Out client have 2 x HP ML370G4 X3.4 2MB Hi Perf EU Rack Srvr with 384954-B21 RHEL 4 AS Std Both systems gain about 10 minutes per hour. Where do I start looking for the problem ?
Are you using NTP? Are your timezone correctly configured? Check your BIOS time and ensure that the /etc/sysconfig/clock, ZONE UTC and ARC settings are correct.
Por que hacerlo dificil si es posible hacerlo facil? - Why do it the hard way, when you can do it the easy way?
Thank you. The site is in Tanzania and I am in South Africa. It will take about 18 hours to get tried, so hopefully by to-morrow night I will know. They claim that they are not using NTP, but I was not on site during installation
I think what Ivan is trying to say is that you SHOULD be running ntpd.
Check the man pages on ntpd and ntpdate. In a nutshell, here are the commands I call from a cronjob to set the time. Note that ntpd has to be stopped before running netdate.
I concur ntp is the way forward. You may find that tock.nml.CSIR.co.za gives you a better ntp response than the global pool for your Tanzanian location It is fed by the CSIR atomic clocks in Pretoria which are part of the UTC aggregation .
I now that the earlier posts are years ago, never the less I am experiancing a "time drift" too.
I and using ntp and the tock.... CSIR time server too on a HP ServerTower ProLiant ML350 G6, QC Xeon E5620(2.40GHz)
12MB, 8 x SFF, P410i/256MB, 6GB(3x2GB), 460W PSU, DVD-ROM using VMware ESX, 4.1.0, 320092 and am experiancing a once of 10min time drift that only appears when the server has to sync the time with the CSIR time server after a reboot of the VMware host. Any suggestions?
