Cacti is not Graphing 10 GigE Links

[maxclark]

The poller debug is below - everything looks fine and the other ethernet and cpu poller does graph properly.

02/13/2006 06:28:01 PM - POLLER: Poller[0] CACTI2RRD: /usr/local/bin/rrdtool update /usr/local/cacti-0.8.6h/rra/bbr01_traffic_in_362.rrd --template traffic_in:traffic_out 1139855259:531867192950523:925473438483294
OK u:0.01 s:0.01 r:20.92
02/13/2006 06:28:01 PM - POLLER: Poller[0] CACTI2RRD: /usr/local/bin/rrdtool update /usr/local/cacti-0.8.6h/rra/bbr01_5min_cpu_361.rrd --template 5min_cpu 1139855259:21
OK u:0.01 s:0.01 r:20.92
02/13/2006 06:28:01 PM - POLLER: Poller[0] CACTI2RRD: /usr/local/bin/rrdtool update /usr/local/cacti-0.8.6h/rra/bbr01_traffic_in_329.rrd --template traffic_in:traffic_out 1139855259:3069924812:3231124154
OK u:0.01 s:0.01 r:20.92

[gandalf]

Please have a look at the Data Source, see MAXIMUM value. Perhaps this is wrong? Then "rrdtool update" won't update the files. To check, please run

Code: Select all

rrdtool fetch <rrd file for 10GE> AVERAGE

and look at the last few lines (=most recent updates)
Reinhard

[maxclark]

These are the last two lines:

1139858100: NaN NaN
1139858400: NaN NaN

The device is configured SNMP v2 and the graph was created as a 64 bit counter. Do I need to change something here?

[gandalf]

These are the last two lines:

1139858100: NaN NaN
1139858400: NaN NaN

The device is configured SNMP v2 and the graph was created as a 64 bit counter. Do I need to change something here?

Would need about last 10 lines (this depends on the time difference to the last poller update). If all is NaN, this makes sure that updating is not really performed.
This may be due to a given MAXIMUM number for the data sources. To get the actual maximum, perform a

Code: Select all

rrdtool info <rrd file>

and look at the first few (20 or so) lines:

Code: Select all

ds[a].min = NaN
ds[a].max = NaN

NaN would be "no limit". If limit is set too low, updating is performed but no data is filled in! To change this, use rrdtool tune
Reinhard

[maxclark]

Thanks - this did the trick.

[ealcaniz]

I got the same

1366289100: nan
1366289400: nan
1366289700: nan
1366290000: nan
1366290300: nan
1366290600: nan
1366290900: nan
1366291200: nan
1366291500: nan
1366291800: nan
1366292100: nan
1366292400: nan
1366292700: nan
1366293000: nan
1366293300: nan
1366293600: nan
1366293900: nan
1366294200: nan
1366294500: nan
1366294800: nan


[root@ui-sc4-nagios1 rra]# rrdtool tune ia-pro-app-att1__cpu_13166.rrd
DS[cpu] typ: GAUGE hbt: 600 min: 0.0000 max: 100.0000

any ideas? happens with a lot of host, I incremented Maximum Concurrent Poller Processes from 1 to 3 and Maximum Threads per Process from 1 to 5. I have a cacti deployed with 404 host.

some servers doesn't get graphics well and I get NAN.

[gandalf]

[root@ui-sc4-nagios1 rra]# rrdtool tune ia-pro-app-att1__cpu_13166.rrd
DS[cpu] typ: GAUGE hbt: 600 min: 0.0000 max: 100.0000

I don't understand this bit.
404 hosts are not too much, by the way.
You might find additional help at 2nd link of my sig
R.