To configure NTP on the service console, you must:
Edit the following configuration files, as described in the following sections:
/etc/ntp.conf
/etc/ntp/step-tickers
/etc/hosts
For ESX Server 3.0 only, run the following command. This opens the appropriate ports and enables the NTP daemon to talk with the external server.
# esxcfg-firewall --enableService ntpClient
Restart ntpd.
Editing /etc/ntp.conf
Specify a pool of NTP servers to which your ESX Server system will sync (for more information about using NTP server pools, see ).
Log on to the service console as the root user.
You are going to replace the content of /etc/ntp.conf, so make a backup copy of this file:
cp /etc/ntp.conf /etc/ntp.conf.bk
Edit the contents of the default /etc/ntp.conf, which the ESX Server installation creates for you, so the file looks like this:
restrict 127.0.0.1
restrict default kod nomodify notrap
server 0.vmware.pool.ntp.org
server 1.vmware.pool.ntp.org
server 2.vmware.pool.ntp.org
driftfile /var/lib/ntp/drift
restrict - The first restrict is required because ntpd needs to resolve hostname addresses via the loopback network. If this entry does not exist, the system log ( /var/log/messages) will show something like this:
ntpd_initres: ntpd returns a permission denied error
For systems directly connected to the Internet, the second restrict line provides a basic level of protection from general UDP spoofing of NTP.
kod - Sends a KoD packet when an access violation occurs.
nomodify - Denies ntpq and ntpdc queries that attempt to modify the run-time configuration of the server. Queries that return information are permitted.
notrap - Declines to provide mode 6 control message trap service to matching hosts. The trap service is a subsystem of the ntpdq control message protocol, which is intended for use by remote event logging programs.
For better protection, you may want to add noquery, which prevents remote queries, and nopeer, which prevents a host from trying to peer with your server and to allow a rogue server to control the clock.
Refer to for a full description of these access control commands.
The 0, 1, and 2.vmware.pool.ntp.org names point to a random set of servers that change every hour.
The driftfile line indicates the name of the file where the value for the system's clock drift (frequency error) is stored. For a more complete definition of driftfile, see .
Editing /etc/ntp/step-tickers
Perform the following steps as the root user on the service console.
If you want, make a backup copy of your /etc/ntp/step-tickers file:
cp /etc/ntp/step-tickers /etc/ntp/step-tickers.bk
Edit /etc/ntp/step-tickers so that the file looks like the following example:
0.vmware.pool.ntp.org
1.vmware.pool.ntp.org
2.vmware.pool.ntp.org
Editing /etc/hosts
Perform the following step as the root user on the service console.
Edit /etc/hosts.
Add the following list of NTP pool servers to the end of your existing hosts list.
0.vmware.pool.ntp.org
1.vmware.pool.ntp.org
2.vmware.pool.ntp.org
Adding this list of NTP servers minimizes the impact of DNS lookup failures during NTP synchronization.
ESX 3.0 Only: Enabling NTP Client for Firewall
As noted earlier in this article, remember to run the following command. This opens the appropriate ports and enables the NTP daemon to talk with the external server.
# esxcfg-firewall --enableService ntpClient
Restarting and Monitoring the NTP Service
Perform these steps as root on the service console.
After you have edited the previous files, restart the NTP service.
To restart the service now, run:
service ntpd restart
To enable the NTP daemon to autostart when the server is rebooted, run:
chkconfig --level 345 ntpd on
Now you can set the local hardware clock to the NTP synchronized local system time. Run:
hwclock --systohc
As ntpd successfully polls NTP servers, the kernel automatically sets the hardware clock to the system clock time periodically.
Monitor the NTP service as desired:
To see the offset (in seconds) between the local clock and the source clock, run:
ntpdate -q time_server_name_or_ip_address
If the correction resulting from synchronizing the local clock with the time server is large enough, it could affect the operating systems or applications running in virtual machines when they synchronize their clocks with the ESX Server system on which they are running.
To watch the status of the ntpd process, run:
watch "ntpq -p"
Press Ctrl-c to stop watching the process.
Note the information in the following columns:
The character in the first column indicates the quality of the source.
The asterisk ( * ) indicates the source is the current reference.
remote lists the IP address or host name of the source.
when indicates how many seconds have passed since the source was polled.
poll indicates the polling interval. This value increases depending on the accuracy of the local clock.
reach is an octal number that indicates reachability of the source. A value of 377 indicates the source has answered the last eight consecutive polls.
offset is the time difference between the source and the local clock in milliseconds.
Note: If ntpq -p returns ntpq: read: Connection refused, check /var/log/messages for a more detailed error message.
Additional documentation for NTP is available at .
ESX Sever 2.0.0
To install and configure NTP on the console operating system (service console):
Log on to the console as the root user.
Mount the ESX Server CD on the console:
mount /mnt/cdrom
Change to the /mnt/cdrom/RedHat/RPMS directory.
Install the NTP package:
rpm -Uhv ntp-*.i386.rpm
Change to the /etc directory.
You should be able to find a file called ntp.conf (after the rpm installation) in the /etc directory (you changed into this directory in step 5). Edit this file using:
vi ntp.conf
Find the the line that reads:
server 127.127.1.0 # local clock
Change it to:
server 192.6.38.127 # This is an example only
Save the file.
Create a file named step-tickers in the /etc/ntp directory. In this file, list the host name of your reference time server.
To enable the ntp daemon to autostart when the server is rebooted, run:
chkconfig ntpd on
To start it now without rebooting, run:
/etc/rc.d/init.d/ntpd start
The time drift corrects after a while.
Wait a few seconds (up to a minute), then run:
ntpq -p
This lists the current status.
Use umount to unmount the ESX Server CD:
umount /mnt/cdrom
These examples use a source server IP address obtained from a list of open access NTP servers. You may select one that suits you from .
