SANLink2 Articles

How to make SANLink 2 10GB work with Mac OS High Sierra 10.13.4?

2018-05-30T08:03:58.9270000

After an update to OSX 10.13.4, the Mac may crash or not work when SANLink 2 is conected. This might be a driver issue and could be fixed by removing the driver and reinstall the latest driver version again.

To delete a driver, use Terminal command:

sudo rm -rf /Library/Extensions/SANLink2-10GE.kext

After removing a driver successfully, go to Promise website to download and install the latest driver again: https://www.promise.com/support. It might take several Mac reboots to make it work.

How to Enable LACP and Jumbo Frames on a SANLink2 10Gb Base-T

2016-10-13T21:06:42.6200000

How to Enable LACP and Jumbo Frames - (Windows & OS X)

SANLink2 10Gb Drivers and Utility (OS X Only)

SANLink2 10G Ethernet Network Adapter Driver for Windows 64bit v20.3
SANLink2 10G BaseT/10G SFP+ Mac Driver v1.8.1
SANLink2 Utility for Mac OS X v1.04.0000.08

Table Of Contents

I. Create LACP with Default MTU 1500 - Windows
II. Enable Jumbo Frames on Windows WHQL certified driver for Windows 7 SP1 / Windows 8.1 and Windows 10. (All 64bit ONLY) MTU 9000
III. Create LACP with Default MTU 1500 - OS X
IV. Enable Jumbo Frames MTU 9000 - OS X

I. Create LACP with Default MTU 1500 - Windows

1. Make sure to identify your 10 Gb Network Ports from the SANLink2
2. Go to Change adapter settings

3. You can also verify the connected devices by going to My Computer > Properties > Device Manager

4. Locate Intel(R) Ethernet Controller X540-AT2
5. Right Click on Intel(R) Ethernet Controller X540-AT2 and select Properties

6. Navigate to the Teaming Tab

7. Select the checkbox, "Team this adapter with other adapters" and click on New Team

8. Specify a name for the team: LACP_Example

9. Click Next, and select both Intel(R) Ethernet Controller X540-AT2 and #2

10. Select a team type: IEEE 802.3ad Dynamic Link Aggregation. (There are other options to choose from, you may select the team type that best suits yours needs)

11. Select a profile to apply to the team: Standard Server

12. Click Finish and allow it to the configure the team.

13. Go back to the Intel(R) Ethernet Controller X540-AT2 Properties and select the Teaming Tab

14. Go to Properties
15. In the Team LACP_Example Properties, go to the Settings tab.

16. Validate that both 10Gb ports are Active as the screenshot above.
17. Open Network and Sharing [Control Panel\All Control Panel Items\Network and Sharing Center] and go back to Change adapter settings.

18. Depending on how your Windows Networking is configured (Public and Private), you will see a new Local Area Connection listed. (In this configuration, Local Connection 4 appeared under the Work Network due to the Local Policies set on this machine)
19. To validate the team configuration, click on Local Area Connection 4 (Or the number assigned to your connection)
You should see an aggregate of 20Gbps for the connection. The link speed is listed as 20 Gbps which signifies that (2) x 10Gb Ethernet ports are actually teamed together.

20. Click Properties. You can also verify in this dialogue that it is Connecting Using: TEAM: LACP_Example

21. Select IPv4 (Double-Click) or click on Properties.
22. In this example, I'm configuring a static IP. Click OK to Save.

23. Make sure to ping other devices on the same network to test your connection to your Network Switch.
24. You may also rename the connection. Go to Change adapter settings and right click Rename on Local Area Connection 4
25. Rename the connection with a unique identifiable name of your choice.

II. Enable Jumbo Frames on Windows WHQL certified driver for Windows 7 SP1 / Windows 8.1 and Windows 10. (All 64bit ONLY)

Enable Jumbo Frames only if devices across the network support them and are configured to use the same frame size. When setting up Jumbo Frames on other network devices, be aware that different network devices calculate Jumbo Frame sizes differently. Some devices include the header information in the frame size while others do not.

1. To verify the current MTU set on the Ethernet Port Team, open Command Prompt and type the following:

netsh interface ipv4 show subinterfaces
2. In this example, we see that the configured team interface, LACP_Example has an MTU of 1500 (Default)
3. Open Network and Sharing Center and select your teamed interface (LACP_Example)

4. Click on Properties and click Configure

5. Go to the Settings tab and select one of the Adapters in the the team.
6. Click on Adapter Properties.

7. Go to the Advanced tab and locate Jumbo Packet.

8. Change the Value to 9014 Bytes

9. Click OK.
10. You will receive a dialogue that states:
"All devices in a team must have the same value for Jumbo Frames/Jumbo Packets. The Jumbo Frames/Jumbo Packets setting for devices in this team will be changed automatically."

11. Click OK and you will be presented with another dialogue:
"The team will lose connectivity momentarily. Do you want to continue?" Click Yes

12. Allow the teamed interface to reconfigure itself and assign the newly MTU of 9000.

13. Open Command Prompt:

netsh interface ipv4 show subinterfaces

Now you have verified that the teamed interface now has an MTU of 9000.

Note: These changes are revertible.

Advanced Settings for Intel® Ethernet Adapters

III. Create LACP with Default MTU 1500 - OS X

1. Identify the location for each 10Gb Port by opening Network Utility

Network Utility Location = /System/Library/CoreServices/Applications/Network\ Utility.app

You can also Spotlight Search for it as well.

2. Locate each 10Gb interface location:

3. Open System Preferences

 > System Preferences > Network

4. Click the Gear button to access the Menu

5. Select Manage Virtual Interfaces

6. Click on the + symbol and select New Link Aggregate

7. Enter the Aggregate Name and select both 10Gb Thunderbolt Ethernet Ports.

8. Click Create.
9. LACP_BOND_EXAMPLE has now been assigned to BSD Name bond0. Click Done

10. Click Apply to Save the Changes.

11. Now the LACP bond will have a yellow indicator after you click Apply. In this example, the bond configured itself with an IP Address of 169.254.67.24 with a class B subnet mask of 255.255.0.0. This will happen by default when there is no DHCP server and the client uses the self-configured IP address. All you need to do is assign the correct IP address and subnet mask that corresponds to your 10Gb network.

12. Before changing the IP address on the LACP bond, check to make sure both 10Gb ports are active. Go to Advanced

13. Verify the both ports in the Bond Status tab are both active (Sending/Receiving) and have an OK status.

14. Click OK to exit.
15. If you will be configuring a static IP Address, Configure IPv4 Manually

16. After assigning the IP Address and subnet mask on the LACP bond, click Apply. The indicator status light will turn to green after you click Apply.
(I'm not using a Router IP Address in this example)

17. Verify the MTU by going back to Advanced and under the Hardware tab. The configuration is set to Automatically; therefore, it will negotiate at the MTU packet size that the switch is configured to transmit.

18. You can also verify the same in the Terminal with the following command:

ifconfig bond(x) = x being the number assigned to the bond.

In this example:

ifconfig bond019. Verifying the MTU with the SANLink Utility

Note: RSC (Receive Side Coalescing) will be enabled by default, if your Network is configured on a Tagged VLAN, you may see an increase in performance by disabling it. It has also been documented in our previous SR 2.6 Release Notes.

Download Release Notes SR 2.6

IV. Enable Jumbo Frames MTU 9000 - OS X

1. In Network Preferences, select your LACP bond (LACP_BOND_EXAMPLE)
2. Click Advanced and locate the Hardware Tab
3. Change Configure from Automatically to Manually
4. Change the MTU to Jumbo (9000) and click OK

5. Click Apply

6. Verify the MTU change via Terminal.

ifconfig bond(x) = x being the number assigned to the bond.

In this example:

ifconfig bond07. Verifying the MTU with the SANLink Utility

If you have any questions or concerns, please open a support request via https://support.promise.com

Using SANLink2 8Gb Fibre Channel to connect with VTrak A-Class may get "no export path" error

2014-04-23T17:08:32.0000000

I just added SANLink2 with my New Mac Pro to my VTrak A-Class SAN. I’m trying to mount the file system on the client and the client software tells me there is “no export path”.

Normally Vtrak A-Class will automatically LUN Map any new client workstation using SANLink2.
SANLink1 is fine as well as other Fibre Channel Host bus adapters. You will have to manually add SANLink2 to the VTrak A-Class LUN Mapping table using our CLI (command line interface).
The next driver release for SANLink2 will address the auto LUN Mapping with our GUI.

Meanwhile follow these steps:

1. Look up the SANLink2 WWPN from your system profiler “Fibre Channel”properties on your Mac Host:

Example:
FibreChannel Domain 0:
PortWorld Wide Name: 24:01:00:01:55:57:D4:76
Fibre Channel Domain 1:
Port World Wide Name: 24:02:00:01:55:57:D4:76

Example:
Open the Vtrak A-Class CLI: ssh administrator@10.81.49.132

Login using the default administrator and password

Lastlogin: Tue Apr 1 12:51:55 on ttys000
manojme6410:~ test$ ssh administrator@10.81.49.132
administrator@10.81.49.132's password:

-------------------------------------------------------------
Promise VTrak Command Line Interface (CLI) Utility
Version: 1.09.0000.18 Build Date: Mar 17, 2014
-------------------------------------------------------------

-------------------------------------------------------------
Type help or ? to display all the available commands
-------------------------------------------------------------

administrator@cli>

2. You want to add the WWPN of the SANLink2 to list of devices using the “initiator”command:

administrator@cli> ? initiator
initiator

Usage: initiator [-a <action>] [-i <Index>] [-c <Count>]
       initiator -a add [-i <Index>]-n <Name>
       initiator -a del -i <Index>

Summary:
   Initiator will display information about the current initiator list
   as well as the ability to add or delete an initiator.

Options:

-a <action>          Which action to perform.

list                (Default)Displays the current initiator list.

add                 Add an initiator to the list.

del                 Delete an initiator from the list.

-i <Index>           Used to specify the index of the initiator. For -a list
  0..2047            option,it is the starting index and may be used with -c
                     option.For other options, it is the specific index.
                     Caution:For -a add option, if the index specified is
                     already in use, the existing initiator name is
                     overwritten with new name.

-c <Count>           Used to specify the number of initiators to be listed.
                     Only used with -a list option.

-n <Name>            Used to specify the name of the initiator.
                     For an iSCSI host interface product, the name should be
                     the initiator's iSCSI name, e.g.
                       iqn.vendorcompany.com
                     Fora Fibre Channel host interface product, the name
                     should be the initiator's WWPN in hex format, e.g.
                       aa-bb-cc-dd-ee-ff-11-22
                     Fora SAS host interface product, the name
                     should be the initiator's SAS address in hex format,
                     e.g.aa-bb-cc-dd-ee-ff-11-22
                     For slot based lun mapping product, the first byte
                     is slot id. For example, for slot 2, the name is
                     02-00-00-00-00-00-00-00

Examples:
       initiator -i 1 -c 2
       initiator -a add -niqn.vendorcompany.com

Type " | more" at the end of each command, to display info page bypage.

Example:
administrator@cli> initiator -a add -n 24-01-00-01-55-57-D4-76

administrator@cli> initiator -a add -n 24-02-00-01-55-57-D4-76

NOTE: notice I replaced the “:” with “-”

Step 3. Check to make sure new “index” is present by typing “initiator” for the initiator in question:

administrator@cli> initiator

-------------------------------------------------------------------------------
Index: 0 Name: 24-01-00-01-55-57-d4-76

-------------------------------------------------------------------------------
Index: 1 Name: 24-02-00-01-55-57-d4-76

4. Next step is to look up the “lunmap” command by typing “? lunmap”

administrator@cli> ? lunmap
lunmap

Usage: lunmap [-a <action>] [-i <InitiatorId>] [-t <TargetId>][-r <CtrlId>]
              [-p<PortId>] [-c <Count>]

       lunmap -a addld -i <InitiatorId>[-l <LdIdList>] [-m <LunMap>]
       lunmap -a delld -i <InitiatorId>[-l <LdIdList>]
       lunmap -a add [-i <InitiatorId>]-n <Name> [-l <LdIdList>] [-m <LunMap>]
       lunmap -a del -i <InitiatorId>
       lunmap -a enable
       lunmap -a disable

Summary:
   lunmap command displays information about the current LUN mapping and
   masking (LMM) table information and enables you to add, modify,
   and delete LMM entries. LMM can be enabled or disabled.

Options:

-a <action>          Which action to perform.

list                (Default)Displays the current LMM table.

enable              Enables LMM.

disable             Disables LMM.

add                 Adds an LMM entry and its LUN maps to the table.

del                 Deletes an LMM entry from the table.

addld               Adds or modifies an LUN map for an existing LMM entry.

delld               Deletes a LUN map for an existing LMM entry.

mod                 Specifies LUN mapping type.
                     Please note that target based LUN mapping can only
                     be specified for iSCSI host interface.

-r <CtrlId>          Specifies the Ctrl ID for a port-based LMM entry.
                     Valid only for Fibre Channel host interface.

-p <PortId>          Specifies the Port ID for a port-based LMM entry.
                      Valid only for Fibre Channel host interface.

-i <InitiatorId>     Specifies the initiator ID for an initiator based
    0..2047           LMMentry. For -a list option, it is the starting index.
                     Maybe used with -c option.

-t <TargetId>        Specifies the target ID for target based LMM entry.
   0..2047           Valid only for iSCSI host interface.

-c <Count>           Specifies the number of LMM entries to be listed.
                     Only used with -a list option.

-n <Name>            Specifies the initiator name.
                     For iSCSI host interface, the name is the initiator's
                     iSCSIname, such as iqn.vendorcompany.com
                     For Fibre Channel host interface, the name is the
                     initiator's WWPN in hex format,
                     such as aa-bb-cc-dd-ee-ff-11-22
                     For SAS host interface, the name is the SAS address,
                     such as aa-bb-cc-dd-ee-ff-11-22
                     For slot-based LUN mapping, the first byte is the
                     slotID. Example for slot 2,
                     the name is 02-00-00-00-00-00-00-00

-l <Ld ID list>      Specifies the logical drive IDs.
   0..1023

-m <LUN map list>    Specifies the LUN mapping values.
   0..1023           Please check the maximum number of LUNs supported by
                     hostOS.

-s "<option>=<value>"
                     Specifies settings for LMM entry.
                     Modifies an LMM entry.
   type=
    initiator        For initiator-based LUN mapping.
    target           For target-based LUN mapping.
    port             For port-based LUN mapping.

Examples:
       lunmap -i 1 -c 2
       lunmap -a addld -i 1 -l 2 -m 2
       lunmap -a delld -i 1 -l 2
       lunmap -a enable
       lunmap -a add -n iqn.promise.com -l0,1 -m 0,1

Type " | more" at the end of each command, to display info page by page.

administrator@cli>

5. You want to map only the data LUNs and not expose your meta data LUN.
In my case I have two data LUNs I want to map. You can also issue the “logdrv”command
To view your existing LUNs on the VTrak A-Class
Example:

lunmap-a add -n 24-01-00-01-55-57-d4-76 -l 1,2 -m 1,2
lunmap -a add -n 24-02-00-01-55-57-d4-76 -l 1,2 -m 1,2

-------------------------------------------------------------------------------
Initiator 0 Name: 24-01-00-01-55-57-d4-76
===============================================================================
| LdId   LUN    | LdId   LUN    | LdId   LUN    | LdId   LUN    |
===============================================================================
| 1      1      | 2     2      |

-------------------------------------------------------------------------------
Initiator 1 Name: 24-02-00-01-55-57-d4-76
===============================================================================
| LdId   LUN    | LdId   LUN    | LdId   LUN    | LdId   LUN    |
===============================================================================
| 1      1      | 2     2      |

6. Proceed to mount the file system using the VTrak AClass Client software utility.

SANLink2 unable to detect all Fibre Channel Target Devices

2014-04-23T16:53:55.0000000

I’m using a SANLink2 8Gb Fibre Channel adapter to connect to my Fibre Channel SAN.

I’m not able to detect all the Fibre Channel SCSI Target Devices and see all my LUNs.

This is a known issue with the first release of the driver.

If you are interested in a beta driver, please contact Promise Technical Support via Phone : (408) 228-1500 or web support.promise.com.

A new point release is coming soon to the Promise website www.promise.com

I want to access my Pegasus R4/R6 connected to my SANLink. How do I access the LUNs across the Fiber Channel SAN?

2012-05-01T12:37:27.0000000

SANLink only functions as a Fiber Channel Initiator. Although it allows local access to Pegasus Direct Attached (DAS) to SAN Link, there are no plans at this point to change its implementation.

SANLink is geared to allow connectivity to Mac Book Pro, Mac Book Air, iMac and Mac Mini to a Fiber Channel SAN. Pegasus can be connected to SANLink as an “End Device”.

Recap:

Can Pegasus storage be accessible across the Fibre Channel SAN?

No, it can only be accessed locally on the respective host connected to Pegasus.