[rrd-users] Re: Finding time for latest primary data point

[Anders Nygren]

On Thursday 29 January 2004 02:16, Alex van den Bogaerdt wrote:
> On Wed, Jan 28, 2004 at 07:46:04PM -0600, Anders Nygren wrote:
> > What I am planning to do is
> > 1, "normal" data collection, and storing the data in rrdtool
> > 2, a separate process that periodically, (after the update), gets the
> > latest values from rrdtool and does threshold checks for generating
> > alarms.
>
> Inbetween (1) and (2), time will move forward.
>
> If 't' is T modulo R, time _could_ go from (t+n) to (t) and you will
> be looking at the wrong interval.  This won't happen often however it
> is possible and it is something to take care of.

But in (2) I will know the time used in the update, so I can use the same
time in the fetch. That would work, right?

> > rrdtool report <file> <CF> --resolution <R> --start <T> --end <T>
> > where <T> is the most recent time that is an integer multiple of <R>.
>
> Do you mean fetch where you write report, or do I need to brush up
> on my rrdtool knowledge?

Yes, of course, I meant fetch, or possibly xport if I need to do some 
calculations on the data.

/Anders

--
Unsubscribe mailto:rrd-users-request at list.ee.ethz.ch?subject=unsubscribe
Help        mailto:rrd-users-request at list.ee.ethz.ch?subject=help
Archive     http://www.ee.ethz.ch/~slist/rrd-users
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi