check_snmp_netint.pl + pnp4nagios produce weird output

[klajosh2]

Hi,

I am using check_snmp_netint.pl plug-in to qurey interfaces on various devices. (mainly cisco).

And on a cisco 6509 switch I see graphs with many gaps but in contract graphs from a cisco 4506 switch are ok.
the parameters of the check_snmp_netint.pl plug-in are the same in both cases.

Code: Select all

command_line    $USER1$/check_snmp_netint.pl -H $HOSTADDRESS$ -2 -C $USER11$ -f -k -q -g -B -M -w 800,800,10,10,10,10 -c 950,950,15,15,15,15 -d 180 -Y -F /var/nagios/check_snmp_netint/tmp.int.$HOSTADDRESS$.$SERVICEDESC$ -P --label -o 65535 -n "$ARG1$"

Plugin outputs are look normal also:
GigabitEthernet1/1:UP (in=1.0Mbps/out=6.2Mbps/errors-in=0.0/errors-out=0.0/discard-in=0.0/discard-out=0.0):(1 UP): OK --- graphs are bad
FastEthernet0/1:UP (in=8.3Mbps/out=2.2Mbps/errors-in=0.0/errors-out=0.0/discard-in=0.0/discard-out=0.0):(1 UP): OK --- graphs are good

I attach both graphs (bad and good).

Please advise,

Klajosh

[sreinhardt]

Generally this is going to be due to gaps in data within the rrd, most likely due to data not being returned from the plugin\device. What interval are you checking this device at?

[klajosh2]

Hi,

thanks for getting back.

I am checking the interfaces in every 3 minutes.
So data should get back in every three minutes.
The heartbeat of the RRD files are 8460.

Here is the first part of the rrdtool info output of the rrd file what produces wrong output:

Code: Select all

filename = "ge1-1.rrd"
rrd_version = "0003"
step = 60
last_update = 1390832510
ds[1].type = "COUNTER"
ds[1].minimal_heartbeat = 8460
ds[1].min = 0.0000000000e+00
ds[1].max = NaN
ds[1].last_ds = "5426021"
ds[1].value = 4.9836527778e+04
ds[1].unknown_sec = 0
ds[2].type = "COUNTER"
ds[2].minimal_heartbeat = 8460
ds[2].min = 0.0000000000e+00
ds[2].max = NaN
ds[2].last_ds = "6027619"
ds[2].value = 5.9644896556e+08
ds[2].unknown_sec = 0
ds[3].type = "COUNTER"
ds[3].minimal_heartbeat = 8460
ds[3].min = NaN
ds[3].max = NaN
ds[3].last_ds = "81861863507"
ds[3].value = 3.2205547778e+07
ds[3].unknown_sec = 0
rra[0].cf = "AVERAGE"
rra[0].rows = 2880
rra[0].cur_row = 2663
rra[0].pdp_per_row = 1
rra[0].xff = 5.0000000000e-01
rra[0].cdp_prep[0].value = NaN
rra[0].cdp_prep[0].unknown_datapoints = 0
rra[0].cdp_prep[1].value = NaN
rra[0].cdp_prep[1].unknown_datapoints = 0
rra[0].cdp_prep[2].value = NaN
rra[0].cdp_prep[2].unknown_datapoints = 0

[klajosh2]

Hi,

additional info: it hast to be related to the pnp4nagios templates. I haunted down a check_snmp_netint pnp4nagios template
from the web and I started to use it. For couriosity I removed the template and delted the old rrd graphs so pnp4nagios has
to regenerate them and with default template I can see the correct values.
Can somebody help me how templates and rrd databases related ? I question this because even if I delete the template only
the graphs were wrong.
Another question how can I setup the default rrd DS type in pnp4nagios? Because interface measurement I would like to see COUNTER
type rather than GAUGE or DERIVE. I know there are more related to pnp4nagios but maybe somebody can help me here also.

Thanks,

klajosh2

[abrist]

Templates are just used for displaying the data, so if removing the template did not fix the problem, but deleting the rrd did, I would assume that the number or type of sources changed at some point, stopping the rrd from being able to save more data. A breakdown of templates and rrds can be found:
http://docs.pnp4nagios.org/pnp-0.6/tpl