I tried using the trends feature last week and today I had a look at the
numbers.
The type of graph I get (see screenshot) is useless to get the insight I
was looking for. The funniest thing is the fact I restarted all servers
yesterday which is shown as a series of small spikes due to the fact that
the 'uptime' shows a greater accuracy for small numbers.
I added a similar graph for available memory last week but that thing is
effectively inaccessible. Looking at the below graph I think this thing has
a 'line' for every measurement because each time a measurement is done the
number of available bytes is unique.
Did I do something wrong or is this a bug in the system?
We just discovered almost the same thing but then with swapmemory. It
looks like trends work best with discrete facts that don't have too many
steps.
If its supposed to work like this then a RFE might be to be able to
define steps or buckets for this kind of facts.
Regards,
Joop
ยทยทยท
On 9-4-2015 12:11, Niels Basjes wrote:
>
> Hi,
>
> I tried using the trends feature last week and today I had a look at
> the numbers.
>
> The type of graph I get (see screenshot) is useless to get the insight
> I was looking for. The funniest thing is the fact I restarted all
> servers yesterday which is shown as a series of small spikes due to
> the fact that the 'uptime' shows a greater accuracy for small numbers.
>
> I added a similar graph for available memory last week but that thing
> is effectively inaccessible. Looking at the below graph I think this
> thing has a 'line' for every measurement because each time a
> measurement is done the number of available bytes is unique.
>
> Did I do something wrong or is this a bug in the system?
>
>
