[rrd-users] Reversed values

[Andy Riebs]

Marc,

If the data that you are gathering are the total number of current users
on a set of systems, for example, you probably want to use "GAUGE"
instead of "DERIVE" in your rrd.  DERIVE (and "COUNTER", for that
matter) both assume numbers that should constantly increase.

hth!
/andy

On Mon, 2007-12-03 at 15:25 +0000, Marc T. Kaplan wrote:
> Andy,
> 
> 
>    You are right, I was not sure if it was a specific issue or general
> given RRDtools. I posted parts of the script you referenced in your
> reply below. I have run different values and tried reading up on RRD
> to understand it better, given the specific database I am trying to
> create. This is more or less a project that I was thrown into head
> first and I am still trying to understand everything related to RRD.
> If you see anything specific regarding the data below please tell me
> what might be wrong. 
> 
> ******************************************************************************
> 
>                 rrdtool create $rrd_dir/usr.rrd -s 86400 \
> DS:alpha:DERIVE:172800:U:U DS:delta:DERIVE:172800:U:U
> DS:eta:DERIVE:172800:U:U DS:epsilon:DERIVE:172800:U:U \ 
> DS:gamma:DERIVE:172800:U:U DS:iota:DERIVE:172800:U:U
> DS:kappa:DERIVE:172800:U:U DS:lambda:DERIVE:172800:U:U \
> DS:omega:DERIVE:172800:U:U DS:theta:DERIVE:172800:U:U
> DS:xi:DERIVE:172800:U:U DS:zeta:DERIVE:172800:U:U \ 
>                         RRA:AVERAGE:0.5:2:90; \
> 
> ******************************************************************************
> 
> alpha=$(grep alpha $fileloc | awk -F': ' '{print $2}') \
> delta=$(grep delta $fileloc | awk -F': ' '{print $2}') \
> eta=$(grep ' eta' $fileloc | awk -F': ' '{print $2}') \
> epsilon=$(grep epsilon $fileloc | awk -F': ' '{print $2}') \ 
> gamma=$(grep gamma $fileloc | awk -F': ' '{print $2}') \
> iota=$(grep iota $fileloc | awk -F': ' '{print $2}') \
> kappa=$(grep kappa $fileloc | awk -F': ' '{print $2}') \ 
> lambda=$(grep lambda $fileloc | awk -F': ' '{print $2}') \
> omega=$(grep omega $fileloc | awk -F': ' '{print $2}') \
> theta=$(grep theta $fileloc | awk -F': ' '{print $2}') \ 
> xi=$(grep xi $fileloc | awk -F': ' '{print $2}') \
> zeta=$(grep zeta $fileloc | awk -F': ' '{print $2}')
> 
> rrdtool update $rrd_dir/usr.rrd N:$alpha:$delta:$eta:$epsilon:$gamma:
> $iota:$kappa:$lambda:$omega:$theta:$xi:$zeta; 
> 
> *****************************************************************************
> 
> rrdtool graph usr.rrd.gif --title="Users per server" \
> --start -1m  \
> --vertical-label=Users \
> --alt-autoscale \ 
> --units-length 10 \
> --units-exponent 0 \
> DEF:pta=usr.rrd:alpha:AVERAGE LINE1:pta#000000:Alpha \
> DEF:ptd=usr.rrd:delta:AVERAGE LINE1:ptd#151B54:Delta \
> DEF:pte=usr.rrd:eta:AVERAGE LINE1:pte#463E41:Eta \
> DEF:ptep=usr.rrd:epsilon:AVERAGE LINE1:ptep#800517:Epsilon \
> DEF:ptg=usr.rrd:gamma:AVERAGE LINE1:ptg#8E35EF:Gamma \
> DEF:pti=usr.rrd:iota:AVERAGE LINE1:pti#F778A1:Iota \
> DEF:ptk=usr.rrd:kappa:AVERAGE LINE1:ptk#6698FF:Kappa \ 
> DEF:ptl=usr.rrd:lambda:AVERAGE LINE1:ptl#667C26:Lambda \
> DEF:pto=usr.rrd:omega:AVERAGE LINE1:pto#87F717:Omega \
> DEF:ptt=usr.rrd:theta:AVERAGE LINE1:ptt#FFFC17:Theta \
> DEF:ptx=usr.rrd:xi:AVERAGE LINE1:ptx#C47451:Xi \ 
> DEF:ptz=usr.rrd:zeta:AVERAGE LINE1:ptz#806517:Zeta
> 
> 
> On Dec 3, 2007 10:18 AM, Andy Riebs <andy.riebs at hp.com> wrote:
>         Marc,
>         
>         "Reversed values" is not the norm :)
>         
>         It would help to send along the commands you are using the
>         create the
>         graph, and, perhaps, the commands that you are using to store
>         the data.
>         
>         /andy 
>         
>         
>         On Mon, 2007-12-03 at 15:11 +0000, Marc T. Kaplan wrote:
>         > Hello all,
>         >
>         >
>         >     Just a quick question, I have gotten my RRD setup and it
>         is
>         > running and storing correctly but upon graphing the data I
>         am 
>         > receiving data that is a bit strange. It seems that
>         everything is
>         > getting graphed in reverse, that is that everything is
>         showing as a
>         > negative when it should be positive. Is there anything
>         specific I 
>         > should look up or do to change the values so they are
>         plotted
>         > correctly?
>         
>         --
>         Andy Riebs,     HP High Performance Computing R&D
>                     Technology for better business outcomes 
>         Linux Software,      <http://www.hp.com/go/clusters>
>         Catalyst Program,   <http://www.hp.com/go/catalysts>
>         (w) +1.603.884.1521, andy.riebs at hp.com
>         (h) andy at candooz.com
>         My opinions are not necessarily those of HP
>         
> 
-- 
Andy Riebs,     HP High Performance Computing R&D
             Technology for better business outcomes
Linux Software,      <http://www.hp.com/go/clusters>
Catalyst Program,   <http://www.hp.com/go/catalysts>
(w) +1.603.884.1521, andy.riebs at hp.com
(h) andy at candooz.com
My opinions are not necessarily those of HP