Time Drift on HA Cluster

[jonadam5]

Hi,

I have a curly issue that I have been unable to solve.
I have 2 SPLAT PRO gateways running on IBM x206 Servers. The problem is that the clocks are running too fast. The 2 gateways stay in perfect time sync with each other but not with anything else on our Network.
I have gone through the time zone settings under sysconfig on a couple of occasions so I'm sure they are set correctly. I have also used the ntp -n command to sync the time against the separate Smart centre server. There is also an accept rule in my policy to allow this traffic.

If anyone has any ideas I would really appreciate it.
Jon

[melipla]

First, set up your ntp file with an ntp server, I've displayed mine as a reference. You can find a more complete list of ntp servers here. The interval specifies how often to sync the time in seconds.

# cat /etc/sysconfig/ntp
INTERVAL=3600
SERVER1=pool.ntp.org
USE_NTP=true

Then start the ntp service:

# ntpstart
#

Check the status:

# ntpstat
ntp is running
#

You should set this up on all of your gateways irregardless of whether or not they're exhibiting time drifts.

[RobertGraham]

Don't forget to create a drift file, that's important. But, melipla is correct.

Not that familiar with SPLAT, but Linux has its own version of the clock that is different from the hardware clock on the board. Most flavors have a hwclock command for setting this. It should be available in expert mode.

Check the manpage on the web for more information on this command.

[jonadam5]

Thanks for the Feedback.
I have edited my NTP file and am now getting success. The main problem was the NTP servers I previously tried syncing from. I had tried a couple and felt sure they were ok as I've used them previously from other Windows servers.

I'm not sure about the drift file, I can't find much information about it. Does it tell SPLAT how to deal with the time skew?