Policy Type targeted Enforcing Mode Enforcing However the daemon does run when executed from the terminal. I am not sure if selinux, chrony or systemd are the main issue here?
Created attachment 559470 [details] chrony.conf as used when chrony.service fails to start
Could you attach the output of ausearch -m avc -ts recent After the chronyd fails to start. Does it fail to start in permissive mode?
Dan I am really sorry but this machine is one that I maintain for a relative who lives 200 miles from me and has just left to go back home after a weekend visit during which I updated the machine - and they only connect to the net via a tethered phone so I can't ssh in to run that test until they next visit me. They are not admin expert enough to run the test on my behalf - However I did set the machine to permissive two days back when I first came across the issue, as a test, and it was exactly the same (i.e. trying to start the daemon with systemctl) - and for good measure I stopped iptables as well and it was also exactly the same - which was why I was not convinced that the main problem was selinux related - unfortunately I did not run the chronyd command manually with selinux set permissive - and that may well have been selinux related but was a secondary issue since I wanted the machine to go back with them with a functional chronyd started with systemd at boot - The machine is i386 and had a standard f16 install which until now had no issues related to chrony as far as I know. If progress on this needs the output of the command you give in #2 then unfortunately it will need to wait until their next visit around the end of March. It would be useful if anyone else had a similar or the same problem who could progress this?
I just installed and started it on F17/Rawhide and it seems to be working.
Seeing the same on rawhide (chrony-1.26-4.20110831gitb088b7.fc17.x86_64) with selinux disabled on boot. Active: failed since Mon, 06 Feb 2012 21:29:09 +0100; 27min ago Process: 19861 ExecStartPost=/usr/libexec/chrony-helper add-dhclient-servers (code=exited, status=1/FAILURE)
Did it log any errors to /var/log/messages?
Is there a server in /var/lib/dhclient/chrony.servers.* which is also in /etc/chrony.conf? That would be a dup of bug #787042.
It is a dup of #787042 for me. Both chrony.servers.{eth0,eth1} had the same server ip...
Here is the package fixing bug #787042. http://koji.fedoraproject.org/koji/buildinfo?buildID=298706
I'm assuming this is a dup of bug #787042 for the original reporter too. Please reopen if that's not the case. *** This bug has been marked as a duplicate of bug 787042 ***