how to use/display INTEGER values, not AVERAGE

Post general support questions here that do not specifically fall into the Linux or Windows categories.

Moderators: Developers, Moderators

Post Reply
fdessart
Posts: 4
Joined: Tue Oct 08, 2002 7:29 am

how to use/display INTEGER values, not AVERAGE

Post by fdessart »

Hello,

I use cacti to monitor number of connected user on a Cisco NAS.

I think Cacti polls SNMP OID every 5 minutes.
On the graph, I can see an average for this data, not the real number of connected users. So I have a decimal in my data, which is not very serious for an integer value. How can I do to display the entire value and not the average?

Thanks for your help.

Francois.
raX
Lead Developer
Posts: 2243
Joined: Sat Oct 13, 2001 7:00 pm
Location: Carlisle, PA
Contact:

Post by raX »

To get exact integers, select a graph item of a graph you want to make this change to. Then select 'Exact Numbers' under 'GPRINT Options'. This should force rrdtool to display an exact integer instead of a decimal.

-Ian
fdessart
Posts: 4
Joined: Tue Oct 08, 2002 7:29 am

Post by fdessart »

Thanx raX.

Ok, it works for the GPRINT values.

However I think now my problem is somewhere else.
The MAX value for 5min, 30min, 2hour, 1day averages are different.
The RRD file aggregates only average values, I would like to have the instantaneous value, got at the time of SNMP polling. So my graph would display a sequence of instantaneous values. In other words, I would like to disable the "averaging" function.

Is it possible?

Thanks.

Francois.
raX
Lead Developer
Posts: 2243
Joined: Sat Oct 13, 2001 7:00 pm
Location: Carlisle, PA
Contact:

Post by raX »

I think the best way to get rid of the averaging altogether is to go into 'Round Robin Archives' and unselected AVERAGE for each one. The only downside to this is you will have delete your rrd files and let cacti re-create them. Also while you're making that change, make sure to select the LAST CF as well.

I am not absolutely sure if you have to unselect these CF's from each data source as well, but you can experiment with that if needed.

-Ian
Post Reply

Who is online

Users browsing this forum: No registered users and 2 guests