Opsview Knowledge Center

Service Checks, Service Groups and Host Templates

Overview of Service Checks within Opsview Monitor

Introduction

This section provides an overview of Service Checks, Service Groups and Host templates, how they interact, how to add and configure them, and how to analyze the Service Check data returned. This document will explain to readers the concept of service groups, Service Checks and Host templates, how the three interact, how to create them, how to apply them and finally how to analyze the gathered data.

After reading the User Guide, you should be able to create your own Service Checks, add them to Service Groups or Host templates and then apply the gathered data within the Host Groups, Hosts and Services section.

Opsview Monitor uses the concept of Hosts to define a device, server, website, database or more. As covered in Host Groups & Hosts Overview, Hosts are defined as:

' an autonomous computing device, such as (but not limited to), a server, virtual server, a slave server, database server, workstation, PC, network device, storage device, sensor, tablet or mobile device.' Essentially a host is logical endpoint, meaning a host could be your VMware host, your Oracle database, a Cisco switch or more. It is very flexible.'

Within Section 4.3, we covered the monitoring of Hosts using host check commands' i.e. checking to see if a Host is UP, DOWN or UNREACHABLE.

This is good practice and provides an insight into the status of the Hosts in your environment, i.e. all my servers are responding, but what if you want to know information about the Hosts ' i.e. how busy is the Apache server, how much disk space does my database server have, etc. This additional information is known as services, and as such a User must use a Service Checkto gather the data and display it within Opsview Monitor's analytical views.

Service Checks can be one of four types:

The first two, SNMP Polling and Plugins, are 'active checks'. The latter two are referred to as 'passive checks'. The difference between the two is that active checks are configured to go to the Host (from Opsview Monitor) and ask for information at predetermined, regular intervals ' whereas passive checks are configured in a way that Opsview Monitor will do nothing unless the monitored host or service sends data into it.

Active check

Active checks mean that the monitoring of a Service Check has been initiated by Opsview Monitor; this means that a check interval has been hit (i.e. five minutes has elapsed) and the Opsview Monitor software uses the determined plugin and arguments OR SNMP OID to retrieve data from the host. The data is then stored, analyzed and actioned if need be (notification sent, status changed, etc).

Passive Check

With passive checks such as SNMP traps or a Passive Check (i.e. NSCA), Opsview Monitor does not have a check interval which it uses to determine when to regularly monitor a Host. This puts the onus on the monitored Host to tell Opsview Monitor when a problem has occurred, i.e. 'An interface has gone down' or 'I'm running out of disk space'. On receiving these messages, Opsview Monitor will store them and then action them again ' changing a status, sending Notifications and more.

The main benefit of a passive check is that it removes the constant 'Opsview: Are you ok? 'Host: Yes i'm fine', 'Opsview: How about now?' 'Host:Yes i'm fine' messages.

However, Passive Checks can be tricky to configure and as passive checks are normally monitored purely on a status level and not a performance level ' in terms of there is no historical data to analyze as the metrics are not being gathered at a regular interval, i.e. with active checks, we are asking the host every 30 seconds 'how busy are you?' and it is providing us load average metrics. With a passive check, we will receive no data until the host tells us 'Hey, i'm really busy' ' this means that we cannot use graphs, historical analysis etc on that Host, unless the Host is sending the passive results regularly (much like an active check.)

To summarize, Service Checks allow Opsview Monitor to go deeper than 'Are you up?' monitoring, and allows you the ability to get real world metrics on their Hosts, such as load, active requests, network throughput and more ' data that can be used in various ways throughout Opsview Monitor. In the coming sections, we will cover the different types of Service Checks, how they work and how to configure them.

Service Checks, Service Groups and Host Templates

Overview of Service Checks within Opsview Monitor