Stange Graphs Items + graphs generation problem

[gronono]

Hi all,

Please help me...

I am using Cacti to generate AP Association graphs based on a CISCO 4400 MIB.
I noticed yesterday that some of these association graphs have strange Y scaling Items (1000m, 800m...).

Why I obtain 800 "milli-associations", I supposed it is due to the small number of users associated on these Access Points but shouldn't I get integers items (0,1,2,..) ?
Moreover, I really don't understand since the most part of AP assoc graphs were generated by the same way and are correct.
So, in order to find a solution to correct these strange items, I navigated on cacti and Cacti Forums and finally I modified AP Assoc graphs template and data templates parameters but unfortunately it didn't change anything about these items...(any solutions ?) Worst, I noticed that even if I rolled back parameters as initialy, CACTI does not want to continue AP association graphs generation.
Any idea to help me recover graphs generation ?

I also noticed that if I recreate graphs, the new graphs works correctly but I lose all my previous data and as I am using CACTI in production environnement, I rather like to find another way.

I precise that I am replacing someone who installed CACTI and I am a real CACTI newby so don't hesitate to detail your replies :O)

Thx

[gandalf]

Hi all,

Please help me...

I am using Cacti to generate AP Association graphs based on a CISCO 4400 MIB.
I noticed yesterday that some of these association graphs have strange Y scaling Items (1000m, 800m...).

Why I obtain 800 "milli-associations", I supposed it is due to the small number of users associated on these Access Points but shouldn't I get integers items (0,1,2,..) ?

This is a feature of rrdtool. See e.g. http://docs.cacti.net/manual:087:8_rrdtool#rrdtool for a better understanding.

Any idea to help me recover graphs generation ?

Please see 2nd link of my sig for debugging

I also noticed that if I recreate graphs, the new graphs works correctly but I lose all my previous data and as I am using CACTI in production environnement, I rather like to find another way.

That's the result when deleting rrd files.
R.