This is usual, each new version adds new mibs and metrics we collect, which slowly expands the footprint of the platform.

If you're having performance issues, I'd recommend disabling the fdb module first, if that data isn't critical for you, since its moderately heavy compared to the utility of the data.

Adam.


Sent from BlueMail

On 21 Oct 2016, at 07:37, William Bauer <bbauer@scripps.edu> wrote:
I upgraded our community edition on Monday from the previous release 
dated 5/4/2016 to the latest release.  I'm using observium to monitor 
the VM on which it is running, which is CentOS 6.

The context switches graph has gone from hovering around 3500 to about 
4200 immediately after the upgrade.  System interrupts have increased 
by about the same percentage, and System I/O and Load Average have 
bumped up slightly.

Are all of these typical of this upgrade?  Nothing else was done to 
the system or to the devices being monitored.  The increase in all 
these graphs (and a corresponding significant decrease in Observium 
performance) all correspond exactly in time to the upgrade.

I've been wondering why it's performing worse, and then I noticed 
these items.  Just looking for ideas or any helpful suggestions, and 
if this is to be expected.

Thanks.

Bill


observium mailing
list
observium@observium.org
http://postman.memetic.org/cgi-bin/mailman/listinfo/observium