y-axis scaling change from 1.0.46 to 1.2.12?

[Christian Boehme]

Hello,

I use cacti as a front end to rrdtool. When upgrading the cacti server I
noticed a change in the graphs which seems to be caused by the upgrade
of rrdtool from 1.0.46 to 1.2.12. With the older rrdtool the graphs look
like this:

http://wwwuser.gwdg.de/~cboehme1/cacti_old.png

This is how this graph is produced:

/usr/bin/rrdtool graph - \
--imgformat=PNG \
--start=-86400 \
--end=-300 \
--title="LSF - gwdg-pcpar Availability" \
--rigid \
--base=1000 \
--height=120 \
--width=500 \
--upper-limit=252 \
--lower-limit=0 \
--vertical-label="" \
DEF:a="/var/lib/cacti/full_11.rrd":full:AVERAGE \
DEF:b="/var/lib/cacti/full_11.rrd":resv:AVERAGE \
DEF:c="/var/lib/cacti/full_11.rrd":free:AVERAGE \
DEF:d="/var/lib/cacti/full_11.rrd":avail:AVERAGE \
CDEF:cdefc=a,252,/,100,* \
CDEF:cdeff=b,252,/,100,* \
CDEF:cdefi=c,252,/,100,* \
CDEF:cdefbb=d,252,/,100,* \
AREA:a#FFF200:"Used Slots"  \
GPRINT:a:LAST:"          Current\:%8.0lf"  \
GPRINT:cdefc:AVERAGE:"Average(%%)\:%8.2lf %s\n"  \
STACK:b#FF7D00:"Reserved Slots"  \
GPRINT:b:LAST:"      Current\:%8.0lf"  \
GPRINT:cdeff:AVERAGE:"Average(%%)\:%8.2lf %s\n"  \
STACK:c#FF0000:"Free Slots"  \
GPRINT:c:LAST:"          Current\:%8.0lf"  \
GPRINT:cdefi:AVERAGE:"Average(%%)\:%8.2lf %s\n"  \
LINE2:d#0000FF:"User Available Slots"  \
GPRINT:d:LAST:"Current\:%8.0lf"  \
GPRINT:cdefbb:AVERAGE:"Average(%%)\:%8.2lf %s"

With the newer rrdtool it looks like this:

http://wwwuser.gwdg.de/~cboehme1/cacti_new.png

This is the rrdtool command.

/usr/bin/rrdtool graph - \
--imgformat=PNG \
--start=-86400 \
--end=-300 \
--title="LSF - gwdg-pcpar Availability" \
--rigid \
--base=1000 \
--height=120 \
--width=500 \
--upper-limit=252 \
--lower-limit=0 \
--vertical-label="" \
--slope-mode \
--font TITLE:12: \
--font AXIS:8: \
--font LEGEND:10: \
--font UNIT:8: \
DEF:a="/srv/www/cacti-0.8.7b/rra/full_11.rrd":full:AVERAGE \
DEF:b="/srv/www/cacti-0.8.7b/rra/full_11.rrd":full:MAX \
DEF:c="/srv/www/cacti-0.8.7b/rra/full_11.rrd":resv:AVERAGE \
DEF:d="/srv/www/cacti-0.8.7b/rra/full_11.rrd":resv:MAX \
DEF:e="/srv/www/cacti-0.8.7b/rra/full_11.rrd":free:AVERAGE \
DEF:f="/srv/www/cacti-0.8.7b/rra/full_11.rrd":free:MAX \
DEF:g="/srv/www/cacti-0.8.7b/rra/full_11.rrd":avail:AVERAGE \
DEF:h="/srv/www/cacti-0.8.7b/rra/full_11.rrd":avail:MAX \
CDEF:cdefc=a,252,/,100,* \
CDEF:cdeff=c,252,/,100,* \
CDEF:cdefi=e,252,/,100,* \
CDEF:cdefbb=g,252,/,100,* \
AREA:a#FFF200FF:"Used Slots"  \
GPRINT:a:LAST:"          Current\:%8.0lf"  \
GPRINT:cdefc:AVERAGE:"Average(%%)\:%8.2lf %s\n"  \
AREA:c#FF7D00FF:"Reserved Slots":STACK \
GPRINT:c:LAST:"      Current\:%8.0lf"  \
GPRINT:cdeff:AVERAGE:"Average(%%)\:%8.2lf %s\n"  \
AREA:e#FF0000FF:"Free Slots":STACK \
GPRINT:e:LAST:"          Current\:%8.0lf"  \
GPRINT:cdefi:AVERAGE:"Average(%%)\:%8.2lf %s\n"  \
LINE2:g#0000FFFF:"User Available Slots"  \
GPRINT:g:LAST:"Current\:%8.0lf"  \
GPRINT:cdefbb:AVERAGE:"Average(%%)\:%8.2lf %s"

The changes in resolution and fonts are caused by cacti, but what causes
the y-axis to reach beyond the upper boundary (causing unwanted unfilled
space in the graph)? Can I somehow reproduce the old behavior? Many thanks!

Christian




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

[Valter Douglas Lisbôa Jr.]

Looking at the graphs, I see the font size has changed by default when you
change the rrdtool. I don't use cacti but you can dig into it to change
the --font on rrdtool graph to fit your needs (man rrdgraph for more).

On Thursday 03 July 2008 06:35:54 Christian Boehme wrote:

> Hello,
>
> I use cacti as a front end to rrdtool. When upgrading the cacti server I
> noticed a change in the graphs which seems to be caused by the upgrade
> of rrdtool from 1.0.46 to 1.2.12. With the older rrdtool the graphs look
> like this:
>
> http://wwwuser.gwdg.de/~cboehme1/cacti_old.png
>
> This is how this graph is produced:
>
> /usr/bin/rrdtool graph - \
> --imgformat=PNG \
> --start=-86400 \
> --end=-300 \
> --title="LSF - gwdg-pcpar Availability" \
> --rigid \
> --base=1000 \
> --height=120 \
> --width=500 \
> --upper-limit=252 \
> --lower-limit=0 \
> --vertical-label="" \
> DEF:a="/var/lib/cacti/full_11.rrd":full:AVERAGE \
> DEF:b="/var/lib/cacti/full_11.rrd":resv:AVERAGE \
> DEF:c="/var/lib/cacti/full_11.rrd":free:AVERAGE \
> DEF:d="/var/lib/cacti/full_11.rrd":avail:AVERAGE \
> CDEF:cdefc=a,252,/,100,* \
> CDEF:cdeff=b,252,/,100,* \
> CDEF:cdefi=c,252,/,100,* \
> CDEF:cdefbb=d,252,/,100,* \
> AREA:a#FFF200:"Used Slots"  \
> GPRINT:a:LAST:"          Current\:%8.0lf"  \
> GPRINT:cdefc:AVERAGE:"Average(%%)\:%8.2lf %s\n"  \
> STACK:b#FF7D00:"Reserved Slots"  \
> GPRINT:b:LAST:"      Current\:%8.0lf"  \
> GPRINT:cdeff:AVERAGE:"Average(%%)\:%8.2lf %s\n"  \
> STACK:c#FF0000:"Free Slots"  \
> GPRINT:c:LAST:"          Current\:%8.0lf"  \
> GPRINT:cdefi:AVERAGE:"Average(%%)\:%8.2lf %s\n"  \
> LINE2:d#0000FF:"User Available Slots"  \
> GPRINT:d:LAST:"Current\:%8.0lf"  \
> GPRINT:cdefbb:AVERAGE:"Average(%%)\:%8.2lf %s"
>
> With the newer rrdtool it looks like this:
>
> http://wwwuser.gwdg.de/~cboehme1/cacti_new.png
>
> This is the rrdtool command.
>
> /usr/bin/rrdtool graph - \
> --imgformat=PNG \
> --start=-86400 \
> --end=-300 \
> --title="LSF - gwdg-pcpar Availability" \
> --rigid \
> --base=1000 \
> --height=120 \
> --width=500 \
> --upper-limit=252 \
> --lower-limit=0 \
> --vertical-label="" \
> --slope-mode \
> --font TITLE:12: \
> --font AXIS:8: \
> --font LEGEND:10: \
> --font UNIT:8: \
> DEF:a="/srv/www/cacti-0.8.7b/rra/full_11.rrd":full:AVERAGE \
> DEF:b="/srv/www/cacti-0.8.7b/rra/full_11.rrd":full:MAX \
> DEF:c="/srv/www/cacti-0.8.7b/rra/full_11.rrd":resv:AVERAGE \
> DEF:d="/srv/www/cacti-0.8.7b/rra/full_11.rrd":resv:MAX \
> DEF:e="/srv/www/cacti-0.8.7b/rra/full_11.rrd":free:AVERAGE \
> DEF:f="/srv/www/cacti-0.8.7b/rra/full_11.rrd":free:MAX \
> DEF:g="/srv/www/cacti-0.8.7b/rra/full_11.rrd":avail:AVERAGE \
> DEF:h="/srv/www/cacti-0.8.7b/rra/full_11.rrd":avail:MAX \
> CDEF:cdefc=a,252,/,100,* \
> CDEF:cdeff=c,252,/,100,* \
> CDEF:cdefi=e,252,/,100,* \
> CDEF:cdefbb=g,252,/,100,* \
> AREA:a#FFF200FF:"Used Slots"  \
> GPRINT:a:LAST:"          Current\:%8.0lf"  \
> GPRINT:cdefc:AVERAGE:"Average(%%)\:%8.2lf %s\n"  \
> AREA:c#FF7D00FF:"Reserved Slots":STACK \
> GPRINT:c:LAST:"      Current\:%8.0lf"  \
> GPRINT:cdeff:AVERAGE:"Average(%%)\:%8.2lf %s\n"  \
> AREA:e#FF0000FF:"Free Slots":STACK \
> GPRINT:e:LAST:"          Current\:%8.0lf"  \
> GPRINT:cdefi:AVERAGE:"Average(%%)\:%8.2lf %s\n"  \
> LINE2:g#0000FFFF:"User Available Slots"  \
> GPRINT:g:LAST:"Current\:%8.0lf"  \
> GPRINT:cdefbb:AVERAGE:"Average(%%)\:%8.2lf %s"
>
> The changes in resolution and fonts are caused by cacti, but what causes
> the y-axis to reach beyond the upper boundary (causing unwanted unfilled
> space in the graph)? Can I somehow reproduce the old behavior? Many thanks!
>
> Christian

--
Valter Douglas Lisbôa Jr.
Sócio-Diretor
Trenix - IT Solutions
"Nossas Idéias, suas Soluções!"
www.trenix.com.br
contato@...
Tel. [phone]
Cel. +55 19 9183.4244

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

[Rene Fritzsche]

Hi,

> I use cacti as a front end to rrdtool. When upgrading the cacti server I
> noticed a change in the graphs which seems to be caused by the upgrade
> of rrdtool from 1.0.46 to 1.2.12. With the older rrdtool the graphs look
> like this:
>
> http://wwwuser.gwdg.de/~cboehme1/cacti_old.png
[...]
> With the newer rrdtool it looks like this:
>
> http://wwwuser.gwdg.de/~cboehme1/cacti_new.png
[...]
>
> The changes in resolution and fonts are caused by cacti, but what causes
> the y-axis to reach beyond the upper boundary (causing unwanted unfilled
> space in the graph)? Can I somehow reproduce the old behavior? Many thanks!

what amazes me more is the fact, that not only the drawing itself
changed (like the extra-space at the top, which seems to be produced
by that little arrow ontop of the y-axis) but the average values too.
is there some reason, why these calculations lead to different values
although its the same formular?

Greets,
Rene.

____________________________________________________________________
Ihre Messenger, Communities und E-Mails jetzt in einem Programm!
WEB.DE MultiMessenger http://www.produkte.web.de/messenger/?did=3071

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

[Valter Douglas Lisbôa Jr.]

Sorry man I read better your problem now. Is not the font, its the fit of
graph. woldn't --rigid be supposed to correct this? My graphs in 1.2.x has
the same behavior, it only fit exactly when I use HREF and --rigid together.

> > The changes in resolution and fonts are caused by cacti, but what causes
> > the y-axis to reach beyond the upper boundary (causing unwanted unfilled
> > space in the graph)? Can I somehow reproduce the old behavior? Many
> > thanks!
> >
> > Christian



--
Valter Douglas Lisbôa Jr.
Sócio-Diretor
Trenix - IT Solutions
"Nossas Idéias, suas Soluções!"
www.trenix.com.br
contato@...
Tel. [phone]
Cel. +55 19 9183.4244

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

[Christian Boehme]

Hello,

Rene Fritzsche schrieb:
> what amazes me more is the fact, that not only the drawing itself
> changed (like the extra-space at the top, which seems to be produced
> by that little arrow ontop of the y-axis) but the average values too.
> is there some reason, why these calculations lead to different values
> although its the same formular?

Well, the change in averages is due to the two graphs being exported at
slightly different times, i.e., the data is not completely identical. I
though no one would notice :). However, the extra space can be even
somewhat larger in other graphs, so I do not think it is caused by the
y-arrow.

Valter Douglas Lisbôa Jr. schrieb:
> Sorry man I read better your problem now. Is not the font, its the fit of
> graph. woldn't --rigid be supposed to correct this? My graphs in 1.2.x has
> the same behavior, it only fit exactly when I use HREF and --rigid together.

I used rigid for both graphs, so that does not help. What does HREF do?
I do not find it in the man pages, but it rather seems to be related to
text?

Any other ideas? Thank you

Christian



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

[Valter Douglas Lisbôa Jr.]

Sorry again , bad typo (Many clients calling me together and a major internet
provider backbone is down in my country, 8-0 ).

I mean HRULE. When I used it the graph adjust exactly, with --upper-limit
and --rigid It does not. I don't botter to much with this issue, so I use the
graphs in this state. But its curious the 1.2.x have thats "upper margin" in
the y axis, like thats does not appear shows any errors, I think it's a
aesthetic difference.

On Thursday 03 July 2008 11:29:42 Christian Boehme wrote:

> Hello,
>
> Rene Fritzsche schrieb:
> > what amazes me more is the fact, that not only the drawing itself
> > changed (like the extra-space at the top, which seems to be produced
> > by that little arrow ontop of the y-axis) but the average values too.
> > is there some reason, why these calculations lead to different values
> > although its the same formular?
>
> Well, the change in averages is due to the two graphs being exported at
> slightly different times, i.e., the data is not completely identical. I
> though no one would notice :). However, the extra space can be even
> somewhat larger in other graphs, so I do not think it is caused by the
> y-arrow.
>
> Valter Douglas Lisbôa Jr. schrieb:
> > Sorry man I read better your problem now. Is not the font, its the fit of
> > graph. woldn't --rigid be supposed to correct this? My graphs in 1.2.x
> > has the same behavior, it only fit exactly when I use HREF and --rigid
> > together.
>
> I used rigid for both graphs, so that does not help. What does HREF do?
> I do not find it in the man pages, but it rather seems to be related to
> text?
>
> Any other ideas? Thank you
>
> Christian

--
Valter Douglas Lisbôa Jr.
Sócio-Diretor
Trenix - IT Solutions
"Nossas Idéias, suas Soluções!"
www.trenix.com.br
contato@...
Tel. [phone]
Cel. +55 19 9183.4244

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

[Christian Boehme]

Valter Douglas Lisbôa Jr. schrieb:
> I mean HRULE. When I used it the graph adjust exactly, with --upper-limit
> and --rigid It does not. I don't botter to much with this issue, so I use the

This might work, but you cannot set an HRULE value in the template to be
used as a per graph value, so in my case - I have quite a few graphs of
this template - it would be very cumbersome.

> graphs in this state. But its curious the 1.2.x have thats "upper margin" in
> the y axis, like thats does not appear shows any errors, I think it's a
> aesthetic difference.

Well, in a stacked graph it is very nice to immediately see that it is
not filled to the maximum value. In the old graphs the free space would
indicate just this. In the new graphs you have to look closely. IMHO
this is an annoying difference. Thanks for your answers, though.

Christian


--
Dr. Christian Boehme
email: Christian.Boehme@...
phone: [phone]
---------------------------------------
Gesellschaft für wissenschaftliche
Datenverarbeitung mbH Göttingen (GWDG)
Am Fassberg 11, 37077 Göttingen
URL:    http://www.gwdg.de
E-Mail: gwdg@...
Tel.:   [phone]
Fax:    [phone]
Geschäftsführer:
Prof. Dr. Bernhard Neumair
Aufsichtsratsvorsitzender:
Prof. Dr. Christian Griesinger
Sitz der Gesellschaft: Göttingen
Registergericht: Göttingen
Handelsregister-Nr. B 598
---------------------------------------



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