I have a KVM setup. Host is 6.3. I have a bridged client running 5.9
Since upgrading to 5.9 I noticed that ntpd is being restarted every 12 hours.
Jan 20 08:00:25 mercury ntpd[16103]: ntpd exiting on signal 15 Jan 20 20:00:26 mercury ntpd[27462]: ntpd exiting on signal 15 Jan 21 08:00:27 mercury ntpd[11343]: ntpd exiting on signal 15 Jan 21 20:00:28 mercury ntpd[25148]: ntpd exiting on signal 15 Jan 22 08:00:29 mercury ntpd[6254]: ntpd exiting on signal 15 Jan 22 20:00:30 mercury ntpd[20424]: ntpd exiting on signal 15 Jan 23 08:00:31 mercury ntpd[2101]: ntpd exiting on signal 15
Looking at logs this appears to be corrolated to boot.log entries
Jan 22 08:00:29 mercury NET[20506]: /sbin/dhclient-script : updated /etc/resolv.conf Jan 22 20:00:30 mercury NET[2080]: /sbin/dhclient-script : updated /etc/resolv.conf Jan 22 20:00:30 mercury NET[2188]: /sbin/dhclient-script : updated /etc/resolv.conf Jan 23 08:00:31 mercury NET[16383]: /sbin/dhclient-script : updated /etc/resolv.conf Jan 23 08:00:31 mercury NET[16485]: /sbin/dhclient-script : updated /etc/resolv.conf
Hmm, looking at messages
Jan 23 07:58:26 mercury dhclient: 5 bad udp checksums in 5 packets Jan 23 07:58:38 mercury dhclient: DHCPREQUEST on eth0 to 255.255.255.255 port 67 (xid=0x3265b988) Jan 23 07:59:13 mercury dhclient: DHCPREQUEST on eth0 to 255.255.255.255 port 67 (xid=0x3265b988) Jan 23 07:59:22 mercury dhclient: DHCPREQUEST on eth0 to 255.255.255.255 port 67 (xid=0x3265b988) Jan 23 07:59:22 mercury dhclient: 5 bad udp checksums in 5 packets Jan 23 07:59:30 mercury dhclient: DHCPREQUEST on eth0 to 255.255.255.255 port 67 (xid=0x3265b988) Jan 23 08:00:13 mercury dhclient: DHCPREQUEST on eth0 to 255.255.255.255 port 67 (xid=0x3265b988) Jan 23 08:00:13 mercury dhclient: 5 bad udp checksums in 5 packets Jan 23 08:00:31 mercury NET[16383]: /sbin/dhclient-script : updated /etc/resolv.conf Jan 23 08:00:31 mercury dhclient: DHCPDISCOVER on eth0 to 255.255.255.255 port 67 interval 5 (xid=0x41b67d6c) Jan 23 08:00:31 mercury dhclient: DHCPOFFER from 10.0.0.134 Jan 23 08:00:31 mercury dhclient: DHCPREQUEST on eth0 to 255.255.255.255 port 67 (xid=0x41b67d6c) Jan 23 08:00:31 mercury dhclient: DHCPACK from 10.0.0.134 (xid=0x41b67d6c) Jan 23 08:00:31 mercury NET[16485]: /sbin/dhclient-script : updated /etc/resolv.conf Jan 23 08:00:31 mercury dhclient: bound to 10.0.0.135 -- renewal in 17809 seconds.
And the dhcp server (on the host) Jan 23 07:59:39 penfold dhcpd: DHCPACK on 10.0.0.135 to 00:1d:09:1c:c8:7e via br0 Jan 23 07:59:51 penfold dhcpd: DHCPREQUEST for 10.0.0.135 from 00:1d:09:1c:c8:7e via br0 Jan 23 07:59:51 penfold dhcpd: DHCPACK on 10.0.0.135 to 00:1d:09:1c:c8:7e via br0 Jan 23 08:00:04 penfold dhcpd: DHCPREQUEST for 10.0.0.135 from 00:1d:09:1c:c8:7e via br0 Jan 23 08:00:04 penfold dhcpd: DHCPACK on 10.0.0.135 to 00:1d:09:1c:c8:7e via br0 Jan 23 08:00:13 penfold dhcpd: DHCPREQUEST for 10.0.0.135 from 00:1d:09:1c:c8:7e via br0 Jan 23 08:00:13 penfold dhcpd: DHCPACK on 10.0.0.135 to 00:1d:09:1c:c8:7e via br0 Jan 23 08:00:31 penfold dhcpd: DHCPOFFER on 10.0.0.135 to 00:1d:09:1c:c8:7e via br0 Jan 23 08:00:31 penfold dhcpd: DHCPREQUEST for 10.0.0.135 (10.0.0.134) from 00:1d:09:1c:c8:7e via br0 Jan 23 08:00:31 penfold dhcpd: DHCPACK on 10.0.0.135 to 00:1d:09:1c:c8:7e via br0
So it's definitely looking like dhclient doesn't like dhcpd.
Server is dhcp-4.1.1-31.0.1.P1.el6.centos.1.x86_64 qemu-kvm-0.12.1.2-2.295.el6_3.10.x86_64
Client is dhclient-3.0.5-31.el5_8.1
Other than being noisy, it doesn't look like things are breaking, but it's definitely wrong :-)