SCIENTIFIC-LINUX-USERS Archives

May 2011

SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV
Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
Re: clock drift under Hyper-V
From:
Dag Wieers <[log in to unmask]>
Reply To:
Dag Wieers <[log in to unmask]>
Date:
Wed, 1 Jun 2011 03:20:23 +0200
Content-Type:
TEXT/PLAIN
Parts/Attachments:
TEXT/PLAIN (23 lines) 
On Tue, 31 May 2011, Werf, C.G. van der (Carel) wrote:

> In my opinion you should NEVER use NTPD in a virtual machine.....
>
> Check out how NTPD works... it tries to adjust "software cycles"to "hardware cycles".
>
> "Hardware cycles" are a bit undefined in a virtual machine.

It keeps the (virtual) system clock in sync with an external source. This 
works better than using a timer interrupt, especially when the hypervisor 
cannot commit to a certain interrupts per second. (Which used to be a 
problem eg when HZ set to 1000 in VMs)

Using ntpd is definitely a best practice we have been using even before 
VMware recommended it. We had more troubles with VMware's host-guest 
synchronization. You just have the configure ntpd correctly.

-- 
-- dag wieers, [log in to unmask], http://dag.wieers.com/
-- dagit linux solutions, [log in to unmask], http://dagit.net/

[Any errors in spelling, tact or fact are transmission errors]

ATOM RSS1 RSS2