need explanation on graphs

[bibi23]

Hi,

I want to create graph, but i don't understand all functionnality.
is someone could explain me the difference between all, avg, min, max selecteur for simple graph ?
What is an aggregated graph ?

i don't find any info in the documentation.

Thanks

[nelsonab]

Bump, I've been looking around on this too. I tried looking in the docs and reading the source but I'm drawing blanks about an aggregated graph's purpose and use.

[nelsonab]

Bump.. I hate to be a butt on this, but I'd really like to understand the aggregated graphs better.

For instance the following is a graph I created for processor load. I'm still kinda scratching my head to understand what it's showing me. I know this may seem like a silly question to some, but I know I'm not the only one who's scratched my head on this one, and the documentation is almost non-existant and overall there's little description about this graph type in the forum.

[rolandsym]

trying to understand myself

I got the below from http://www.jffnms.org/docs/polling.html which is a network management system. But it does not sync up with how zabbix interprets the data. Looks like cacti looks at it the same way as jffnms.

Could we get more an explanation in the documentation? It's empty on the graphs section.

-----------------------------------------
Aggregate Graphs

An aggregate graph is one that is able to show simultaneously data from multiple graphs of the same Interface Type. The filename for the aggregate is like the standard graph, except it has _aggregate in it. For example if the graph filename is traffic.inc.php, the aggregate filename will be traffic_aggregate.inc.php

The function arguments are similar to that of the standard graph function except that the interfaces are in an array with an index of the interface ID. The array contains the standard Interface Parameters (see table * )


Sincerely,
Rolandsym

[bbrendon]

I won't get too involved but ...

The line in the top graph is the average. The top and bottom areas are min and max.

The reason you see the shading sometimes is the way you have these set for the item:
"Keep history (in days)"
"Keep trends (in days)"

Once it has been longer than history, then the data turns into a trend. A trend is not as accurate and is summarized which is why the two graphs are different.

[nelsonab]

Ok I get that, but I'm still not quite understanding what the graph is aggregating.

Is it aggregating n number of periods of the graph? Example if the graph is 1 hour, then it will aggregate the last n hours? Here's a strange one from my development box, it's over 7 days. Again the top is the aggregated one the bottom is the real values (simple graph). Why do I show aggregated data when I don't have real data in that time area and vice versa? I also show a similar situation on a 24 hour graph.

[Alexei]

Ok I get that, but I'm still not quite understanding what the graph is aggregating.

When selecting a long period, a single pixel (X axis) of the graph represents 10 minutes, maybe 2 days, depending on the period. The graphs displays min, max, avg value for the 10 minutes (2 days, whatever).

[nelsonab]

Alexi, thanks for replying. I'm glad to see some good discussion about this little understood feature.

Ok so as I understand it you are saying that the "aggregation window" or the period for which data aggregation is done is determined by the graph period. The smallest the window can be though is 10 minutes. Correct?

Would the last graph I presented be correct and expected? I ask this because the aggregation shows up in areas where there is no physical data. The data gap was from when the development VMWare image was shut down.

[rue]

riddle solved?

Ok I get that, but I'm still not quite understanding what the graph is aggregating.

Is it aggregating n number of periods of the graph? Example if the graph is 1 hour, then it will aggregate the last n hours? Here's a strange one from my development box, it's over 7 days. Again the top is the aggregated one the bottom is the real values (simple graph). Why do I show aggregated data when I don't have real data in that time area and vice versa? I also show a similar situation on a 24 hour graph.

Just wondered about the same question. What exactly does type "aggregated" for an item in a graph mean?

It seems to be a means to compare values of the current period (displayed in the graph) with min/avg/max from overlapping periods before (and excluding) the current period displayed (which explains, why you see aggregated data although your agent didn't run at all... ).

It only makes sense if you want to compare the current data with periods before that or if you want to get an, well, aggregated overview of the data over the last eg. 5 (default) days or weeks.

Would be nice to change the xtics in the graph to reflect the function... .

Just for reference.

Cheers

Rü

[tdomhan]

When selecting a long period, a single pixel (X axis) of the graph represents 10 minutes, maybe 2 days, depending on the period. The graphs displays min, max, avg value for the 10 minutes (2 days, whatever).

how does that apply to a short period(say 1hour), or does the selected function (avg/min/max) not influence the graph in a short period of time