Nov 22, 2009 5:39 AM
Nagios-like "Alert Histogram" - how to get it?
-
Like (0)
Does Zenoss provide a built-in way to graphing alerts by time? Something similar to Nagios "alert histogram" feature - number of alerts per time of day or day of week.
Thanks in advance,
Vitaly
Thank you! But it's not exactly what I'm looking for.
I want similar graph but not per collector; I need it per device/service/group....
There is a feature I would LOVE to see added to Zenoss.
I accomplished a limited version of this by creating an external data source (shell script attached). I then created a new Device template called NumEvents. My Data Source command template is simply:
/home/zenoss/query.sh -H $devname
The Data Point is a GUAGE, and I had to use a custom Create Cmd:
RRA:AVERAGE:0.5:1:600
RRA:AVERAGE:0.5:6:600
RRA:AVERAGE:0.5:24:600
RRA:AVERAGE:0.5:288:600
RRA:MAX:0.5:1:600
RRA:MAX:0.5:6:600
RRA:MAX:0.5:24:600
RRA:MAX:0.5:288:600
RRA:LAST:0.5:1:600
RRA:LAST:0.5:6:600
RRA:LAST:0.5:24:600
RRA:LAST:0.5:288:600
When graphing, I use the MAX data point, and changed the data value to: '%.lf' (This gets rid of all the decimal place crap)
This gives me a new PERF graph of the number of events every 5 minutes:
+1, yes please post on the Wiki.
--
James Pulver
Information Technology Area Supervisor
LEPP Computer Group
Cornell University
Ryan Matte wrote, On 11/24/2009 3:44 PM:
Very nice, you should post that on the Wiki section of the site.
+1 - Yes, very nice workaround!!
But need to be adapted to the different Linux-Distributions, because the path to mysql is hard-coded.
For example for the rPath-Virtual-Appliance in the line:
COUNT='/usr/local/zenoss/mysql/bin/mysql --skip-column-names ...
it must be change to
COUNT='mysql --skip-column-names ...
Another Tip, if you copy the file "query.sh" over windows or open it with a Windows-Editor you should remove LF/CR with this command as zenoss-user:
mv query.sh query.bak; tr -d '\r' < query.bak > query.sh
make it executable with:
chmod 754 query.sh
Then it works well
Regards
dbuck
Nice idea!! Shouldn't be too hard really...individual data points for each event severity catagory, and a stacked multi-graph report...
Yeah...
nelliot, thank you again for for script! I added such graph to my basic "device" template and created a milti-graph report for all my production servers.
In the same time I have to say that it'll be nice if Zenoss will provide more flexible report tool/wizard, which will allow us to create events reports based on devices, systems/groups, severity of events and so on.
My first attempt at a ZenPack, this build has a stacked graph that is color coded by event severity: docs/DOC-4538
(Edit: This version is an "All Events" graph showing the total number of events, not broken down by specific device)
Follow Us On Twitter »
|
Latest from the Zenoss Blog » | Community | Products | Services Resources | Customers Partners | About Us | ||
Copyright © 2005-2011 Zenoss, Inc.
|
||||||||