Re: [Dovecot] Time jumped forwards
On Fri Jan 14 07:23:32 EET 2011, Eric Shubert wrote:
In particular, see NTP Recommendations and VMware Tools bits at the bottom.
Thanks, I've read through that document a couple of times now & I'm fairly sure my system 'ticks all the boxes' yet the warning messages continues.
I can see no noticeable time discrepancy when I do periodic time checks from the command line comparing against another system's clock - certainly not 2 minute error margin
Could this be a time mismatch between imap client & server?
Ian.
On 14/01/11 15:14, Ian B wrote:
On Fri Jan 14 07:23:32 EET 2011, Eric Shubert wrote:
In particular, see NTP Recommendations and VMware Tools bits at the bottom.
Thanks, I've read through that document a couple of times now & I'm fairly sure my system 'ticks all the boxes' yet the warning messages continues.
There is no checklist on the page, but a list of recommendations. that does not mean that you can use all of them at the same time.
I had this problem too when I setup a VMware guest last month (noticed through strange spikes in munin's ntp monitoring, not through dovecot logs). After some research I found out that I had VMware time synchronisation (host to guest) enabled, and NTP in the guest running. The ESX host clock was some 90+ seconds behind, so VMware stalled my clock once in a while, after which NTP corrected it again.
I asked the ESX admin to fix the clock, but disabled the VMware synchronisation and now use only ntp in the guest. Never looked back.
So in stead of 'ticking all the boxes', explain your setup. You have NTP enabled, but what does VMware do?
Could this be a time mismatch between imap client & server?
No.
-- Tom
On 2011-01-14 9:32 AM, Tom Hendrikx wrote:
I asked the ESX admin to fix the clock, but disabled the VMware synchronisation and now use only ntp in the guest. Never looked back.
Interesting... I thought this was the one thing that you were *never* supposed to do??
--
Best regards,
Charles
On 14/01/11 18:18, Charles Marcus wrote:
On 2011-01-14 9:32 AM, Tom Hendrikx wrote:
I asked the ESX admin to fix the clock, but disabled the VMware synchronisation and now use only ntp in the guest. Never looked back.
Interesting... I thought this was the one thing that you were *never* supposed to do??
I enabled (VMware tools) time sync in the first place because I had heard something about not running NTP in a virtualized guest OS, but after further reading I found out that in recent virtualization setups, there are no reasons not to use ntp. It is even advised to use ntp and not period synchronisation with the host, as ntp results in much more accurate timekeeping.
Running VMWare tools time sync and ntp together is not good. This happened to me because I erroneously assumed that time sync was not working when my clock was 90+ seconds off. I enabled ntp before investigating why time sync resulted in a faulty time...
When feeling the need for a deeper read: http://www.vmware.com/files/pdf/Timekeeping-In-VirtualMachines.pdf
-- Regards, Tom
On 2011-01-14 1:44 PM, Tom Hendrikx wrote:
When feeling the need for a deeper read: http://www.vmware.com/files/pdf/Timekeeping-In-VirtualMachines.pdf
Cool... thanks for the explanation and link...
--
Best regards,
Charles
participants (3)
-
Charles Marcus
-
Ian B
-
Tom Hendrikx