Unexpected behaviour of dashboard's surveillance view

Description

The surveillance view shows an unexpected behaviour. Even though there are some services down and the corresponding table cell is colored red, it shows 0 of XX services down (see screenshot).

I'm using the following configuration for the surveillance view "default":

<view name="default" refresh-seconds="300" >
<rows>
<row-def label="IBM1" >
<category name="IBM1"/>
</row-def>
<row-def label="IBM2" >
<category name="IBM2" />
</row-def>
<row-def label="Dienste" >
<category name="dienste" />
</row-def>
</rows>
<columns>
<column-def label="Status" >
<category name="everything" />
</column-def>
</columns>
</view>

As our infrastructure is not clearly divided in production, testing and development, I changed the view a bit. Every node automatically gets the Surveillance Category "everything" assigned.

Environment

OpenNMS Horizon 15.0.1 OpenJDK Runtime Environment (1.7.0_75-b13) Debian 7.8 / Linux 3.2.0-4-amd64 (amd64)

Acceptance / Success Criteria

None

Attachments

1

is duplicated by

Lucidchart Diagrams

Activity

Show:

David Hustace November 2, 2015 at 4:17 PM

Since this seems to be a custom page, it's really impossible for us to reproduce. It also appears to me to be a point of confusion with all the components shown on this single page. The surveillance view is showing that there are 0 of 10 nodes down meeting the category set of IBM1 ∩ Everything and the Node status which shows that there are 2 of 2 services down on a node. I'm not sure why that doesn't indicate a "node down" but seems a different issue than what is presented here.

I hope this clarification helps with understanding the differences between the SV and the Node Status view.

Ronny Trommer July 9, 2015 at 3:03 AM

Is it possible to verify if this issue is fixed with

Cannot Reproduce

Details

Assignee

Reporter

Labels

Components

Affects versions

Priority

PagerDuty

Created April 16, 2015 at 4:41 AM
Updated June 23, 2016 at 3:49 PM
Resolved November 2, 2015 at 4:17 PM