How to enable Host Numeric Server Info monitoring in the vmware probe

Products

DX Unified Infrastructure Management (Nimsoft / UIM) Unified Infrastructure Management for Mainframe CA Unified Infrastructure Management SaaS (Nimsoft / UIM)

Issue/Introduction

There is a set of information made available by vCenter through the WebServices API that the vmware probe does not present by default. The collection of information is known as HostNumericServerInfo, or sometimes NumericServerInfo in the literature. This includes information about server fans, running temperatures, power supplies, etc. UIM does not present this for monitoring by default. There is a configuration hand-edit that can enable the data for monitoring. This document will explain how to enable these monitors. However, there are some reasons the monitors are disabled by default. That discussion will be considered first.

The HostNumericServerInfo content in the vmware Managed Object Browser (MOB) is found behind a link labeled ‘runtime’. All labeling beyond this link is subject to relabeling at any time by the vmware server. The issue this causes for the probe is mostly related to alarms. It is quite possible for alarms to be issued by the probe for which the naming content changes unexpectedly. When this happens, the Clear alarm is formed with mismatching naming. The original alarm is now orphaned, and must be cleared by hand.

One other issue is that NumericSensorInfo monitors are only supported in the Auto-Configuration mode. 'Static' is not supported for these elements.

Environment

Component: vmware

Cause

- Host Numeric Sensor info not supported via the Admin Console

Resolution

To enable the NumericServerInfo monitors, manually edit the mondef.cfg file on the probe robot host or edit the probe_schema.cfg on older versionf of vmware. (As always, make a backup first.) Search down for ‘using = NumericSensorInfo’. You should find three occurrences of this string. They will be associated with three configuration blocks: <UnitModifier>, <HealthState>, and <CurrentReading>. Three lines below each, find a line like ‘active = no’. Set this line to ‘active = yes’, save the file, and restart the probe.

The new monitors will be found at the individual Host level of the tree. The actual monitors available are configurable in the vmware interfaces. What is actually presented is outside UIM control. Depending on this external configuration, the list may be hundreds of lines long. That list is outside the scope of this document. You can shrink the list by enabling only one or two of the ‘active =’ specifiers.

As mentioned before, these new monitors are not supported as Static monitors. Do not enable them at the Host level. Instead, add them to a Template and deploy that Template to the Auto Configuration node, as described in the probe documentation.

QOS: QOS_VMWARE_HOST_NUMERIC_SENSOR_INFO

Additional Information

IMPORTANT: Host Numeric Server Info monitoring is ONLY available when using the Infrastructure Manager (IM) and is NOT available/accessible via the web Admin Console.

https://techdocs.broadcom.com/us/en/ca-enterprise-software/it-operations-management/ca-unified-infrastructure-management-probes/GA/monitoring/clouds-containers-and-virtualization/vmware-vmware-monitoring/vmware-metrics.html

In larger environments which commonly use the vmware probe for storage monitoring of hundreds or thousands of ESX Hosts, it is NOT recommended to use IM, as the Admin Console scales and performs much better than the Infrastructure Manager for the vmware probe. Normally the vmware probe needs very large min and max memory setting values just to operate in a somewhat acceptable manner. It can get bogged down easily in IM, so we highly recommend using the Admin Console as the preferred method of vmware probe administration and configuration.

In the Admin Console, the Template Builder and ability to copy templates from one environment to another, e.g., DEV to Production, is a great advantage.

vmware configuration using IM versus AC

In the vmware raw configuration, if you set enable_bulk_configuration = false then you use IM to configure the vmware probe.

If you set enable_bulk_configuration to a value of pobc, it defaults to using the web-based Admin Console.

Using the Admin Console (AC) - is not compatible with configuration through the Infrastructure Manager interface. Do not mix or interchange configuration methods! If you do, the result will be unpredictable monitoring of your VMware hosts and/or VMs and/or corruption of your vmware configuration.

When you enable bulk configuration, Infrastructure Manager is disabled and the Template Editor appears in the Admin Console GUI.

Once you enable bulk configuration, there is no supported process for going back to manual configuration. Be sure that you fully understand and accept the consequences of enabling bulk configuration before enabling it.

vmware FAQS

https://knowledge.broadcom.com/external/article/189579/vmware-frequently-asked-questions-faqs.html