GPRINTS differ from graph [solved]

[sepp]

Hi community,

I am stucked again... I created a graph showing the dsp-usage in our net. The "funny" part is that the graph seems to have correct values, where the GPRINTS tell really wrong things -- except the "last" value which seems to be correct. I cannot see the reason why this happens. Here is the output from the "little-tool-button":

/usr/bin/rrdtool graph - \
--imgformat=PNG \
--start=1211189315 \
--end=1211275715 \
--title="Overview of DSP usage 1" \
--base=1000 \
--height=350 \
--width=850 \
--alt-autoscale-max \
--lower-limit=0 \
COMMENT:"From 2008/05/19 11\:28\:35 To 2008/05/20 11\:28\:35\c" \
COMMENT:" \n" \
--vertical-label="" \
--slope-mode \
--font TITLE:12: \
--font AXIS:8: \
--font LEGEND:10: \
--font UNIT:8: \
DEF:a="/var/lib/cacti/rra/xxxvgx1_###_net_dsp_used_453.rrd":dsp_used:LAST \
DEF:b="/var/lib/cacti/rra/yyy-vgx-1_###_net_dsp_used_1165.rrd":dsp_used:LAST \
DEF:c="/var/lib/cacti/rra/xxx-vgx-2_dsp_used_1143.rrd":dsp_used:LAST \
DEF:d="/var/lib/cacti/rra/xxx-vgx-3_###_net_dsp_used_1146.rrd":dsp_used:LAST \
DEF:e="/var/lib/cacti/rra/xxx-vgx-4_###_net_dsp_used_1160.rrd":dsp_used:LAST \
DEF:f="/var/lib/cacti/rra/xxx-vgx-7_dsp_used_1131.rrd":dsp_used:LAST \
DEF:g="/var/lib/cacti/rra/xxx-vgx-5_###_net_dsp_used_1152.rrd":dsp_used:LAST \
DEF:h="/var/lib/cacti/rra/xxx-vgx-6_dsp_used_1137.rrd":dsp_used:LAST \
DEF:i="/var/lib/cacti/rra/yyy-vgx-3_###_net_dsp_used_1169.rrd":dsp_used:LAST \
DEF:j="/var/lib/cacti/rra/yyy-vgx-1_###_net_dsp_used_1165.rrd":dsp_used:MIN \
DEF:ba="/var/lib/cacti/rra/yyy-vgx-1_###_net_dsp_used_1165.rrd":dsp_used:MAX \
CDEF:cdefbc=TIME,1211275464,GT,a,a,UN,0,a,IF,IF,TIME,1211275464,GT,c,c,UN,0,c,IF,IF,TIME,1211275464,GT,d,d,UN,0,d,IF,IF,TIME,1211275464,
GT,e,e,UN,0,e,IF,IF,TIME,1211275464,GT,f,f,UN,0,f,IF,IF,TIME,1211275464,GT,g,g,UN,0,g,IF,IF,TIME,1211275464,GT,h,h,UN,0,h,IF,IF,TIME,1211275464,
GT,b,b,UN,0,b,IF,IF,TIME,1211275464,GT,i,i,UN,0,i,IF,IF,+,+,+,+,+,+,+,+ \
CDEF:cdefbd=TIME,1211275464,GT,a,a,UN,0,a,IF,IF,TIME,1211275464,GT,c,c,UN,0,c,IF,IF,TIME,1211275464,GT,d,d,UN,0,d,IF,IF,TIME,1211275464,
GT,e,e,UN,0,e,IF,IF,TIME,1211275464,GT,f,f,UN,0,f,IF,IF,TIME,1211275464,GT,g,g,UN,0,g,IF,IF,TIME,1211275464,GT,h,h,UN,0,h,IF,IF,TIME,1211275464,
GT,b,b,UN,0,b,IF,IF,TIME,1211275464,GT,i,i,UN,0,i,IF,IF,+,+,+,+,+,+,+,+ \
CDEF:cdefbe=TIME,1211275464,GT,j,j,UN,0,j,IF,IF \
CDEF:cdefbf=TIME,1211275464,GT,ba,ba,UN,0,ba,IF,IF \
LINE2:a#FF7D00FF:"xxx-VGX-1 AS5400" \
GPRINTLAST:"active calls int1%8.0lf\n" \
LINE2:c#FF00FFFF:"xxx-VGX-2" \
GPRINTLAST:" active calls int2%8.0lf\n" \
LINE2:d#F51D30FF:"xxx-VGX-3\n" \
LINE2:e#C4FD3DFF:"xxx-VGX-4\n" \
LINE2:f#00234BFF:"xxx-VGX-7\n" \
LINE2:g#AFECEDFF:"xxx-VGX-5\n" \
LINE2:h#4444FFFF:"xxx-VGX-6\n" \
LINE2:b#00FF00FF:"yyy-VGX-1\n" \
LINE2:i#55D6D3FF:"yyy-VGX-3\:\n" \
COMMENT:"total usage" \
AREA:cdefbc#8F005C33:"" \
GPRINT:cdefbd:LAST:" Last\:%8.0lf" \
GPRINT:cdefbe:MIN:"Min\:%8.0lf" \
GPRINT:cdefbf:MAX:"Max%8.0lf" \
GPRINT:cdefbf:AVERAGE:"Avg\:%8.0lf"

has anyone any Idea? I dont understand why vgx-1 and vgx-2 are showing the same numbers (this is definitely wrong) and why the summation for the totals only works for "last" value, but not for max etc. I used CDEF ALL_DATA_SOURCES_NODUPS

[sepp]

hm... I still didn't come any further with this. Has nobody experienced this before?

many thanks,

//seb

[gandalf]

has anyone any Idea? I dont understand why vgx-1 and vgx-2 are showing the same numbers (this is definitely wrong)

LINE2:a#FF7D00FF:"xxx-VGX-1 AS5400" \
GPRINTLAST:"active calls int1%8.0lf\n" \
LINE2:c#FF00FFFF:"xxx-VGX-2" \
GPRINTLAST:" active calls int2%8.0lf\n" \

are the statements to be taken into account. In both cases, you've used the same data source on the GPRINT statement. So the printout is the same. I suppse you wanted to use DEF=a for the first one and DEF=c for the second one. Please pay attention when assigning data sources to a Graph Template Item!

and why the summation for the totals only works for "last" value, but not for max etc. I used CDEF ALL_DATA_SOURCES_NODUPS

You're sure it's ALL_DATA_SOURCES_NODUPS and not SIMILAR_DATA_SOURCES_NODUPS? Did you assign a data source to the GPRINT item referring to that CDEF?
Reinhard

[sepp]

Hi Reinhard,

many many thanks for your hints. I see the point why the data is the same, but I actually dont get around to tell cacti to use def a for the first GPRINT. I tried several ways to do this in the graph template, but no success. Regardless what I do, it always takes b for both gprints.

I am quite sure about the total_data issue, please see attached pics. I don't know why it works for the "last" value, but not for the others.

thanks!

//seb

[gandalf]

When totalling, please take the "consolidation function" used into account. When using LAST as in your exmaple, it will total all LAST data sources. This may not be what you want.
Reinhard

[sepp]

hi,

I "fixed" that while using my own cdef instead of "total all data sources". this cdef is obviously not working for me

thanks!