[rrd-users] problem with lost of data, file size always the same!

[Francesco Andreozzi]

Hi Nuno,
Thankyou for your reply,
here some more information...

# rrdtool info e_m153_o69.rrd

filename = "e_m153_o69.rrd"
rrd_version = "0003"
step = 300
last_update = 1223021717
ds[e_m153_o69].type = "GAUGE"
ds[e_m153_o69].minimal_heartbeat = 600
ds[e_m153_o69].min = 0.0000000000e+00
ds[e_m153_o69].max = NaN
ds[e_m153_o69].last_ds = "UNKN"
ds[e_m153_o69].value = 3.0357484375e+04
ds[e_m153_o69].unknown_sec = 0
rra[0].cf = "MIN"
rra[0].rows = 1440
rra[0].pdp_per_row = 12
rra[0].xff = 5.0000000000e-01
rra[0].cdp_prep[0].value = 1.7854259896e+03
rra[0].cdp_prep[0].unknown_datapoints = 0
rra[1].cf = "MAX"
rra[1].rows = 1440
rra[1].pdp_per_row = 12
rra[1].xff = 5.0000000000e-01
rra[1].cdp_prep[0].value = 1.7857299479e+03
rra[1].cdp_prep[0].unknown_datapoints = 0
rra[2].cf = "AVERAGE"
rra[2].rows = 1440
rra[2].pdp_per_row = 1
rra[2].xff = 5.0000000000e-01
rra[2].cdp_prep[0].value = NaN
rra[2].cdp_prep[0].unknown_datapoints = 0


we want to use a collection time every 300second / 5 minutes

the real rrdtool create is like this

rrdtool create e_m153_o69.rrd --start <my-date> \ (my-date depends, it  
is dinamic)
           DS:e_m153_o69:GAUGE:300:0:U \
           RRA:MIN:0.5:12:1440 \
           RRA:MAX:0.5:12:1440 \
           RRA:AVERAGE:0.5:1:1440

the goal of this rrd is to collect data for a machine (something like  
10years of data) and give the possibility to create data for a  
specific date range, also of some years or month before now. with 5  
minutes resolution

Thankyou for your help

ciao
Francesco


NB i've reat the beginner guide but i'm still confused!



Quoting Nuno Pereira <nunopereira at interacesso.pt>:

>> -----Mensagem original-----
>> De: rrd-users-bounces at lists.oetiker.ch [mailto:rrd-users-
>> bounces at lists.oetiker.ch] Em nome de Francesco Andreozzi
>> Enviada: sexta-feira, 3 de Outubro de 2008 12:08
>> Para: rrd-users at lists.oetiker.ch
>> Assunto: [rrd-users] problem with lost of data, file size always the
>> same!
>>
>> Hi all, this is my first post and i hope to be clear :D
> This is a case where a FTFM would be apropriate
> (http://oss.oetiker.ch/rrdtool/doc/rrdcreate.en.html), but let's try to
> figure a few things out.
>
>
>> i have start to using rrd because seems to be really nice stuff....
>>
>> i create a lot of rrd to collect data from a specific device ...
>>
>> i create an rrd with the command like this
>>
>> rrdtool create <RRDNAME.rrd> --start <my-date> \
>>          DS:<my-obj>:GAUGE:<my-time>0:U \
>>          RRA:MIN:0.5:12:1440 \
>>          RRA:MAX:0.5:12:1440 \
>>          RRA:AVERAGE:0.5:1:1440
>>
>> the collection seems to be good except i'm lost some data!!!...
>
> If I'm not wrong, you have just created an RRD which hás just one sample,
> that is a GAUGE, and you collect 1440 points of the MIN and MAX value of
> each 12 samples. Each sample, you store the sample AVERAGE (which would be
> the same to put there a MAX and MIN, as you have only one value to average).
> Clarify one thing: give us na example of what you put in <my-time> in the
> line " DS:<my-obj>:GAUGE:<my-time>0:U \". You missed a semicolon there,
> between <my-time> (heartbeat) and the 0 (minimal accpetable variable).
>
>> if i do a rrdtool dump I can see min and max value from the starting
>> time, but avarage no, the first values are lost every 5 minutes, when
>> i do the rrdtool update!
> You have 4 sections (or 2, with the last divided in 3). Check out the end of
> the file, or whatch the ouput of "rrdtool info", as you'll see there an
> AVERAGE.
>
>>
>> the file size is always the same and it doesn't increase!
> That happens with all the RRDs that aren't resized.
>
>> when i'm creating graph i really want to use avarage values! but on
>> this way i think i will gona lost all this information!
> Only if you are putting samples with a step that is lower than step (300 in
> your case, which is default), and in this case you could put a smaller step,
> if you need. Or if you don't update the RRD with samples in a period bigger
> than heartbeat (your <my-time>).
>
>
>>
>> Ciao
>> Francesco
> Adeus,
> Nuno Pereira
>
>
>