Re: [rrd-users] Malloc error when creating graph (Debian on

Hi Tobi,

thanks for following up on this. However, your suggestion does not make
any difference.

Here is a simple test case, modeled after my original setup:

rrdtool create temptest.rrd --start 1201691880 --step 60 \
DS:AF:GAUGE:120:U:U \
RRA:AVERAGE:0.5:1:11520

(No data loaded into the rrdtool database!)

rrdtool graph temptest.png \
--end now --start end-48799 -a PNG -t "Temperaturverlauf 4h" \
--vertical-label "Temperatur =B0C" \
--width 800 --height 400 \
DEF:aussen=3Dtemptest.rrd:AF:AVERAGE \
CDEF:aus=3Daussen,10,/ \
LINE1:aus#ff0000:"Aussen"

results in: ERROR: malloc im->gdes[gdi].data

When I change the "start" parameter of the "rrdtool graph" command
to "end-48800" (just one second more!), the expected empty graph is
produced.

Could somebody with a development environment on a NSLU2 running Debian
reproduce this test case and maybe debug what is happening?

Regards, Richard

> Richard,
>
> one idea that just occurs to me is that maybe by reducing the time
> span a smaller higher resolution RRA gets picked from your rrd and
> this actually leads to MORE date eventouch the time span is shorter
> ...
>
> for testing, try an rrd with only one RRA
>
> cheers
> tobi
>
> Today u7l11ey@mail.lrz-muenchen.de wrote:
>
>> Hi all,
>>
>> I get a strange error when creating a graph. My system is a Linksys
>> NSLU2
>> running Debian (Linux slug 2.6.18-5-ixp4xx #1 Sun Dec 23 05:17:39 UTC
>> 2007
>> armv5tel GNU/Linux), I'm using rrdtool 1.2.15. The same problem has also
>> been reported for 1.2.23, see ticket #111 on the rrdtool website.
>>
>> This is the graph definition:
>>
>> rrdtool graph temperatur4.png \
>> --end now --start end-48799 -a PNG -t "Temperaturverlauf 4h" \
>> --vertical-label "Temperatur =B0C" \
>> --width 800 --height 400 \
>> DEF:aussen=3Dtemperatur.rrd:AF:AVERAGE \
>> CDEF:aus=3Daussen,10,/ \
>> LINE1:aus#ff0000:"Aussen"
>>
>> Running it produces "ERROR: malloc im->gdes[gdi].data". Notice the
>> parameter "--start end-48799": The error happens for any values of the
>> subtractor below 48800! Any values equal or above 48800 will produce the
>> graph as desired. It also works if the "CDEF" or the size definition is
>> omitted, but that is not what I want.
>>
>> I am confused since a graph containing a larger time span shoud use more
>> resources and therefore be more likely to fail. But in this case, it
>> fails
>> if the graph contains a shorter time span.
>>
>> Is anybody running rrdtool on their NSLU2 and also having similar
>> problems
>> and/or a solution? I'll be happy to supply a complete test case if it
>> helps.
>>
>> Regards, Richard
>>
>> _______________________________________________
>> rrd-users mailing list
>> rrd-users@lists.oetiker.ch
>> https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users
>>
>>
>
> --
> Tobi Oetiker, OETIKER+PARTNER AG, Aarweg 15 CH-4600 Olten
> http://it.oetiker.ch tobi@oetiker.ch ++41 62 213 9902


_______________________________________________
rrd-users mailing list
rrd-users@lists.oetiker.ch
https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users