[nycbug-talk] dealing with drift

[Isaac Levy]

Thanks yall',

On Feb 28, 2005, at 12:33 PM, George Georgalis wrote:

>> Hi All,
>>
>> I was just wanting to post on some of the post-meeting conversations
>> I've repeatedly over-heard about managing time-drift across clusters 
>> of
>> machines- (you know who you people are...)
>>
>> How do folks like to do this?  Special tools beyond rockin' timed?
>> Anyone know any good urls on the topic (either practical or acidemic)?
>
> Am I missing something here? Isn't the obvious answer, for a cluster, 
> to
> configure the headnode (whether is connected to internet and synced or
> not) to answer ntp queries from nodes, and configure the nodes ntp time
> server to use their headnode?  /usr/ports/net/ntp

On Feb 28, 2005, at 12:43 PM, Dan Langille wrote:

> Sounds like much the same way I manage my home network.  My gateway
> acts as time master, and all other machines on my network sync with
> it.  In turn, the gateway syncs with an external source.


Cool.  My question then changes to this:

Are there any tricks to maintain redundancy with ntp servers?  i.e. is 
there a builtin/traditional way to configure things so that *all* 
machines can act as ntp servers in the event that the ntp box goes el' 
muerte?

Rocket-
.ike