<img height="1" width="1" style="display:none;" alt="" src="https://px.ads.linkedin.com/collect/?pid=1110556&amp;fmt=gif">
July 28, 2022

How to Scale with DX UIM’s Monitoring Configuration Service, Part 2: Key Concepts

by: Steve D'Arcy

To contend with their escalating, intensifying demands, today’s operations teams must constantly be on the quest to boost efficiency. In my prior post, I offered a high-level introduction to DX Unified Infrastructure Management (DX UIM) Monitoring Configuration Service (MCS), outlining how its key features can significantly streamline administration in large-scale enterprise environments.

In this follow up post, I’ll provide more details for teams looking to start working with MCS. Specifically, I’ll offer some information on the four key elements within an MCS implementation: groups, templates, profiles, and alarm policies. The following sections offer more details on each of these elements.

Groups

In MCS, groups represent a superset of elements, including configurations and probes. Following are a few important concepts to be aware of:

  • Groups versus super packages. For users familiar with earlier versions of the product, including CA UIM or Nimsoft, MCS groups are similar to “super packages.” Groups and super packages are both a collection of configurations and probes that can be distributed from a single place. Unlike a super package, however, an MCS group will remove the configuration from a device if the device is removed from the group.
  • Grouping. Grouping is done by attributes. Teams can choose from a drop-down list of available attributes or even execute a SQL query to group devices. In addition, teams can create their own attributes, which can be used to group devices.
  • Prioritization. Within group profiles, users can assign a priority level. If a device belongs to multiple groups, the highest priority profile will be deployed.

Dynamic and Static Groups

When creating a new group, users can choose whether it is static or dynamic. In effect, the behavior and functionality of both types of groups is the same. The key difference is the way in which devices are added:

  • Static. With static groups, if a new device comes online, a user can manually add that device to a group, or they can use API-driven logic to make the change. Once a device is added, monitoring will then be applied according to policies established.
  • Dynamic. With dynamic groups, updates happen automatically. If a device is added to a group, policies will automatically be added and monitoring will be deployed.

For both group types, if a probe is not already present, MCS will deploy it automatically. If a device is removed from either group type, that group’s monitoring policy will also be removed.

Templates

In DX UIM implementations, probes are controlled by configuration files. A template is a specific section of a configuration file. Configuration files are always made up of multiple templates. For example, the configuration file for the CDM probe will include specific templates for CPU, disk, and memory.

Every probe has a specific template for set up, which includes the establishment of log levels, log sizes, and more. With one template, teams can create hundreds or thousands of distinct profiles. Some templates restrict the number of profiles that can be created, for example, on a per-probe or per-device basis. In establishing templates, it is important to ensure that whenever possible, existing profiles are being updated, rather than having multiple copies created.

Following are some key elements to consider in working with templates.

Access Control

In most implementations, teams will want to restrict access to some specific templates. For example, administrators may want to ensure that only administrators will be able to access setup templates. Teams can associate templates with access control lists (ACLs), so users can only access the template they’re authorized to use.

Remote and Local Templates

Templates can either be saved locally or used remotely. Local templates require that robots and probes are installed. In most deployments, around 90% of templates are saved locally. Teams can employ remote templates with net_connect and RSP to do agentless monitoring.

Legacy and Enhanced Templates

Template lists often have specific templates that are doubled, with one being the legacy template and one that is enhanced. Here’s how they differ:

  • Legacy. Legacy templates give complete control to the probe, which does everything, including metrics, thresholds, and alarms.
  • Enhanced. Enhanced templates operate differently. Once alarm policies are enabled, the enhanced template controls metrics only.

To provide some background, the way all probes work is that they pass metrics and alarms to the spooler, which has a connection to the hub and message bus. With enhanced templates, spoolers play a central role. With these templates, intelligence is added to the spooler, which will look up the configuration before publishing to the message bus.

All enhanced template profiles are written to the plugin_metric.cfg file. These templates then turn the metrics on at the probe. This is a key aspect to how enhanced templates now work.

While metrics may be gathered via the probe, the spooler determines what gets published to the message bus and what doesn’t. For example, while 20 different metrics may be captured, the enhanced template configuration may indicate that only seven metrics should be passed to the message bus.

Once an enhanced template is created, a default alarm policy is also generated that is provided directly to the spooler (plugin_metric.cfg). When the spooler receives a metric, the plugin_metric.cfg alarm policies will determine whether to issue an alarm. The spooler can do this for all probes.

Getting Started

To get started with templates, users must select a group or device. In the navigation bar, users click the gear icon, then go to profiles, and then click the templates button. The list of templates available will be unique to each site. Templates come in a downloadable package, and can be installed in MCS, however, the approach usually depends on the probes being used.

Profiles

When administrators or users define monitoring settings for MCS, they do so in what are known as “profiles.” Profiles can be created at the device or group level. Profiles can be pushed to all devices within a group that meet the criteria defined in the template.

Alarm Policies

Alarm policies can span multiple probes and groups. Policies can be employed at other levels, including conditions, devices, or monitoring technology. Policies can include different groups, such as a MySQL and Unix group. As part of app discovery, MCS will see if a device falls into either group, and apply alarm policies accordingly.

Teams can flexibly create and manage policies. Teams can apply multiple thresholds to a policy, modify thresholds, specify whether a group is static or dynamic, and more. When specified conditions are met, alarm messages can be generated immediately. In addition, teams can specify alarm settings, so, for example, alarm messages will only be delivered if three alarms arise in a 15-minute span.

When enhanced templates are built, MCS creates default alarm policies automatically. However, administrators can choose whether to use those default alarm policies. They can disable or delete those policies, or instruct MCS not to create policies at all. Teams can also use these existing policies as a fall back, and set at a low priority, so some alarm policies are always implemented, even if no other higher-priority policies are established.

When creating a new alarm policy, teams must set as a higher priority than the default to ensure it is executed.

Conclusion

By employing the advanced capabilities of MCS, teams can establish a solid foundation that enables intelligent monitoring across dynamic, large-scale environments. To learn more, be sure to view our MCS demo video, which offers a detailed look at many of the steps involved in working with this feature.

Explore More Posts

View All Blog Posts
August 8, 2022

Innovate at Speed With DX NetOps 22.2 Network Monitoring Software

DX NetOps 22.2 represents the most scalable and comprehensive solution to unify monitoring of multi-vendor SD-WAN, providing end-to-end network visibility. Read Now
July 29, 2022

GigaOm, Again Names Broadcom Leader in Radar Report for Network Observability, 2022

Our latest innovations for network observability and digital experience monitoring have once again been recognized as a leader in their category by GigaOm.Our latest... Read Now
July 29, 2022

General Availability: Experience-Driven NetOps, DX NetOps 22.2

DX NetOps 22.2 continues to deliver our industry-leading visibility, scale and modern network coverage...now beyond the network edge. Read Now
July 28, 2022

How to Scale with DX UIM’s Monitoring Configuration Service, Part 2: Key Concepts

In Part 2, we explore the four key elements within an MCS implementation: groups, templates, profiles, and alarm policies. Read Now
July 27, 2022

Hybrid Network Triage for the New Enterprise Network

With Experience-Driven NetOps, it's now possible to have user experience, active testing, and network path analytics for any managed or unmanaged network. Read Now
July 26, 2022

Optimizing Security and Digital Experiences: Why User Experience Monitoring is Key

For any enterprise operating in today’s digital world, navigating the tension between security and user experience is an increasingly difficult task. Read Now
July 25, 2022

Operationalizing Experience in the New Enterprise Network

The DX NetOps 22.2 release will enable teams to operationalize the new enterprise network, focus on user experience, and avoid chasing utilization spikes. Read Now
July 21, 2022

How to Scale with DX UIM’s Monitoring Configuration Service, Part 1: Introduction

Monitoring is a task that keeps getting more vital and more difficult. To respond, operations teams must keep finding ways to boost their operational efficiency. Read Now
July 20, 2022

More Than Half of Organizations Prioritize Security Over User Experience

Network visibility demands a holistic approach to balance user experience with security. Read Now
July 7, 2022

Expert Series: Broadcom IT Shares Their View on the Difference Between Monitoring and Monitoring Correctly

Here, Broadcom IT's Kathy Solomon shares how implementing DX UIM has allowed them to not just monitor high value servers, but monitor them correctly. Read Now