eG Monitoring
 

Measures reported by IBFabricElemTest

A switched fabric is a point-to-point switch-based interconnect designed for fault tolerance and scalability, and is configured, verified and managed by an element manager in the network environment. A point-to-point switch fabric element ensures that every link has exactly one device connected at each end of the link. Thus, the loading and termination operations are well controlled and, with only one device allowed, peak I/O performance of the switch is guaranteed with fabric elements. The fabric elements provide scalability which can be accomplished by adding the target switch to the fabric and connecting more endnodes through the switch. The aggregate bandwidth of the switch increases as the target switch is additional switches are added to the network. Multiple paths between devices keep the aggregate bandwidth high and provide fail-safe, redundant connections. Since fabric elements provide both network connectivity and management capabilities for the InfiniBand Switch, an inoperable fabric element can halt the communication between the target switch and other end-nodes in the environment. To avoid this, administrators should continuously track the status of the fabric elements that are connected to the target switch. This is what exactly the IBFabricElemTest test does!.

This test auto-discovers the fabric elements connected to the target Infiniband switch, and for each fabric element, reports the current operational status and availability of the element manager.

Outputs of the test : One set of results for each fabric element connected to the InfiniBand Switch to be monitored.

The measures made by this test are as follows:

Measurement Description Measurement Unit Interpretation
fabricElemOperStatus Indicates the current operational status of this fabric element.   The values reported by this measure and its numeric equivalents are mentioned in the table below:

Measure Value Numeric Value
Unknown 1
No-attention 2
Attention 3
Note:

By default, this measure reports the Measure Values listed in the table above to indicate the operational status of this fabric element. The graph of this measure however, represents the status of a port using the numeric equivalents only.

fabricElemMgrAlive Indicates the whether/not the element manager is currently available to this fabric element.   The values reported by this measure and its numeric equivalents are mentioned in the table below:

Measure Value Numeric Value
Unknown 0
Available 1
Not available 2
Note:

By default, this measure reports the Measure Values listed in the table above to indicate the availability of element manager to this fabric element. The graph of this measure however, is represented using the numeric equivalents only.