We just noticed that the time was very far off on our sparkly new VMware EXS 3.5 server. When I went to run ntpdate to bring it up to sync, I was suprised to find that it could not make a connection to the time server because outbound UDP 123 traffic was blocked by the internal firewall. Here is what I got:
# /usr/sbin/ntpdate -u time.nist.gov
9 Apr 03:47:53 ntpdate: sendto(18.104.22.168): Operation not permitted
9 Apr 03:47:54 ntpdate: sendto(22.214.171.124): Operation not permitted
9 Apr 03:47:55 ntpdate: sendto(126.96.36.199): Operation not permitted
9 Apr 03:47:56 ntpdate: sendto(188.8.131.52): Operation not permitted
9 Apr 03:47:57 ntpdate: no server suitable for synchronization found
Normally I would just add a rule to the “/etc/sysconfig/iptables” file to allow traffic out on this port, but Vmware ESX server does not use iptables… It uses its own firewall, so I had to figure out how to change it. Happily, it turns out that there is a handy “esxcfg-firewall” command built just for such things.
When many new servers are delivered from the factory, the system clock is way off. Most UNIX systems run “ntpd” to keep the time in sync with internet time servers, which are, in turn synchronized against an atomic clock. This results in a system time that is very very close to the “actual” time of day. The downside, however, is that even a properly configured “ntpd” will not synchronize the system clock if it is too far out of sync with the time server. To remedy this, we first have to run “ntpdate” to get the system clock close to the correct time, and then enable “ntpd” to keep it there.
The first thing we have to do is “ntpd” to free up the port for “ntpdate”:
[root@server /]# /sbin/service ntpd stop
Shutting down ntpd: [ OK ]
This frees up the port for ntpdate. Next we run:
[root@server /]# /usr/sbin/ntpdate time.apple.com
Now the time should be set correctly. We then change the default time servers to something like the following in /etc/ntp.conf: