[7.400] Slow clock in Astaro VM

Monarch · #1 (**permalink**) 03-15-2009, 11:09 PM

Hi,

I'm, using ASG running inside VMware Server 2.0 upon Windows Server 2008. Hardware as noted below.

Since I did a fresh install (using the 7.400 ISO image) the clock gets out of sync. That means it looses a few minutes per day. That wasn't the case after my last fresh installation (using the 7.304 image).

What can I do to correct this issue?

Regards,
Bastian

wingman · #2 (**permalink**) 03-15-2009, 11:13 PM

Quote:

Originally Posted by Monarch

Hi,

I'm, using ASG running inside VMware Server 2.0 upon Windows Server 2008. Hardware as noted below.

Since I did a fresh install (using the 7.400 ISO image) the clock gets out of sync. That means it looses a few minutes per day. That wasn't the case after my last fresh installation (using the 7.304 image).

What can I do to correct this issue?

Regards,
Bastian

How long are you running the 7.4 version for?
Is ntp enabled and allowed to access the internet?(system logs?)

Monarch · #3 (**permalink**) 03-15-2009, 11:33 PM

V7.4 has been running for over 10 days (uptime). During this time I now corrected the system clock three times manually, it was at least 10 minutes behind the real time.

NTP is set to the default (pool.ntp.org). The only log entries I could find are the following (excerpt):

Code:

var/log/system/2009/03/system-2009-03-13.log.gz:2009:03:13-04:00:44 astaro1 ntpd[4680]: Deleting interface #15 ipsec0, 91.20.96.10#123, interface stats: received=2, sent=2, dropped=0, active_time=85200 secs
/var/log/system/2009/03/system-2009-03-14.log.gz:2009:03:14-04:00:44 astaro1 ntpd[4680]: Listening on interface #17 ipsec0, 91.20.113.170#123 Enabled
/var/log/system/2009/03/system-2009-03-14.log.gz:2009:03:14-04:00:44 astaro1 ntpd[4680]: Deleting interface #16 ipsec0, 91.20.105.233#123, interface stats: received=3, sent=3, dropped=0, active_time=86400 secs
/var/log/system.log:2009:03:15-04:00:44 astaro1 ntpd[4680]: Listening on interface #18 ipsec0, 91.20.68.89#123 Enabled
/var/log/system.log:2009:03:15-04:00:44 astaro1 ntpd[4680]: Deleting interface #17 ipsec0, 91.20.113.170#123, interface stats: received=3, sent=3, dropped=0, active_time=86400 secs

Regards,
Bastian

RFCat_vk · #4 (**permalink**) 03-16-2009, 08:52 AM

Hi,
from my understanding you can't run a vm partition as a time reference due to loss/non use of processor cycles. You have to have a real network element as the time reference then refer the vm guest to it.

Ian M

BAlfson · #5 (**permalink**) 03-16-2009, 12:25 PM

Ian, I don't understand. How does one configure the Astaro to obviate this problem? Or, is this a VMware setting we need to check?

Thanks - Bob

RFCat_vk · #6 (**permalink**) 03-16-2009, 07:59 PM

Hi folks,
my comment comes from the experience we are having with time slip on a vmesx box at work.
From my understanding you need to ensure the host hardware clock is in synch with the vm host software clock, but referencing a real ntp server.

We have been advised by "experts" to setup a real device as an ntp server on our network and that we shouldn't use a vmguest as a time server because of the cycle loss.

That is as much as I know.

Ian M

Monarch · #7 (**permalink**) 03-16-2009, 10:30 PM

Indeed, and therefore I'm not using the virtual ASG as a time source in my network. But it's not good when times in reports and logfiles aren't accurate.

How can the VM's clock be held in sync with the host's clock? As I already wrote, in my last installation I didn't have that problem at all

Hmm, I just remember an article about that problem I read a long time ago:
HELP: VMware - Fixing Time Keeping Problems with the Guest OS : VMblog.com - Virtualization Technology News and Information for Everyone

I'll try these settings

BarryG · #8 (**permalink**) 03-17-2009, 11:01 PM

Using the official Astaro VMWare image is recommended for issues like this; however I don't know if there's one for 7.4 yet.

Barry

Monarch · #9 (**permalink**) 03-18-2009, 04:21 PM

I think the last link was the solution.

Clock is now in sync (bang on time) since Monday.

Full content of my config.ini:

Code:

authd.client.port = "902"
authd.proxy.vim = "vmware-hostd:hostd-vmdb"
authd.proxy.nfc = "vmware-hostd:ha-nfc"
host.cpukHz = "2500000"
host.noTSC = "TRUE"
ptsc.noTSC = "TRUE"

Regards,
Bastian