eG Monitoring
 

Measures reported by LHStorageNodeTest

This test auto-discovers the storage nodes on the storage cluster and for each storage node, reports the current health, overall size, and the load-balancing capability of each storage node in the storage system. With the help of this test, administrators can not only identify overloaded nodes, but can also predict the potential failure of the node, so that efforts can be undertaken to avert the same. In addition, the test also points administrators to nodes that are handling more I/O requests than the rest, thus shedding light on irregularities in the distribution of I/O load across disks and prompting administrators to fine-tune the load-balancing algorithm.

Outputs of the test : One set of results for each storage node on the target storage cluster being monitored.

The measures made by this test are as follows:

Measurement Description Measurement Unit Interpretation
Status Indicates the current health of this storage node.   The values reported by this measure and their corresponding numeric equivalents are mentioned in the table below:

Measure Value Numeric Value
Up 1
Down 2
Note:

This measure reports the Measure Values listed in the table above to indicate the current health of this storage node. However, in the graph of this measure, the state is indicated using only the Numeric Values listed in the above table.
Condition Indicates the current storage condition of this storage node.   The values reported by this measure and their corresponding numeric equivalents are mentioned in the table below:

Measure Value Numeric Value
NotReady 1
Unoperable 2
Overloaded 3
Ready 4
Note:

This measure reports the Measure Values listed in the table above to indicate the current storage condition of this storage node. However, in the graph, the values of this measure are indicated using only the Numeric Values listed in the above table.
Total_size Indicates the total size of this storage node. GB  
Provisioned_size Indicates the space (allocated) provisioned for this storage node. GB  
Used_size Indicates the space that was already utilized in this storage node. GB A low value is desired for this measure. If the value of this measure is close to the Total space measure, then, it indicates that the storage node is running out of space. Administrators should therefore, free up the space in the storage node or allocate additional resources to the storage node.
Free_size Indicates the space that is currently available for use in this storage node. GB A high value is desired for this measure.
Total_space_util Indicates the space utilized in this storage node. Percent A low value is desired for this measure.
Allocated_space_used Indicates the percentage of space that is utilized from the provisioned space of this storage node. Percent  
Tot_free_space_pct Indicates the percentage of space that is available for use in this storage node. Percent  
Allocated_free_pct Indicates the percentage of space that is available for use from the provisioned space of this storage node. Percent  
IO_reads Indicates the rate at which read operations were performed on this storage node during the last measurement period. Operations/Sec Compare the value of this measure across storage nodes to know which node handled the maximum number of I/O read operations and which handled the least. If the gap between the two is very high, then it indicates serious irregularities in loadbalancing across storage nodes.
IO_writes Indicates the rate at which write operations were performed on this storage node during the last measurement period. Operations/Sec Compare the value of this measure across storage nodes to know which node handled the maximum number of I/O write operations and which handled the least. If the gap between the two is very high, then it indicates serious irregularities in loadbalancing across storage nodes.
Data_reads Indicates the rate at which data was read from this storage node during the last measurement period. MB/sec Compare the value of these measures across storage nodes to identify the slowest node in terms of servicing read and write requests (respectively).
Data_writes Indicates the rate at which data was written to this storage nod during the last measurement period. MB/sec
Pending_IO_operation Indicates the I/O operations that were pending on this storage node during the last measurement period. Operations/Sec A consistent increase in this value indicates a potential processing bottleneck with the storage node.
Read_latency Indicates the time taken to complete the read operations from this storage node during the last measurement period. Millisecs  
Write_latency Indicates the time taken to complete the write operations on this storage node during the last measurement period. Millisecs  
IO_latency Indicates the time taken to complete the read and write operations on this storage node during the last measurement period. Millisecs Ideally, this value should be low. If not, it implies that the storage node is slow.