[rrd-users] Re: How to save and graph NON-averages
Sam Howard
Sam.Howard at trcinc.com
Mon Sep 25 22:16:37 MEST 2000
Ok, I got most of that... :)
Your comment about time != steptime got me looking at the RRD itself.
When I dump the database, with a steptime=86400 (1 day), I notice that it
chooses 20:00 (8:pm) as the time, even though I am running the create script
at 16:00 (4:pm). As seen by:
<!-- 1997-10-05 20:00:00 EDT --> <row><v> NaN </v><v> NaN </v></row>
But, here's where it get's weird (and maybe why my values are getting
"helped")...after a few days, in the database, the time *changes*:
<!-- 1997-10-25 20:00:00 EDT --> <row><v> NaN </v><v> NaN </v></row>
<!-- 1997-10-26 19:00:00 EST --> <row><v> NaN </v><v> NaN </v></row>
It goes back and forth every so many months!!
Now, if I can get the database to remain consistent with respect to its
timestamp for each record, I can work around the time skew of my record -- I
am already lying about the time of day, as I am only submitting one value
per day (I can say it's 20:00, if I want to...no big deal).
I notice that if I reduce the steptime, to anything other than 3600 or 7200,
that I get time fluctuations...
I hate to waste so many records (24 per day) just to work around this...
What's up?
Thanks,
Sam
-----Original Message-----
From: Alex van den Bogaerdt [mailto:alex at slot.hollandcasino.nl]
Sent: Monday, September 25, 2000 3:46 PM
To: Sam.Howard at trcinc.com
Cc: rrd-users at list.ee.ethz.ch
Subject: [rrd-users] Re: How to save and graph NON-averages
Sam Howard wrote:
>
> I'm trying to create some very simple graphs of our backup totals (i.e.
> bytes backed up per night), but I can't seem to get RRD to NOT average out
> the values...
>
> For example, I am creating my database as:
>
> rrdtool create nsr.rrd \
> --start '07/01/2000' \
> --step 86400 \
> DS:incrbytes:GAUGE:86400:U:U \
> DS:fullbytes:GAUGE:86400:U:U \
> RRA:LAST:0.5:1:1000
>
> Inserting data as:
>
> + rrdtool update nsr.rrd 967392000:77736128:0
> + rrdtool update nsr.rrd 967478400:0:1447213300
You use a time that is not an *exact* multiplication of your step time.
If (time % steptime) != 0, interpolation (better: normalizing) will occur.
If you need/want to keep the original values:
use GAUGE and make sure to update on RRDtool boundaries. If you want to
update at Sun Aug 27 16:00:00 2000 UTC, you need to use a step time of
3600. Make sure to use an appropriate xff !
regards,
--
__________________________________________________________________
/ alex at slot.hollandcasino.nl alex at ergens.op.het.net \
| work private |
| My employer is capable of speaking therefore I speak only for myself |
+----------------------------------------------------------------------+
| Technical questions sent directly to me will be nuked. Use the list. |
+----------------------------------------------------------------------+
| http://faq.mrtg.org/ |
| http://rrdtool.eu.org --> tutorial |
+----------------------------------------------------------------------+
--
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
--
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
More information about the rrd-users
mailing list