XIO Targets Test

A target is a physical port located on the storage controller of the XtremIO Storage Array.

The XtremIO Storage Array supports the following target types:

  • iSCSI - a 10GbE NIC port for connecting to iSCSI networks. There are two iSCSI targets per Storage Controller.
  • FC - an FC HBA port for connecting to fiber optic cable networks. There are two FC targets per Storage Controller.

The cluster targets form the XtremIO Storage Array’s front-end to which application servers connect for receiving storage services. The I/O requests from the host/server are primarily received through these targets, and the responses from the storage array are communicated to the host/server through these targets. The targets are the primary handlers of the I/O request processing in the storage array. By periodically checking the target port status and measuring the I/O load on the ports, you can identify overloaded ports, and thus proactively detect potential/existing load-balancing irregularities and/or processing bottlenecks with the storage controller. The XIO Targets test facilitates this port check. For every target port configured on the storage controller, this test reports the port state, the I/O load on the ports, the processing ability of the ports, and the errors encountered by each target port. In the process, the test not only points administrators to overloaded ports, but also puts a finger on ports that are slow when processing I/O requests and the ports that are erroneous.

Target of the test : An EMC XtremIO Storage array

Agent deploying the test : A remote agent

Outputs of the test : One set of results for each X-Brick:Storage Controller:target port of the target EMC XtremIO being monitored .

Configurable parameters for the test
Parameter Description

Test Period

How often should the test be executed.

Host

The IP address of the storage device for which this test is to be configured.

Port

The port number at which the storage array listens. The default is NULL.

XtremIO User and XtremIO Password

Provide the credentials of a user who has read only privileges to access the XtremIO storage array in the XtremIO User and XtremIO Password text boxes.

Confirm Password

Confirm the password by retyping it here.

XMS IP

This parameter is applicable only for EMC XtremIO 4.x. By default, None will be chosen from this list. If the target EMC XtremIO storage array is within a XMS Management Server that is auto-discovered, then the IP or host name of that XMS Management Server will be displayed in this list. Select that particular XMS IP to configure this test. If you wish to monitor an EMC XtremIO Storage Array that is either not an integral part of the auto-discovered XMS Management Server or a brand new EMC XtremIO Storage Array, choose the Other option. This will enable you to add a new XMS Managament Server. To know how to add a new XMS Management Server, refer to Adding a new XMS.

SSL

The eG agent collects performance metrics by invoking Restful APIs on the target Storage array. Typically, the Restful APIs can be invoked through the HTTP or the HTTPS mode. By default, the eG agent invokes the Restful APIs using the HTTPS mode. This is why, the SSL flag is set to Yes by default. If the target storage array is not SSL-enabled, then the Restful APIs can be accessed through the HTTP mode only. In this case, set the SSL flag to No.

Detailed Diagnosis

To make diagnosis more efficient and accurate, the eG Enterprise suite embeds an optional detailed diagnostic capability. With this capability, the eG agents can be configured to run detailed, more elaborate tests as and when specific problems are detected. To enable the detailed diagnosis capability of this test for a particular server, choose the On option. To disable the capability, click on the Off option.

The option to selectively enable/disable the detailed diagnosis capability will be available only if the following conditions are fulfilled:

  • The eG manager license should allow the detailed diagnosis capability
  • Both the normal and abnormal frequencies configured for the detailed diagnosis measures should not be 0.
Measurements made by the test
Measurement Description Measurement Unit Interpretation

Port state

Indicates the current state of this target port.

 

The values reported by this measure and their numeric equivalents are available in the table below:

Measure Value Numeric Value
Up 0
Down 1

Note:

This measure reports the Measure Values listed in the table above to indicate the current state of this target. However, in the graph, this measure is indicated using the Numeric Values listed in the above table.

Target health

Indicates the current health of this target port.

 

The values reported by this measure and their numeric equivalents are available in the table below:

Measure Value Numeric Value
Level 1 clear 0
Unknown 1

Note:

This measure reports the Measure Values listed in the table above to indicate the health of this target. However, in the graph, this measure is indicated using the Numeric Values listed in the above table.

Bandwidth

Indicates the amount of data utilized for performing I/O operations per second through this target port.

MB/sec

 

Read bandwidth

Indicates the amount of data utilized for performing read I/O operations per second through this target port.

MB/Sec

By comparing the values of these measures across the target ports, you can isolate those target ports that consume bandwidth excessively, and also understand where exactly was too much of bandwidth consumed - while reading? or writing?

Write bandwidth

Indicates the amount of data utilized for performing write I/O operations were performed per second through this target port.

MB/Sec

IOPS

Indicates the total number of reads and writes through this target port per second.

IOPS

Compare the value of this measure across the target ports to know the target port that handled the maximum number of I/O operations and the port that handled the least. By constantly analyzing these values, administrators can keep a check on the load balancing issues across the target ports.

Read IOPS

Indicates the number of read I/O operations performed through this target port per second.

IOPS

 

Write IOPS

Indicates the number of write I/O operations performed through this target port per second.

IOPS

 

Reads

Indicates the number of reads per second through this target port during the last measurement period.

Reads/sec

Compare the value of this measure across target ports to identify the port that is handling the maximum number of read requests and the port that is handling the least.

Writes

Indicates the number of writes handled by this target port per second during the last measurement period.

Writes/sec

Compare the value of this measure across target ports to identify the port that is handling the maximum number of write requests and the port that is handling the least.

Data reads

Indicates the rate at which data was read through this target port during the last measurement period.

MB/Sec

Compare the value of these measures across the target ports to identify the slowest target port in terms of servicing read and write requests respectively.

Data written

Indicates the rate at which data was written through this target port during the last measurement period.

MB/Sec

Average read size

Indicates the average amount of data read through this target port per I/O operation during the last measurement period.

MB/Op

 

Average write size

Indicates the average amount of data written through this target port per I/O operation during the last measurement period.

MB/Op

 

Loss of signals

Indicates the number of times the signal was lost on this target port during the last measurement period.

Number

Ideally, the value of this measure should be zero. A non-zero value for this measure indicates that the port detected a loss of the electrical or optical signal used to transfer data on the port.

This is likely an indicator for a faulty connector or cable. These are also caused when the device connected to the port is restarted, replaced or being serviced when the Fiber Channel cable connected to the port is temporarily disconnected. 

If the port is in the “loss of signal” state for longer than a specific period, the port will get into the link failure state which could degrade the performance of the Fiber Channel link.

This measure is applicable only when the target port is a FC port.

Rate of Signal loss

Indicates number of signals lost on this target port per second during the last measurement period.

Signals/sec

A low value is desired for this measure.

This measure is applicable only when the target port is a FC port.

Loss of Syncs

Indicates the number of times this target port failed to synchronize during the last measurement period.

Number

Ideally, the value of this measure should be zero. A non-zero value for this measure indicates that port went into the “loss of synchronization” state, where it encountered continuous Disparity errors.

This is likely an indicator for a faulty connector or cable. These are also caused when the device connected to the port is restarted, replaced or being serviced when the Fiber Channel cable connected to the port is temporarily disconnected. 

If the port is in the “loss of synchronization” state for longer than a specific period, the port will get into the link failure state which could degrade the performance of the Fiber Channel link.

This measure is applicable only when the target port is a FC port.

Loss syncs rate

Indicates the number of times this target port failed to synchronize per second during the last measurement period.

Sync/sec

Ideally, the value of this measure should be zero.

This measure is applicable only when the target port is a FC port.

Invalid CRCs

Indicates the number of invalid CRCs that occurred on this target port during the last measurement period.

Number

This refers to the number of frames handled by this target that contains checksum errors. Ideally, the value of this measure should be zero.

These are usually recoverable errors and will not degrade system performance unless their occurrence is sustained when the data cannot be relayed after retransmissions.

This measure is applicable only when the target port is a FC port.

Rate of invalid CRCs

Indicates the rate at which invalid CRCs occurred on this target port during the last measurement period.

CRC/sec

Ideally, the value of this measure should be low. A high value for this measure indicates poor health of the target port.

This measure is applicable only when the target port is a FC port.

Primitive sequence protocol errors

Indicates the number of Primitive Sequence protocol errors that occurred on this target port during the last measurement period.

Number

Ideally, the value of this measure should be zero.

This measure is applicable only when the target port is a FC port.

Rate of PSP errors

Indicates the number of Primitive Sequence protocol errors occurred per second on this target port during the last measurement period.

Errors/sec

Ideally, the value of this measure should be zero.

 

This measure is applicable only when the target port is a FC port.

Link failures

Indicates the number of link failures experienced by this target during the last measurement period.

Number

Ideally, the value of this measure should be zero. A non-zero value indicates that Fiber Channel connectivity with this target was “broken” that many times. This is likely an indicator for a faulty connector or cable. These are also caused when the device connected to this target is restarted, replaced or being serviced when the Fiber Channel cable connected to this target is temporarily disconnected.

These measures are applicable only when the target port is a FC port.

Rate of link failures

Indicates the number of link failures experienced per second by this target during the last measurement period.

Failures/sec

Dumped frames

Indicates the number of frames dumped by this target port due to lack of buffer credit during the last measurement period.

Percent

Buffer credits, also called buffer-to-buffer credits (BBC) are used as a flow control method by Fiber Channel technology and represent the number of frames a port can store.

Each time a port transmits a frame that port's BB Credit is decremented by one; for each R_RDY received, that port's BB Credit is incremented by one. Transmission of an R_RDY indicates that the port has processed a frame, freed a receive buffer, and is ready for one more. If the BB Credit is zero, the corresponding node cannot transmit until an R_RDY is received back. A high value for this measure therefore indicates that an R_RDY was not received by the FC port for a long time. This is a cause for concern, as until the R_RDY is received, the FC port will not resume communication. 

The solution for this problem is to allocate optimal buffer credits to the FC port. The optimal number of buffer credits is determined by the distance (frame delivery time), the processing time at the receiving port, the link signaling rate, and the size of the frames being transmitted. As the link speed increases, the frame delivery time is reduced and the number of buffer credits must be increased to obtain full link utilization, even in a short-distance environment. Smaller frame sizes need more buffer credits.

This measure is applicable only when the target port is a FC port.