Managing Network Adapters

This chapter includes the following sections:

Configuring Network Adapter Properties

Before you begin

  • The server must be powered on, or the properties will not display.

Procedure

Step 1

In the Navigation pane, click the Networking menu.

Step 2

In the Networking menu, click Adapter Card 1 or Adapter Card 2 or Adapter Card MLOM.

Step 3

In the Adapter Card Properties area, review the following information:

Name

Description

PCI Slot field

The PCI slot in which the adapter is installed.

Vendor field

The vendor for the adapter.

Product Name field

The product name for the adapter.

Product ID field

The product ID for the adapter.

Serial Number field

The serial number for the adapter.

Version ID field

The version ID for the adapter.

Hardware Revision field

The hardware revision for the adapter.

Cisco IMC Management Enabled field

If this field displays yes, then the adapter is functioning in Cisco Card Mode and passing Cisco IMC management traffic through to the server Cisco IMC.

Configuration Pending field

If this field displays yes, the adapter configuration has changed in Cisco IMC but these changes have not been communicated to the host operating system.

To activate the changes, an administrator must reboot the adapter.

ISCSI Boot Capable field

Whether iSCSI boot is supported on the adapter.

CDN Capable field

Whether CDN is supported on the adapter.

usNIC Capable field

Whether the adapter and the firmware running on the adapter support the usNIC.

Note 

usNIC support is not available for C125 servers.

Description field

A user-defined description for the adapter.

You can enter between 1 and 63 characters.

Enable FIP Mode check box

If checked, then FCoE Initialization Protocol (FIP) mode is enabled. FIP mode ensures that the adapter is compatible with current FCoE standards.

Note 

We recommend that you use this option only when explicitly directed to do so by a technical support representative.

Enable LLDP check box

Note 

For LLDP change to be effective, it is required that you reboot the server.

In case of S3260 chassis with two nodes, ensure to reboot the secondary node after making LLDP changes in the primary node.

If checked, then Link Layer Discovery Protocol (LLDP) enables all the Data Center Bridging Capability Exchange protocol (DCBX) functionality, which includes FCoE, priority based flow control.

By default, LLDP option is enabled.

Note 

We recommend that you do not disable LLDP option, as it disables all the DCBX functionality.

Note 

This option is available only on some UCS C-Series servers.

Enable VNTAG Mode check box

If VNTAG mode is enabled:

  • vNICs and vHBAs can be assigned to a specific channel.

  • vNICs and vHBAs can be associated to a port profile.

  • vNICs can fail over to another vNIC if there are communication problems.

Step 4

In the Firmware area, review the following information:

Name Description

Running Version field

The firmware version that is currently active.

Backup Version field

The alternate firmware version installed on the adapter, if any. The backup version is not currently running. To activate it, administrators can click Activate Firmware in the Actions area.

Note 

When you install new firmware on the adapter, any existing backup version is deleted and the new firmware becomes the backup version. You must manually activate the new firmware if you want the adapter to run the new version.

Startup Version field

The firmware version that will become active the next time the adapter is rebooted.

Bootloader Version field

The bootloader version associated with the adapter card.

Status field

The status of the last firmware activation that was performed on this adapter.

Note 

The status is reset each time the adapter is rebooted.

Step 5

In the External Ethernet Interfaces area, review the following information:

Note 

You may not see all the fields listed in the table on your screen. Click the icon on the right top corner and choose columns that you want to view.

Name Description

Port column

The uplink port ID.

Admin Speed column

The data transfer rate for the port. This can be one of the following:

  • 1 Gbps

  • 10 Gbps

  • 40 Gbps

  • 4 x 10 Gbps

  • Auto

  • 40 Gbps

  • 4 x 10 Gbps

Note 

You can edit the Admin Speed column. Select the port for which you want to edit the Admin Speed and click on the icon above the Port column. Click on save to save your changes. You need to choose 40 Gbps as the port speed if you are using a 40 Gbps switch.

Link Training column

Indicates if link training is enabled on the port.

MAC Address column

The MAC address of the uplink port.

Link State column

The current operational state of the uplink port. This can be one of the following:

  • Fault

  • Link Up

  • Link Down

  • SFP ID Error

  • SFP Not Installed

  • SFP Security Check Failed

  • Unsupported SFP

Encap column

The mode in which adapter operates. This can be one of the following:

  • CE—Classical Ethernet mode.

  • VNTAG—VNTAG mode.

Operating Speed column

The operating rate for the port. This can be one of the following:

  • 1 Gbps

  • 10 Gbps

  • 40 Gbps

  • 4 x 10 Gbps

  • Auto

  • 40 Gbps

  • 4 x 10 Gbps

Note 

You need to choose 40 Gbps as the port speed if you are using a 40 Gbps switch.

Connector Present column

Indicated whether or not the connector is present. This can be one of the following:

  • Yes—Connector is present.

  • No—Connector not present.

Note 

This option is only available for some adapter cards.

Connector Supported column

Indicates whether or not the connector is supported by Cisco. This can be one of the following:

  • Yes—The connector is supported by Cisco.

  • No—The connector is not supported by Cisco.

If the connector is not supported then the link will not be up.

Note 

This option is only available for some adapter cards.

Connector Type column

The type of the connector.

Note 

This option is only available for some adapter cards.

Connector Vendor column

The vendor for the connector.

Note 

This option is only available for some adapter cards.

Connector Part Number column

The part number of the connector.

Note 

This option is only available for some adapter cards.

Connector Part Revision column

The part revision number of the connector.

Note 

This option is only available for some adapter cards.

Viewing Storage Adapter Properties

Before you begin

  • The server must be powered on.

Procedure

Step 1

In the Navigation pane, click the Storage menu.

Step 2

In the Storage menu, click the appropriate LSI MegaRAID or HBA controller.

Step 3

In the Controller area, the Controller Info tab displays by default.

Step 4

In the Work pane's Health/Status area, review the following information:

Name

Description

Composite Health field

The combined health of the controller, the attached drives, and the battery backup unit. This can be one of the following:

  • Good

  • Moderate Fault

  • Severe Fault

  • N/A

Controller Status field

The current status of the controller. This can be one of the following:

  • Optimal — The controller is functioning properly.

  • Failed — The controller is not functioning.

  • Unresponsive — The controller is down.

RAID Chip Temperature field

Temperature of the controller in degree centigrade.

TTY Log Status field

The current status of the TTY log download. This can be one of the following:

  • Not Downloaded
  • In Progress
  • Complete
Step 5

In the Firmware Versions area, review the following information:

Name

Description

Product Name field

The name of the MegaRAID controller.

Serial Number field

The serial number of the MegaRAID controller.

Firmware Package Build field

The active firmware package version number.

For the firmware component version numbers, see the Running Firmware Images area.

Step 6

In the PCI Info area, review the following information:

Name

Description

PCI Slot field

The name of the PCIe slot in which the controller is located.

Vendor ID field

The PCI vendor ID, in hexadecimal.

Device ID field

The PCI device ID, in hexadecimal.

SubVendor ID field

The PCI subvendor ID, in hexadecimal.

SubDevice ID field

The PCI subdevice ID, in hexadecimal.

Step 7

In the Manufacturing Data area, review the following information:

Name

Description

Manufactured Date field

The date the MegaRAID card was manufactured, in the format yyy-mm-dd.

Revision No field

The board revision number, if any.

Step 8

In the Boot Drive area, review the following information:

Name

Description

Boot Drive field

The number of the boot drive.

Boot Drive is PD field

If this field displays true, the boot drive is a physical drive.

Step 9

In the Running Firmware Images area, review the following information:

Name

Description

BIOS Version field

The BIOS option PROM version number.

Firmware Version field

The active firmware version number.

Preboot CLI Version field

The pre-boot CLI version number.

WebBIOS Version field

The Web BIOS version number.

NVDATA Version field

The non-volatile data (NVDATA) version number.

Boot Block Version field

The boot block version number.

Boot Version field

The firmware boot loader version number on the LSI controller.

Step 10

In the Startup Firmware Images area, review the following information:

Name

Description

Startup BIOS Version field

The BIOS option PROM version that will become active when the host server reboots, if different from the current version.

Startup Firmware Version field

The firmware version that will become active when the host server reboots, if different from the current version.

Startup Preboot CLI Version field

The pre-boot CLI version that will become active when the host server reboots, if different from the current version.

Startup WebBIOS Version field

The Web BIOS version that will become active when the host server reboots, if different from the current version.

Startup NVDATA Version field

The non-volatile data version that will become active when the host server reboots, if different from the current version.

Startup Boot Block Version field

The boot block version that will become active when the host server reboots, if different from the current version.

Startup Boot Version field

The firmware boot loader version that will become active when the host server reboots, if different from the current version.

Step 11

In the Virtual Drive Count area, review the following information:

Name

Description

Virtual Drive Count field

The number of virtual drives configured on the controller.

Degraded Drive Count field

The number of virtual drives in a degraded state on the controller.

Offline Drive Count field

The number of virtual drives that have failed on the controller.

Step 12

In the Physical Drive Count area, review the following information:

Name

Description

Disk Present Count field

The number of physical drives present on the controller.

Degraded Disk Count field

The number of physical drives in a degraded state on the controller.

Failed Disk Count field

The number of physical drives that have failed on the controller.

Step 13

In the Settings area, review the following information:

Name

Description

Predictive Fail Poll Interval field

The number of seconds between predictive failure polls.

During each poll, the controller examines the Self-Monitoring Analysis and Reporting Technology (SMART) data on all physical drives to determine if any is about to fail.

Rebuild Rate field

The rate at which the controller rebuilds degraded RAID volumes.

This rate is shown as a percentage of the total bandwidth available.

Patrol Read Rate field

The rate at which the controller performs a background read of the physical drives looking for inconsistent data.

This rate is shown as a percentage of the total bandwidth available.

Consistency Check Rate field

The rate at which the controller scans the virtual drives looking for redundant data inconsistencies and fixing them.

This rate is shown as a percentage of the total bandwidth available.

Reconstruction Rate field

The rate at which virtual drives are reconstructed when the capacity or RAID level needs to be changed.

This rate is shown as a percentage of the total bandwidth available.

Cache Flush Interval field

The number of seconds waits before flushing the cache memory to the physical drives.

Max Drives To Spin Up At Once field

The number of drives that can be spun up simultaneously after the server is powered on.

Delay Among Spinup Groups field

The number of seconds to wait before the controller spins up the next set of drives.

Physical Drive Coercion Mode field

Whether the controller rounds the size of physical drives down to a round number. This can be one of the following:

  • None—The controller does not do any rounding.

  • 128 MB—Drive sizes are rounded down to the closest multiple of 128 MB.

  • 1GB—Drive sizes are rounded down to the closest multiple of 1GB.

Cluster Mode field

If this field displays true, the drives on this controller are shared with controllers on other servers.

Battery Warning field

If this field displays true, missing battery warnings are disabled.

ECC Bucket Leak Rate field

The error correcting code (ECC) single-bit error bucket leak rate, in minutes.

With ECC, the controller increments an error counter when it encounters a single bit error while reading from a physical drive. The controller decrements the error counter each time the number of minutes defined in this field passes.

If the error counter reaches a system-defined maximum, the controller sends an event message to the system.

Expose Enclosure Devices field

If this field displays true, enclosure devices are visible to the host drivers.

Maintain PD Fail History field

If this field displays true, the controller remembers which physical drives were determined to be bad across server reboots.

Enable Copyback on SMART field

If this field displays true, the controller copies the contents of the drive to a spare drive if Self-Monitoring Analysis and Reporting Technology (SMART) reports an error.

Enable Copyback to SSD on SMART Error field

If this field displays true, the controller copies the contents of an SSD card to a spare card if SMART reports an error.

Native Command Queuing field

If this field displays true, Native Command Queuing (NCQ) is disabled.

JBOD field

If this field displays true, JBOD is enabled.

Enable Spin Down of Unconfigured Drives field

If this field displays true, the controller spins down unconfigured drives.

Enable SSD Patrol Read field

If this field displays true, the controller performs patrol reads on SSD cards.

Auto Enhanced Import field

If this field displays true, foreign configurations are automatically imported when the controller boots.

Step 14

In the Capabilities area, review the following information:

Name

Description

RAID Levels Supported field

The RAID levels supported by the controller. This can be one or more of the following:

  • Raid 0—Simple striping.

  • Raid 1—Simple mirroring.

  • Raid 5—Striping with parity.

  • Raid 1E—Integrated offset strip mirroring

  • Raid 6—Striping with two parity drives.

  • Raid 10—Spanned mirroring.

  • Raid 50—Spanned striping with parity.

  • Raid 60—Spanned striping with two parity drives.

  • Raid srl-03—Spanned secondary RAID level

  • Raid 00—Spanned striping.

  • Raid 1e-rlq0—Integrated adjacent strip mirroring with no span.

  • Raid 1e0-rlq0—Integrated adjacent strip mirroring with span.

Step 15

In the HW Configuration area, review the following information:

Name

Description

SAS Address field

A MegaRAID controller can have up to 16 serial-attached SCSI (SAS) addresses. This field displays the first 8 SAS addresses, if they are in use.

BBU Present field

If this field displays true, the battery backup unit is present.

NVRAM Present field

If this field displays true, the NVRAM is present.

NVRAM Size field

The size of the NVRAM, in kilobytes.

Serial Debugger Present field

If this field displays true, a serial debugger is attached to the RAID card.

Memory Present field

If this field displays true, memory is present.

Flash Present field

If this field displays true, flash memory is present.

Flash Size field

The size of the flash memory, in megabytes.

Memory Size field

The size of the memory, in megabytes.

Cache Memory Size field

The size of the cache memory, in megabytes.

Number of Backend Ports field

The number of SATA or SAS ports on the controller.

Step 16

In the Error Counters area, review the following information:

Name

Description

Memory Correctable Errors field

The number of correctable errors in the controller memory.

Memory Uncorrectable Errors field

The number of uncorrectable errors in the controller memory.

Managing vHBAs

Guidelines for Managing vHBAs

When managing vHBAs, consider the following guidelines and restrictions:

  • The Cisco UCS Virtual Interface Cards provide two vHBAs and two vNICs by default. You can create up to 14 additional vHBAs or vNICs on these adapter cards.

    The Cisco UCS 1455 and 1457 Virtual Interface Cards, in non-port channel mode, provide four vHBAs and four vNICs by default. You can create up to 10 additional vHBAs or vNICs on these adapter cards.


    Note

    If Network Interface Virtualization (NIV) mode is enabled for the adapter, you must assign a channel number to a vHBA when you create it.

  • When using the Cisco UCS Virtual Interface Cards in an FCoE application, you must associate the vHBA with the FCoE VLAN. Follow the instructions in the Modifying vHBA Properties section to assign the VLAN.

  • After making configuration changes, you must reboot the host for settings to take effect.

Viewing vHBA Properties

Procedure

Step 1

In the Navigation pane, click the Networking menu.

Step 2

In the Adapter Card pane, click the vHBAs tab.

Step 3

In the vHBAs pane, click fc0 or fc1.

Step 4

In the General area of vHBA Properties, review the information in the following fields:

Name Description

Name field

The name of the virtual HBA.

This name cannot be changed after the vHBA has been created.

Target WWNN field

The WWNN associated with the vHBA.

To let the system generate the WWNN, select AUTO. To specify a WWNN, click the second radio button and enter the WWNN in the corresponding field.

Target WWPN field

The WWPN associated with the vHBA.

To let the system generate the WWPN, select AUTO. To specify a WWPN, click the second radio button and enter the WWPN in the corresponding field.

FC SAN Boot check box

If checked, the vHBA can be used to perform a SAN boot.

Enable Persistent LUN Binding check box

If checked, any LUN ID associations are retained in memory until they are manually cleared.

Uplink Port field

The uplink port associated with the vHBA.

Note 

This value cannot be changed for the system-defined vHBAs fc0 and fc1.

MAC Address field

The MAC address associated with the vHBA.

To let the system generate the MAC address, select AUTO. To specify an address, click the second radio button and enter the MAC address in the corresponding field.

Default VLAN field

If there is no default VLAN for this vHBA, click NONE. Otherwise, click the second radio button and enter a VLAN ID between 1 and 4094 in the field.

Class of Service drop-down list

The CoS for the vHBA.

Select an integer between 0 and 6, with 0 being lowest priority and 6 being the highest priority.

Note 

This option cannot be used in VNTAG mode.

Rate Limit field

The data rate limit for traffic on this vHBA, in Mbps.

If you want this vHBA to have an unlimited data rate, select OFF. Otherwise, click the second radio button and enter an integer between 1 and 10,000.

Note 

This option cannot be used in VNTAG mode.

PCIe Device Order field

The order in which this vHBA will be used.

To let the system set the order, select ANY. To specify an order, select the second radio button and enter an integer between 0 and 17.

EDTOV field

The error detect timeout value (EDTOV), which is the number of milliseconds to wait before the system assumes that an error has occurred.

Enter an integer between 1,000 and 100,000. The default is 2,000 milliseconds.

RATOV field

The resource allocation timeout value (RATOV), which is the number of milliseconds to wait before the system assumes that a resource cannot be properly allocated.

Enter an integer between 5,000 and 100,000. The default is 10,000 milliseconds.

Max Data Field Size field

The maximum size of the Fibre Channel frame payload bytes that the vHBA supports.

Enter an integer between 256 and 2112.

Channel Number field

The channel number that will be assigned to this vHBA.

Enter an integer between 1 and 1,000.

Note 

VNTAG mode is required for this option.

Port Profile drop-down list

The port profile that should be associated with the vHBA, if any.

This field displays the port profiles defined on the switch to which this server is connected.

Note 

VNTAG mode is required for this option.

Step 5

In the Error Recovery area, review the information in the following fields:

Name Description

Enable FCP Error Recovery check box

If checked, the system uses FCP Sequence Level Error Recovery protocol (FC-TAPE).

Link Down Timeout field

The number of milliseconds the uplink port should be offline before it informs the system that the uplink port is down and fabric connectivity has been lost.

Enter an integer between 0 and 240,000.

Port Down I/O Retries field

The number of times an I/O request to a port is returned because the port is busy before the system decides the port is unavailable.

Enter an integer between 0 and 255.

I/O Timeout Retry field

The time period till which the system waits for timeout before retrying. When a disk does not respond for I/O within the defined timeout period, the driver aborts the pending command, and resends the same I/O after the timer expires.

Enter an integer between 1 and 59.

Port Down Timeout field

The number of milliseconds a remote Fibre Channel port should be offline before informing the SCSI upper layer that the port is unavailable.

Enter an integer between 0 and 240,000.

Step 6

In the Fibre Channel Interrupt area, review the information in the following fields:

Name Description

Interrupt Mode drop-down list

The preferred driver interrupt mode. This can be one of the following:

  • MSIx—Message Signaled Interrupts (MSI) with the optional extension. This is the recommended option.

  • MSI—MSI only.

  • INTx—PCI INTx interrupts.

Step 7

In the Fibre Channel Port area, review the information in the following fields:

Name Description

I/O Throttle Count field

The number of I/O operations that can be pending in the vHBA at one time.

Enter an integer between 1 and 1,024.

LUNs per Target field

The maximum number of LUNs that the driver will export. This is usually an operating system platform limitation.

Enter an integer between 1 and 1,024. The recommended value is 1024.

LUN Queue Depth field

The number of commands that the HBA can send or receive in a single chunk per LUN. This parameter adjusts the initial queue depth for all LUNs on the adapter.

Default value is 20 for physical miniports and 250 for virtual miniports.

Step 8

In the Fibre Channel Port FLOGI area, review the information in the following fields:

Name Description

FLOGI Retries field

The number of times that the system tries to log in to the fabric after the first failure.

To specify an unlimited number of retries, select the INFINITE radio button. Otherwise select the second radio button and enter an integer into the corresponding field.

FLOGI Timeout field

The number of milliseconds that the system waits before it tries to log in again.

Enter an integer between 1,000 and 255,000.

Step 9

In the Fibre Channel Port PLOGI area, review the information in the following fields:

Name Description

PLOGI Retries field

The number of times that the system tries to log in to a port after the first failure.

Enter an integer between 0 and 255.

PLOGI Timeout field

The number of milliseconds that the system waits before it tries to log in again.

Enter an integer between 1,000 and 255,000.

Step 10

In the SCSI I/O area, review the information in the following fields:

Name Description

CDB Transmit Queue Count field

The number of SCSI I/O queue resources the system should allocate.

Enter an integer between 1 and 8.

CDB Transmit Queue Ring Size field

The number of descriptors in each SCSI I/O queue.

Enter an integer between 64 and 512.

Step 11

In the Receive/Transmit Queues area, review the information in the following fields:

Name Description

FC Work Queue Ring Size field

The number of descriptors in each transmit queue.

Enter an integer between 64 and 128.

FC Receive Queue Ring Size field

The number of descriptors in each receive queue.

Enter an integer between 64 and 128.

Modifying vHBA Properties

Procedure

Step 1

In the Navigation pane, click the Networking menu.

Step 2

In the Adapter Card pane, click the vHBAs tab.

Step 3

In the vHBAs pane, click fc0 or fc1.

Step 4

In the General area, update the following fields:

Name Description

Name field

The name of the virtual HBA.

This name cannot be changed after the vHBA has been created.

Target WWNN field

The WWNN associated with the vHBA.

To let the system generate the WWNN, select AUTO. To specify a WWNN, click the second radio button and enter the WWNN in the corresponding field.

Target WWPN field

The WWPN associated with the vHBA.

To let the system generate the WWPN, select AUTO. To specify a WWPN, click the second radio button and enter the WWPN in the corresponding field.

FC SAN Boot check box

If checked, the vHBA can be used to perform a SAN boot.

Enable Persistent LUN Binding check box

If checked, any LUN ID associations are retained in memory until they are manually cleared.

Uplink Port field

The uplink port associated with the vHBA.

Note 

This value cannot be changed for the system-defined vHBAs fc0 and fc1.

MAC Address field

The MAC address associated with the vHBA.

To let the system generate the MAC address, select AUTO. To specify an address, click the second radio button and enter the MAC address in the corresponding field.

Default VLAN field

If there is no default VLAN for this vHBA, click NONE. Otherwise, click the second radio button and enter a VLAN ID between 1 and 4094 in the field.

Class of Service drop-down list

The CoS for the vHBA.

Select an integer between 0 and 6, with 0 being lowest priority and 6 being the highest priority.

Note 

This option cannot be used in VNTAG mode.

Rate Limit field

The data rate limit for traffic on this vHBA, in Mbps.

If you want this vHBA to have an unlimited data rate, select OFF. Otherwise, click the second radio button and enter an integer between 1 and 10,000.

Note 

This option cannot be used in VNTAG mode.

PCIe Device Order field

The order in which this vHBA will be used.

To let the system set the order, select ANY. To specify an order, select the second radio button and enter an integer between 0 and 17.

EDTOV field

The error detect timeout value (EDTOV), which is the number of milliseconds to wait before the system assumes that an error has occurred.

Enter an integer between 1,000 and 100,000. The default is 2,000 milliseconds.

RATOV field

The resource allocation timeout value (RATOV), which is the number of milliseconds to wait before the system assumes that a resource cannot be properly allocated.

Enter an integer between 5,000 and 100,000. The default is 10,000 milliseconds.

Max Data Field Size field

The maximum size of the Fibre Channel frame payload bytes that the vHBA supports.

Enter an integer between 256 and 2112.

Channel Number field

The channel number that will be assigned to this vHBA.

Enter an integer between 1 and 1,000.

Note 

VNTAG mode is required for this option.

Port Profile drop-down list

The port profile that should be associated with the vHBA, if any.

This field displays the port profiles defined on the switch to which this server is connected.

Note 

VNTAG mode is required for this option.

Step 5

In the Error Recovery area, update the following fields:

Name Description

Enable FCP Error Recovery check box

If checked, the system uses FCP Sequence Level Error Recovery protocol (FC-TAPE).

Link Down Timeout field

The number of milliseconds the uplink port should be offline before it informs the system that the uplink port is down and fabric connectivity has been lost.

Enter an integer between 0 and 240,000.

Port Down I/O Retries field

The number of times an I/O request to a port is returned because the port is busy before the system decides the port is unavailable.

Enter an integer between 0 and 255.

I/O Timeout Retry field

The time period till which the system waits for timeout before retrying. When a disk does not respond for I/O within the defined timeout period, the driver aborts the pending command, and resends the same I/O after the timer expires.

Enter an integer between 1 and 59.

Port Down Timeout field

The number of milliseconds a remote Fibre Channel port should be offline before informing the SCSI upper layer that the port is unavailable.

Enter an integer between 0 and 240,000.

Step 6

In the Fibre Channel Interrupt area, update the following fields:

Name Description

Interrupt Mode drop-down list

The preferred driver interrupt mode. This can be one of the following:

  • MSIx—Message Signaled Interrupts (MSI) with the optional extension. This is the recommended option.

  • MSI—MSI only.

  • INTx—PCI INTx interrupts.

Step 7

In the Fibre Channel Port area, update the following fields:

Name Description

I/O Throttle Count field

The number of I/O operations that can be pending in the vHBA at one time.

Enter an integer between 1 and 1,024.

LUNs per Target field

The maximum number of LUNs that the driver will export. This is usually an operating system platform limitation.

Enter an integer between 1 and 1,024. The recommended value is 1024.

LUN Queue Depth field

The number of commands that the HBA can send or receive in a single chunk per LUN. This parameter adjusts the initial queue depth for all LUNs on the adapter.

Default value is 20 for physical miniports and 250 for virtual miniports.

Step 8

In the Fibre Channel Port FLOGI area, update the following fields:

Name Description

FLOGI Retries field

The number of times that the system tries to log in to the fabric after the first failure.

To specify an unlimited number of retries, select the INFINITE radio button. Otherwise select the second radio button and enter an integer into the corresponding field.

FLOGI Timeout field

The number of milliseconds that the system waits before it tries to log in again.

Enter an integer between 1,000 and 255,000.

Step 9

In the Fibre Channel Port PLOGI area, update the following fields:

Name Description

PLOGI Retries field

The number of times that the system tries to log in to a port after the first failure.

Enter an integer between 0 and 255.

PLOGI Timeout field

The number of milliseconds that the system waits before it tries to log in again.

Enter an integer between 1,000 and 255,000.

Step 10

In the SCSI I/O area, update the following fields:

Name Description

CDB Transmit Queue Count field

The number of SCSI I/O queue resources the system should allocate.

Enter an integer between 1 and 8.

CDB Transmit Queue Ring Size field

The number of descriptors in each SCSI I/O queue.

Enter an integer between 64 and 512.

Step 11

In the Receive/Transmit Queues area, update the following fields:

Name Description

FC Work Queue Ring Size field

The number of descriptors in each transmit queue.

Enter an integer between 64 and 128.

FC Receive Queue Ring Size field

The number of descriptors in each receive queue.

Enter an integer between 64 and 128.

Step 12

Click Save Changes.

Creating a vHBA

The Cisco UCS Virtual Interface Cards provide two vHBAs and two vNICs by default. You can create up to 14 additional vHBAs or vNICs on these adapter cards.

The Cisco UCS 1455 and 1457 Virtual Interface Cards, in non-port channel mode, provide four vHBAs and four vNICs by default. You can create up to 10 additional vHBAs or vNICs on these adapter cards.

Procedure

Step 1

In the Navigation pane, click the Networking menu.

Step 2

In the Adapter Card pane, click the vHBAs tab.

Step 3

In the Host Fibre Channel Interfaces area, choose one of these actions:

  • To create a vHBA using default configuration settings, click Add vHBA.
  • To create a vHBA using the same configuration settings as an existing vHBA, select that vHBA and click Clone vHBA.

The Add vHBA dialog box appears.

Step 4

In the Add vHBA dialog box, enter a name for the vHBA in the Name entry box.

Step 5

Click Add vHBA.

What to do next

  • Reboot the server to create the vHBA.

  • If configuration changes are required, configure the new vHBA as described in Modifying vHBA Properties.

Deleting a vHBA

Procedure

Step 1

In the Navigation pane, click the Networking menu.

Step 2

In the Adapter Card pane, click the vHBAs tab.

Step 3

In the Host Fibre Channel Interfaces area, select a vHBA or vHBAs from the table.

Note 
You cannot delete either of the two default vHBAs, fc0 or fc1.
Step 4

Click Delete vHBAs and click OK to confirm.

vHBA Boot Table

In the vHBA boot table, you can specify up to four LUNs from which the server can boot.

Creating a Boot Table Entry

Procedure

Step 1

In the Navigation pane, click the Networking menu.

Step 2

In the Adapter Card pane, click the vHBAs tab.

Step 3

In the Fibre Channel Interfaces area, scroll down to the Boot Table area.

Step 4

Click the Add Boot Entry button to open the Add Boot Entry dialog box.

Step 5

In the Add Boot Entry dialog box, review the following information and perform the actions specified:

Name Description

Target WWPN field

The World Wide Port Name (WWPN) that corresponds to the location of the boot image.

Enter the WWPN in the format hh:hh:hh:hh:hh:hh:hh:hh.

LUN ID field

The LUN ID that corresponds to the location of the boot image.

Enter an ID between 0 and 255.

Add Boot Entry button

Adds the specified location to the boot table.

Reset Values button

Clears the values currently entered in the fields.

Cancel button

Closes the dialog box without saving any changes made while the dialog box was open.

Deleting a Boot Table Entry

Procedure

Step 1

In the Navigation pane, click the Networking menu.

Step 2

In the Adapter Card pane, click the vHBAs tab.

Step 3

In the Fibre Channel Interfaces area, scroll down to the Boot Table area.

Step 4

In the Boot Table area, click the entry to be deleted.

Step 5

Click Delete Boot Entry and click OK to confirm.

vHBA Persistent Binding

Persistent binding ensures that the system-assigned mapping of Fibre Channel targets is maintained after a reboot.

Viewing Persistent Bindings

Procedure

Step 1

In the Navigation pane, click the Networking menu.

Step 2

In the Adapter Card pane, click the vHBAs tab.

Step 3

In the vHBAs pane, click fc0 or fc1.

Step 4

In the Persistent Bindings dialog box, review the following information:

Name Description

Index column

The unique identifier for the binding.

Target WWPN column

The target World Wide Port Name with which the binding is associated.

Host WWPN column

The host World Wide Port Name with which the binding is associated.

Bus ID column

The bus ID with which the binding is associated.

Target ID column

The target ID on the host system with which the binding is associated.

Rebuild Persistent Bindings button

Clears all unused bindings and resets the ones that are in use.

Close button

Closes the dialog box and saves your changes.
Step 5

Click Close.

Rebuilding Persistent Bindings

Procedure

Step 1

In the Navigation pane, click the Networking menu.

Step 2

In the Adapter Card pane, click the vHBAs tab.

Step 3

In the vHBAs pane, click fc0 or fc1.

Step 4

In the Fibre Channel Interfaces area, scroll down to the Persistent Bindings area.

Step 5

Click the Rebuild Persistent Bindings button.

Step 6

Click OK to confirm.

Managing vNICs

Guidelines for Managing vNICs

When managing vNICs, consider the following guidelines and restrictions:

  • The Cisco UCS Virtual Interface Cards provide two vHBAs and two vNICs by default. You can create up to 14 additional vHBAs or vNICs on these adapter cards.

    The Cisco UCS 1455 and 1457 Virtual Interface Cards, in non-port channel mode, provide four vHBAs and four vNICs by default. You can create up to 10 additional vHBAs or vNICs on these adapter cards.


    Note

    If Network Interface Virtualization (NIV) mode is enabled for the adapter, you must assign a channel number to a vNIC when you create it.

  • After making configuration changes, you must reboot the host for settings to take effect.

Cisco C-series servers use Remote Direct Memory Access (RDMA) over Converged Ethernet (RoCE) for packet transfers. RoCE defines the mechanism of performing RDMA over ethernet, based on the similar mechanism of RDMA over Infiniband. However, RoCE, with its performance oriented characteristics, delivers a superior performance compared to traditional network socket implementation because of the lower latency, lower CPU utilization and higher utilization of network bandwidth. RoCE meets the requirement of moving large amount of data across networks very efficiently.

The RoCE firmware requires the following configuration parameters provided by Cisco UCS Manager for better vNIC performance:

  • Queue Pairs

  • Memory Regions

  • Resource Groups

Guidelines and Limitations for SMB Direct with RoCE

  • Microsoft SMB Direct with RoCE is supported:

    • On Windows 2012 R2.

    • On Windows 2016.

  • Cisco UCS C-Series server does not support more than 4 RoCE-enabled vNICs per adapter.

  • Cisco UCS C-Series server does not support RoCE with NVGRE, VXLAN, VMQ, or usNIC.

  • Maximum number of queue pairs per adapter is 8192.

  • Maximum number of memory regions per adapter is 524288.

  • RoCE configuration is supported between Cisco adapters. Interoperability between Cisco adapters and third party adapters is not supported.


Important

It is required to configure the no-drop QOS policy settings at the switches in the RDMA traffic path.

Viewing vNIC Properties

Procedure

Step 1

In the Navigation pane, click the Networking menu.

Step 2

In the Adapter Card pane, click the vNICs tab.

Step 3

In the vNICs pane, click eth0 or eth1.

Step 4

In the Ethernet Interfaces pane's vNIC Properties area, review the information in the following fields:

Name Description

Name field

The name for the virtual NIC.

This name cannot be changed after the vNIC has been created.

CDN field

The Consistent Device Name (CDN) that you can assign to the ethernet vNICs on the VIC cards. Assigning a specific CDN to a device helps in identifying it on the host OS.

Note 

This feature works only when the CDN Support for VIC token is enabled in the BIOS.

MTU field

The maximum transmission unit, or packet size, that this vNIC accepts.

Enter an integer between 1500 and 9000.

Uplink Port drop-down list

The uplink port associated with this vNIC. All traffic for this vNIC goes through this uplink port.

MAC Address field

The MAC address associated with the vNIC.

To let the adapter select an available MAC address from its internal pool, select Auto. To specify an address, click the second radio button and enter the MAC address in the corresponding field.

Class of Service drop-down list

The class of service to associate with traffic from this vNIC.

Select an integer between 0 and 6, with 0 being lowest priority and 6 being the highest priority.

Note 

This option cannot be used in VNTAG mode.

Trust Host CoS check box

Check this box if you want the vNIC to use the class of service provided by the host operating system.

PCI Order field

The order in which this vNIC will be used.

To specify an order, enter an integer within the displayed range.

Default VLAN field

If there is no default VLAN for this vNIC, click NONE. Otherwise, click the second radio button and enter a VLAN ID between 1 and 4094 in the field.

Note 

This option cannot be used in VNTAG mode.

VLAN Mode drop-down list

If you want to use VLAN trunking, select TRUNK. Otherwise, select ACCESS.

Note 

This option cannot be used in VNTAG mode.

Rate Limit field

If you want this vNIC to have an unlimited data rate, select OFF. Otherwise, click the second radio button and enter a rate limit in the associated field.

Enter an integer between 1 and 10,000 Mbps.

For VIC 13xx controllers, you can enter an integer between 1 and 40,000 Mbps.

For VIC 1455 and 1457 controllers:

  • If the adapter is connected to 25 Gbps link on a Switch, then you can enter an integer between 1 to 25,000 Mbps for the Rate Limit field.

  • If the adapter is connected to 10 Gbps link on a Switch, then you can enter an integer between 1 to 10,000 Mbps for the Rate Limit field.

For VIC 1495 and 1497 controllers:

  • If the adapter is connected to 40 Gbps link on a switch, then you can enter an integer between 1 to 40,000 Mbps for the Rate Limit field.

  • If the adapter is connected to 100 Gbps link on a switch, then you can enter an integer between 1 to 100,000 Mbps for the Rate Limit field.

Note 

This option cannot be used in VNTAG mode.

Enable PXE Boot check box

Check this box if the vNIC can be used to perform a PXE boot.

Channel Number field

Select the channel number that will be assigned to this vNIC.

Note 

VNTAG mode is required for this option.

PCI Link field

The link through which vNICs can be connected. These are the following values:

  • 0 - The first cross-edged link where the vNIC is placed.

  • 1 - The second cross-edged link where the vNIC is placed.

Note 

  • This option is available only on some Cisco UCS C-Series servers.

Port Profile drop-down list

Select the port profile that should be associated with the vNIC.

This field displays the port profiles defined on the switch to which this server is connected.

Note 

VNTAG mode is required for this option.

Enable Uplink Failover check box

Check this box if traffic on this vNIC should fail over to the secondary interface if there are communication problems.

Note 

VNTAG mode is required for this option.

Enable VMQ check box

Check this box to enable Virtual Machine Queue (VMQ).

Note 

Ensure that VMQ is not enabled when SR-IOV or netflow option is enabled on the adapter.

This option is available only on some Cisco UCS C-Series servers.

Enable aRFS check box

Check this box to enable Accelerated Receive Flow steering (aRFS).

This option is available only on some Cisco UCS C-Series servers.

Enable NVGRE check box

Check this box to enable Network Virtualization using Generic Routing Encapsulation.

  • This option is available only on some Cisco UCS C-Series servers.

  • This option is available only on C-Series servers with Cisco VIC 1385 cards.

Enable VXLAN check box

Check this box to enable Virtual Extensible LAN.

  • This option is available only on some Cisco UCS C-Series servers.

  • This option is available only on C-Series servers with Cisco VIC 1385 and VIC 14xx cards.

Advanced Filter check box

Check this box to enable advanced filter options in vNICs.

Failback Timeout field

After a vNIC has started using its secondary interface, this setting controls how long the primary interface must be available before the system resumes using the primary interface for the vNIC.

Enter a number of seconds between 0 and 600.

Note 

VNTAG mode is required for this option.

Step 5

In the Ethernet Interrupt area, review the information in the following fields:

Name Description

Interrupt Count field

The number of interrupt resources to allocate. In general, this value should be equal to the number of completion queue resources.

Enter an integer between 1 and 514.

Coalescing Time field

The time to wait between interrupts or the idle period that must be encountered before an interrupt is sent.

Enter an integer between 1 and 65535. To turn off interrupt coalescing, enter 0 (zero) in this field.

Coalescing Type drop-down list

This can be one of the following:

  • MIN—The system waits for the time specified in the Coalescing Time field before sending another interrupt event.

  • IDLE—The system does not send an interrupt until there is a period of no activity lasting as least as long as the time specified in the Coalescing Time field.

Interrupt Mode drop-down list

The preferred driver interrupt mode. This can be one of the following:

  • MSI-X—Message Signaled Interrupts (MSI) with the optional extension. This is the recommended option.

  • MSI—MSI only.

  • INTx—PCI INTx interrupts.

Step 6

In the Ethernet Receive Queue area, review the information in the following fields:

Name Description

Receive Queue Count field

The number of receive queue resources to allocate.

Enter an integer between 1 and 256.

Receive Queue Ring Size field

The number of descriptors in each receive queue.

Enter an integer between 64 and 4096.

Step 7

In the Ethernet Transmit Queue area, review the information in the following fields:

Name Description

Transmit Queue Count field

The number of transmit queue resources to allocate.

Enter an integer between 1 and 256.

Transmit Queue Ring Size field

The number of descriptors in each transmit queue.

Enter an integer between 64 and 4096.

Step 8

In the Completion Queue area, review the information in the following fields:

Name Description

Completion Queue Count field

The number of completion queue resources to allocate. In general, the number of completion queue resources you should allocate is equal to the number of transmit queue resources plus the number of receive queue resources.

Enter an integer between 1 and 512.

Completion Queue Ring Size field

The number of descriptors in each completion queue.

This value cannot be changed.

Step 9

In the TCP Offload area, review the information in the following fields:

Name Description

Enable TCP Segmentation Offload check box

If checked, the CPU sends large TCP packets to the hardware to be segmented. This option may reduce CPU overhead and increase throughput rate.

If cleared, the CPU segments large packets.

Note 

This option is also known as Large Send Offload (LSO).

Enable TCP Rx Offload Checksum Validation check box

If checked, the CPU sends all packet checksums to the hardware for validation. This option may reduce CPU overhead.

If cleared, the CPU validates all packet checksums.

Enable TCP Tx Offload Checksum Generation check box

If checked, the CPU sends all packets to the hardware so that the checksum can be calculated. This option may reduce CPU overhead.

If cleared, the CPU calculates all packet checksums.

Enable Large Receive check box

If checked, the hardware reassembles all segmented packets before sending them to the CPU. This option may reduce CPU utilization and increase inbound throughput.

If cleared, the CPU processes all large packets.

Step 10

In the Receive Side Scaling area, review the information in the following fields:

Name Description

Enable TCP Receive Side Scaling check box

Receive Side Scaling (RSS) distributes network receive processing across multiple CPUs in multiprocessor systems.

If checked, network receive processing is shared across processors whenever possible.

If cleared, network receive processing is always handled by a single processor even if additional processors are available.

Enable IPv4 RSS check box

If checked, RSS is enabled on IPv4 networks.

Enable TCP-IPv4 RSS check box

If checked, RSS is enabled for TCP transmissions across IPv4 networks.

Enable IPv6 RSS check box

If checked, RSS is enabled on IPv6 networks.

Enable TCP-IPv6 RSS check box

If checked, RSS is enabled for TCP transmissions across IPv6 networks.

Enable IPv6 Extension RSS check box

If checked, RSS is enabled for IPv6 extensions.

Enable TCP-IPv6 Extension RSS check box

If checked, RSS is enabled for TCP transmissions across IPv6 networks.

Modifying vNIC Properties

Procedure

Step 1

In the Navigation pane, click the Networking menu.

Step 2

In the Adapter Card pane, click the vNICs tab.

Step 3

In the vNICs pane, click eth0 or eth1.

Step 4

In the Ethernet Interfaces pane's vNIC Properties area, update the following fields:

Name Description

Name field

The name for the virtual NIC.

This name cannot be changed after the vNIC has been created.

CDN field

The Consistent Device Name (CDN) that you can assign to the ethernet vNICs on the VIC cards. Assigning a specific CDN to a device helps in identifying it on the host OS.

Note 

This feature works only when the CDN Support for VIC token is enabled in the BIOS.

MTU field

The maximum transmission unit, or packet size, that this vNIC accepts.

Enter an integer between 1500 and 9000.

Uplink Port drop-down list

The uplink port associated with this vNIC. All traffic for this vNIC goes through this uplink port.

MAC Address field

The MAC address associated with the vNIC.

To let the adapter select an available MAC address from its internal pool, select Auto. To specify an address, click the second radio button and enter the MAC address in the corresponding field.

Class of Service drop-down list

The class of service to associate with traffic from this vNIC.

Select an integer between 0 and 6, with 0 being lowest priority and 6 being the highest priority.

Note 

This option cannot be used in VNTAG mode.

Trust Host CoS check box

Check this box if you want the vNIC to use the class of service provided by the host operating system.

PCI Order field

The order in which this vNIC will be used.

To specify an order, enter an integer within the displayed range.

Default VLAN field

If there is no default VLAN for this vNIC, click NONE. Otherwise, click the second radio button and enter a VLAN ID between 1 and 4094 in the field.

Note 

This option cannot be used in VNTAG mode.

VLAN Mode drop-down list

If you want to use VLAN trunking, select TRUNK. Otherwise, select ACCESS.

Note 

This option cannot be used in VNTAG mode.

Rate Limit field

If you want this vNIC to have an unlimited data rate, select OFF. Otherwise, click the second radio button and enter a rate limit in the associated field.

Enter an integer between 1 and 10,000 Mbps.

For VIC 13xx controllers, you can enter an integer between 1 and 40,000 Mbps.

For VIC 1455 and 1457 controllers:

  • If the adapter is connected to 25 Gbps link on a Switch, then you can enter an integer between 1 to 25,000 Mbps for the Rate Limit field.

  • If the adapter is connected to 10 Gbps link on a Switch, then you can enter an integer between 1 to 10,000 Mbps for the Rate Limit field.

For VIC 1495 and 1497 controllers:

  • If the adapter is connected to 40 Gbps link on a switch, then you can enter an integer between 1 to 40,000 Mbps for the Rate Limit field.

  • If the adapter is connected to 100 Gbps link on a switch, then you can enter an integer between 1 to 100,000 Mbps for the Rate Limit field.

Note 

This option cannot be used in VNTAG mode.

Enable PXE Boot check box

Check this box if the vNIC can be used to perform a PXE boot.

Channel Number field

Select the channel number that will be assigned to this vNIC.

Note 

VNTAG mode is required for this option.

PCI Link field

The link through which vNICs can be connected. These are the following values:

  • 0 - The first cross-edged link where the vNIC is placed.

  • 1 - The second cross-edged link where the vNIC is placed.

Note 

  • This option is available only on some Cisco UCS C-Series servers.

Port Profile drop-down list

Select the port profile that should be associated with the vNIC.

This field displays the port profiles defined on the switch to which this server is connected.

Note 

VNTAG mode is required for this option.

Enable Uplink Failover check box

Check this box if traffic on this vNIC should fail over to the secondary interface if there are communication problems.

Note 

VNTAG mode is required for this option.

Enable VMQ check box

Check this box to enable Virtual Machine Queue (VMQ).

Note 

Ensure that VMQ is not enabled when SR-IOV or netflow option is enabled on the adapter.

This option is available only on some Cisco UCS C-Series servers.

Enable aRFS check box

Check this box to enable Accelerated Receive Flow steering (aRFS).

This option is available only on some Cisco UCS C-Series servers.

Enable NVGRE check box

Check this box to enable Network Virtualization using Generic Routing Encapsulation.

  • This option is available only on some Cisco UCS C-Series servers.

  • This option is available only on C-Series servers with Cisco VIC 1385 cards.

Enable VXLAN check box

Check this box to enable Virtual Extensible LAN.

  • This option is available only on some Cisco UCS C-Series servers.

  • This option is available only on C-Series servers with Cisco VIC 1385 and VIC 14xx cards.

Advanced Filter check box

Check this box to enable advanced filter options in vNICs.

Failback Timeout field

After a vNIC has started using its secondary interface, this setting controls how long the primary interface must be available before the system resumes using the primary interface for the vNIC.

Enter a number of seconds between 0 and 600.

Note 

VNTAG mode is required for this option.

Step 5

In the Ethernet Interrupt area, update the following fields:

Name Description

Interrupt Count field

The number of interrupt resources to allocate. In general, this value should be equal to the number of completion queue resources.

Enter an integer between 1 and 514.

Coalescing Time field

The time to wait between interrupts or the idle period that must be encountered before an interrupt is sent.

Enter an integer between 1 and 65535. To turn off interrupt coalescing, enter 0 (zero) in this field.

Coalescing Type drop-down list

This can be one of the following:

  • MIN—The system waits for the time specified in the Coalescing Time field before sending another interrupt event.

  • IDLE—The system does not send an interrupt until there is a period of no activity lasting as least as long as the time specified in the Coalescing Time field.

Interrupt Mode drop-down list

The preferred driver interrupt mode. This can be one of the following:

  • MSI-X—Message Signaled Interrupts (MSI) with the optional extension. This is the recommended option.

  • MSI—MSI only.

  • INTx—PCI INTx interrupts.

Step 6

In the Ethernet Receive Queue area, update the following fields:

Name Description

Receive Queue Count field

The number of receive queue resources to allocate.

Enter an integer between 1 and 256.

Receive Queue Ring Size field

The number of descriptors in each receive queue.

Enter an integer between 64 and 4096.

Step 7

In the Ethernet Transmit Queue area, update the following fields:

Name Description

Transmit Queue Count field

The number of transmit queue resources to allocate.

Enter an integer between 1 and 256.

Transmit Queue Ring Size field

The number of descriptors in each transmit queue.

Enter an integer between 64 and 4096.

Step 8

In the Completion Queue area, update the following fields:

Name Description

Completion Queue Count field

The number of completion queue resources to allocate. In general, the number of completion queue resources you should allocate is equal to the number of transmit queue resources plus the number of receive queue resources.

Enter an integer between 1 and 512.

Completion Queue Ring Size field

The number of descriptors in each completion queue.

This value cannot be changed.

Step 9

In the RoCE Properties area, update the following fields:

Name Description

RoCE check box

Check the check box to change the RoCE Properties.

Queue Pairs (1 - 2048) field

The number of queue pairs per adapter. Enter an integer between 1 and 2048. We recommend that this number be an integer power of 2.

Memory Regions (1 - 524288) field

The number of memory regions per adapter. Enter an integer between 1 and 524288. We recommend that this number be an integer power of 2.

Resource Groups (1 - 128) field

The number of resource groups per adapter. Enter an integer between 1 and 128. We recommend that this number be an integer power of 2 greater than or equal to the number of CPU cores on the system for optimum performance.

Step 10

In the TCP Offload area, update the following fields:

Name Description

Enable TCP Segmentation Offload check box

If checked, the CPU sends large TCP packets to the hardware to be segmented. This option may reduce CPU overhead and increase throughput rate.

If cleared, the CPU segments large packets.

Note 

This option is also known as Large Send Offload (LSO).

Enable TCP Rx Offload Checksum Validation check box

If checked, the CPU sends all packet checksums to the hardware for validation. This option may reduce CPU overhead.

If cleared, the CPU validates all packet checksums.

Enable TCP Tx Offload Checksum Generation check box

If checked, the CPU sends all packets to the hardware so that the checksum can be calculated. This option may reduce CPU overhead.

If cleared, the CPU calculates all packet checksums.

Enable Large Receive check box

If checked, the hardware reassembles all segmented packets before sending them to the CPU. This option may reduce CPU utilization and increase inbound throughput.

If cleared, the CPU processes all large packets.

Step 11

In the Receive Side Scaling area, update the following fields:

Step 12

Click Save Changes.

Creating a vNIC

The Cisco UCS Virtual Interface Cards provide two vHBAs and two vNICs by default. You can create up to 14 additional vHBAs or vNICs on these adapter cards.

The Cisco UCS 1455 and 1457 Virtual Interface Cards, in non-port channel mode, provide four vHBAs and four vNICs by default. You can create up to 10 additional vHBAs or vNICs on these adapter cards.

Procedure

Step 1

In the Navigation pane, click the Networking menu.

Step 2

In the Adapter Card pane, click the vNICs tab.

Step 3

In the Host Ethernet Interfaces area, choose one of these actions:

  • To create a vNIC using default configuration settings, click Add vNIC.
  • To create a vNIC using the same configuration settings as an existing vNIC, select that vNIC and click Clone vNIC.

The Add vNIC dialog box appears.

Step 4

In the Add vNIC dialog box, enter a name for the vNIC in the Name entry box.

Step 5

(Optional) In the Add vNIC dialog box, enter a channel number for the vNIC in the Channel Number entry box.

Note 

If NIV is enabled on the adapter, you must assign a channel number for the vNIC when you create it.

Step 6

Click Add vNIC.

What to do next

If configuration changes are required, configure the new vNIC as described in Modifying vNIC Properties.

Deleting a vNIC

Procedure

Step 1

In the Navigation pane, click the Networking menu.

Step 2

In the Adapter Card pane, click the vNICs tab.

Step 3

In the Host Ethernet Interfaces area, select a vNIC from the table.

Note 
You cannot delete either of the two default vNICs, eth0 or eth1.
Step 4

Click Delete vNIC and click OK to confirm.

Managing Cisco usNIC

Overview of Cisco usNIC

The Cisco user-space NIC (Cisco usNIC) feature improves the performance of software applications that run on the Cisco UCS servers in your data center by bypassing the kernel when sending and receiving networking packets. The applications interact directly with a Cisco UCS VIC second generation or later generation adapter, such as the , which improves the networking performance of your high-performance computing cluster. To benefit from Cisco usNIC, your applications must use the Message Passing Interface (MPI) instead of sockets or other communication APIs.

Cisco usNIC offers the following benefits for your MPI applications:

  • Provides a low-latency and high-throughput communication transport.

  • Employs the standard and application-independent Ethernet protocol.

  • Takes advantage of low­latency forwarding, Unified Fabric, and integrated management support in the following Cisco data center platforms:

    • Cisco UCS server

    • Cisco UCS VIC second generation or later generation adapter

    • 10 or 40GbE networks

Standard Ethernet applications use user-space socket libraries, which invoke the networking stack in the Linux kernel. The networking stack then uses the Cisco eNIC driver to communicate with the Cisco VIC hardware. The following figure shows the contrast between a regular software application and an MPI application that uses Cisco usNIC.

Figure 1. Kernel-Based Network Communication versus Cisco usNIC-Based Communication

Viewing and Configuring Cisco usNIC using the Cisco IMC GUI

Before you begin

You must log in to the Cisco IMC GUI with administrator privileges to perform this task. Click Play on this video to watch how to configure Cisco usNIC in CIMC.

Procedure

Step 1

Log into the Cisco IMC GUI.

For more information about how to log into Cisco IMC, see Cisco UCS C-Series Servers Integrated Management Controller GUI Configuration Guide.

Step 2

In the Navigation pane, click the Networking menu.

Step 3

In the Adapter Card pane, click the vNICs tab.

Step 4

In the vNICs pane, click eth0 or eth1.

Step 5

In the Ethernet Interfaces area, select the usNIC area.

Note 

usNIC support is not available for C125 servers.
Step 6

In the Properties area, review and update the following fields:

Name Description

Name

The name for the vNIC that is the parent of the usNIC.

Note 

This field is read-only.

usNIC field

The number of usNICs assigned to the specific vNIC.

Enter an integer between 0 and 225.

To assign additional usNICs to a specified vNIC, enter value higher than the existing value.

To delete usNICs from a specified vNIC, enter value smaller than the existing value.

To delete all the usNICs assigned to a vNIC, enter zero.

Transmit Queue Count field

The number of transmit queue resources to allocate.

Enter an integer between 1 and 256.

Receive Queue Count field

The number of receive queue resources to allocate.

Enter an integer between 1 and 256.

Completion Queue Count field

The number of completion queue resources to allocate. In general, the number of completion queue resources you should allocate is equal to the number of transmit queue resources plus the number of receive queue resources.

Enter an integer between 1 and 512.

Transmit Queue Ring Size field

The number of descriptors in each transmit queue.

Enter an integer between 64 and 4096.

Receive Queue Ring Size field

The number of descriptors in each receive queue.

Enter an integer between 64 and 4096.

Interrupt Count field

The number of interrupt resources to allocate. In general, this value should be equal to the number of completion queue resources.

Enter an integer between 1 and 514.

Interrupt Coalescing Type drop-down list

This can be one of the following:

  • MIN—The system waits for the time specified in the Coalescing Time field before sending another interrupt event.

  • IDLE—The system does not send an interrupt until there is a period of no activity lasting as least as long as the time specified in the Coalescing Time field.

Interrupt Coalescing Timer Time field

The time to wait between interrupts or the idle period that must be encountered before an interrupt is sent.

Enter an integer between 1 and 65535. To turn off interrupt coalescing, enter 0 (zero) in this field.

Class of Service field

The class of service to associate with traffic from this usNIC.

Select an integer between 0 and 6, with 0 being lowest priority and 6 being the highest priority.

Note 

This option cannot be used in VNTAG mode.

TCP Segment Offload check box

If checked, the CPU sends large TCP packets to the hardware to be segmented. This option may reduce CPU overhead and increase throughput rate.

If cleared, the CPU segments large packets.

Note 

This option is also known as Large Send Offload (LSO).

Large Receive check box

If checked, the hardware reassembles all segmented packets before sending them to the CPU. This option may reduce CPU utilization and increase inbound throughput.

If cleared, the CPU processes all large packets.

TCP Tx Checksum check box

If checked, the CPU sends all packets to the hardware so that the checksum can be calculated. This option may reduce CPU overhead.

If cleared, the CPU calculates all packet checksums.

TCP Rx Checksum check box

If checked, the CPU sends all packet checksums to the hardware for validation. This option may reduce CPU overhead.

If cleared, the CPU validates all packet checksums.

Step 7

Click Save Changes.

The changes take effect upon the next server reboot.

Viewing usNIC Properties

Procedure

Step 1

In the Navigation pane, click the Networking menu.

Step 2

In the Adapter Card pane, click the vNICs tab.

Step 3

In the vNICs pane, click eth0 or eth1.

Step 4

In the Host Ethernet Interfaces pane's usNIC Properties area, review the information in the following fields:

Note 

usNIC support is not available for C125 servers.
Name Description

Name

The name for the vNIC that is the parent of the usNIC.

Note 

This field is read-only.

usNIC field

The number of usNICs assigned to the specific vNIC.

Enter an integer between 0 and 225.

To assign additional usNICs to a specified vNIC, enter value higher than the existing value.

To delete usNICs from a specified vNIC, enter value smaller than the existing value.

To delete all the usNICs assigned to a vNIC, enter zero.

Transmit Queue Count field

The number of transmit queue resources to allocate.

Enter an integer between 1 and 256.

Receive Queue Count field

The number of receive queue resources to allocate.

Enter an integer between 1 and 256.

Completion Queue Count field

The number of completion queue resources to allocate. In general, the number of completion queue resources you should allocate is equal to the number of transmit queue resources plus the number of receive queue resources.

Enter an integer between 1 and 512.

Transmit Queue Ring Size field

The number of descriptors in each transmit queue.

Enter an integer between 64 and 4096.

Receive Queue Ring Size field

The number of descriptors in each receive queue.

Enter an integer between 64 and 4096.

Interrupt Count field

The number of interrupt resources to allocate. In general, this value should be equal to the number of completion queue resources.

Enter an integer between 1 and 514.

Interrupt Coalescing Type drop-down list

This can be one of the following:

  • MIN—The system waits for the time specified in the Coalescing Time field before sending another interrupt event.

  • IDLE—The system does not send an interrupt until there is a period of no activity lasting as least as long as the time specified in the Coalescing Time field.

Interrupt Coalescing Timer Time field

The time to wait between interrupts or the idle period that must be encountered before an interrupt is sent.

Enter an integer between 1 and 65535. To turn off interrupt coalescing, enter 0 (zero) in this field.

Class of Service field

The class of service to associate with traffic from this usNIC.

Select an integer between 0 and 6, with 0 being lowest priority and 6 being the highest priority.

Note 

This option cannot be used in VNTAG mode.

TCP Segment Offload check box

If checked, the CPU sends large TCP packets to the hardware to be segmented. This option may reduce CPU overhead and increase throughput rate.

If cleared, the CPU segments large packets.

Note 

This option is also known as Large Send Offload (LSO).

Large Receive check box

If checked, the hardware reassembles all segmented packets before sending them to the CPU. This option may reduce CPU utilization and increase inbound throughput.

If cleared, the CPU processes all large packets.

TCP Tx Checksum check box

If checked, the CPU sends all packets to the hardware so that the checksum can be calculated. This option may reduce CPU overhead.

If cleared, the CPU calculates all packet checksums.

TCP Rx Checksum check box

If checked, the CPU sends all packet checksums to the hardware for validation. This option may reduce CPU overhead.

If cleared, the CPU validates all packet checksums.

Configuring iSCSI Boot Capability

Configuring iSCSI Boot Capability for vNICs

When the rack-servers are configured in a standalone mode, and when the VIC adapters are directly attached to the Nexus 5000 and Nexus 6000 family of switches, you can configure these VIC adapters to boot the servers remotely from iSCSI storage targets. You can configure Ethernet vNICs to enable a rack server to load the host OS image from remote iSCSI target devices.

To configure the iSCSI boot capability on a vNIC:

  • You must log in with admin privileges to perform this task.

  • To configure a vNIC to boot a server remotely from an iSCSI storage target, you must enable the PXE boot option on the vNIC.


Note

You can configure a maximum of 2 iSCSI vNICs for each host.

Configuring iSCSI Boot Capability on a vNIC

You can configure a maximum of 2 iSCSI vNICs for each host.

Before you begin

  • To configure a vNIC to boot a server remotely from an iSCSI storage target, you must enable the PXE boot option on the vNIC.

  • You must log in with admin privileges to perform this task.

Procedure

Step 1

In the Navigation pane, click the Networking menu.

Step 2

In the Adapter Card pane, click the vNICs tab.

Step 3

In the vNICs pane, click eth0 or eth1.

Step 4

In the Ethernet Interfaces area, select the iSCSI Boot Properties area.

Step 5

In the General Area, update the following fields:

Name Description

Name field

The name of the vNIC.

DHCP Network check box

Whether DHCP Network is enabled for the vNIC.

If enabled, the initiator network configuration is obtained from the DHCP server.

DHCP iSCSI check box

Whether DHCP iSCSI is enabled for the vNIC. If enabled and the DHCP ID is set, the initiator IQN and target information are obtained from the DHCP server.

Note 

If DHCP iSCSI is enabled without a DHCP ID, only the target information is obtained.

DHCP ID field

The vendor identifier string used by the adapter to obtain the initiator IQN and target information from the DHCP server.

Enter a string up to 64 characters.

DHCP Timeout field

The number of seconds to wait before the initiator assumes that the DHCP server is unavailable.

Enter an integer between 60 and 300 (default: 60 seconds)

Link Timeout field

The number of seconds to wait before the initiator assumes that the link is unavailable.

Enter an integer between 0 and 255 (default: 15 seconds)

LUN Busy Retry Count field

The number of times to retry the connection in case of a failure during iSCSI LUN discovery.

Enter an integer between 0 and 255. The default is 15.

IP Version field

The IP version to use during iSCSI boot.
Step 6

In the Initiator Area, update the following fields:

Name Description

Name field

A regular expression that defines the name of the iSCSI initiator.

You can enter any alphanumeric string as well as the following special characters:

  • . (period)

  • : (colon)

  • - (dash)

Note 

The name is in the IQN format.

IP Address field

The IP address of the iSCSI initiator.

Subnet Mask field

The subnet mask for the iSCSI initiator.

Gateway field

The default gateway.

Primary DNS field

The primary DNS server address.

Secondary DNS field

The secondary DNS server address.

TCP Timeout field

The number of seconds to wait before the initiator assumes that TCP is unavailable.

Enter an integer between 0 and 255 (default: 15 seconds)

CHAP Name field

The Challenge-Handshake Authentication Protocol (CHAP) name of the initiator.

CHAP Secret field

The Challenge-Handshake Authentication Protocol (CHAP) shared secret of the initiator.
Step 7

In the Primary Target Area, update the following fields:

Name Description

Name field

The name of the primary target in the IQN format.

IP Address field

The IP address of the target.

TCP Port field

The TCP port associated with the target.

Boot LUN field

The Boot LUN associated with the target.

CHAP Name field

The Challenge-Handshake Authentication Protocol (CHAP) name of the initiator.

CHAP Secret field

The Challenge-Handshake Authentication Protocol (CHAP) shared secret of the initiator.
Step 8

In the Secondary Target Area, update the following fields:

Name Description

Name field

The name of the secondary target in the IQN format.

IP Address field

The IP address of the target.

TCP Port field

The TCP port associated with the target.

Boot LUN field

The Boot LUN associated with the target.

CHAP Name field

The Challenge-Handshake Authentication Protocol (CHAP) name of the initiator.

CHAP Secret field

The Challenge-Handshake Authentication Protocol (CHAP) shared secret of the initiator.
Name Description

Configure ISCSI button

Configures iSCSI boot on the selected vNIC.

Unconfigure ISCSI button

Removes the configuration from the selected vNIC.

Reset Values button

Restores the values for the vNIC to the settings that were in effect when this dialog box was first opened.

Cancel button

Closes the dialog box without making any changes.
Step 9

Click Save Changes.

Removing iSCSI Boot Configuration from a vNIC

Before you begin

You must log in with admin privileges to perform this task.

Procedure

Step 1

In the Navigation pane, click the Networking menu.

Step 2

In the Adapter Card pane, click the vNICs tab.

Step 3

In the vNICs pane, click eth0 or eth1.

Step 4

In the Ethernet Interfaces area, select the iSCSI Boot Properties area.

Step 5

Click the Unconfigure ISCSI button at the bottom of the area.

Backing Up and Restoring the Adapter Configuration

Exporting the Adapter Configuration

The adapter configuration can be exported as an XML file to a remote server which can be one of the following:

  • TFTP

  • FTP

  • SFTP

  • SCP

  • HTTP

Before you begin

Obtain the remote server IP address.

Procedure

Step 1

In the Navigation pane, click the Networking menu.

Step 2

Click the Adapter Card tab.

The General tab appears.

Step 3

In the Actions area of the General tab, click Export Configuration.

The Export Adapter Configuration dialog box opens.

Step 4

In the Export Adapter Configuration dialog box, update the following fields:

Name Description

Export to drop-down list

The remote server type. This can be one of the following:

  • TFTP Server

  • FTP Server

  • SFTP Server

  • SCP Server

  • HTTP Server

Note 

If you chose SCP or SFTP as the remote server type while performing this action, a pop-up window is displayed with the message Server (RSA) key fingerprint is <server_finger_print _ID> Do you wish to continue?. Click Yes or No depending on the authenticity of the server fingerprint.

The fingerprint is based on the host's public key and helps you to identify or verify the host you are connecting to.

Server IP/Hostname field

The IPv4 or IPv6 address, or hostname of the server to which the adapter configuration file will be exported. Depending on the setting in the Export to drop-down list, the name of the field may vary.

Path and Filename field

The path and filename Cisco IMC should use when exporting the file to the remote server.

Username

The username the system should use to log in to the remote server. This field does not apply if the protocol is TFTP or HTTP.

Password

The password for the remote server username. This field does not apply if the protocol is TFTP or HTTP.

Step 5

Click Export Configuration.

Note 

After exporting, you must reset the vNIC MAC address by selecting the AUTO radio button from the vNIC configuration screen on the Cisco IMC web UI to generate a new MAC address.

Importing the Adapter Configuration

Procedure

Step 1

In the Navigation pane, click the Networking menu.

Step 2

Click the Adapter Card tab.

The General tab appears.

Step 3

In the Actions area of the General tab, click Import Configuration.

The Import Adapter Configuration dialog box opens.

Step 4

In the Import Adapter Configuration dialog box, update the following fields:

Name Description

Import from drop-down list

The remote server type. This can be one of the following:

  • TFTP Server

  • FTP Server

  • SFTP Server

  • SCP Server

  • HTTP Server

Note 

If you chose SCP or SFTP as the remote server type while performing this action, a pop-up window is displayed with the message Server (RSA) key fingerprint is <server_finger_print _ID> Do you wish to continue?. Click Yes or No depending on the authenticity of the server fingerprint.

The fingerprint is based on the host's public key and helps you to identify or verify the host you are connecting to.

Server IP/Hostname field

The IPv4 or IPv6 address, or hostname of the server on which the adapter configuration file resides. Depending on the setting in the Import from drop-down list, the name of the field may vary.

Path and Filename field

The path and filename of the configuration file on the remote server.

Username

The username the system should use to log in to the remote server. This field does not apply if the protocol is TFTP or HTTP.

Password

The password for the remote server username. This field does not apply if the protocol is TFTP or HTTP.

Step 5

Click Import Configuration.

The adapter downloads the configuration file from the specified path on the TFTP server at the specified IP address. The configuration will be installed during the next server reboot.

What to do next

Reboot the server to apply the imported configuration.

Restoring Adapter Defaults

Procedure

Step 1

In the Navigation pane, click the Networking menu.

Step 2

Click the Adapter Card tab.

The General tab appears.

Step 3

In the Actions area of the General tab, click Reset To Defaults and click OK to confirm.

Note 

Resetting the adapter to default settings sets the port speed to 4 X 10 Gbps. Choose 40 Gbps as the port speed only if you are using a 40 Gbps switch.

Resetting the Adapter

Procedure

Step 1

In the Navigation pane, click the Networking menu.

Step 2

Click the Adapter Card tab.

The General tab appears.

Step 3

In the Actions area of the General tab, click Reset and click Yes to confirm.

Note 

Resetting the adapter also resets the host and requires a reformat.