Sun Microsystems, Inc.  Sun System Handbook - ISO 3.4 June 2011 Internal/Partner Edition
   Home | Current Systems | Former STK Products | EOL Systems | Components | General Info | Search | Feedback

Asset ID: 1-72-1007648.1
Update Date:2009-02-12
Keywords:

Solution Type  Problem Resolution Sure

Solution  1007648.1 :   Sun StorEdge[TM] 3510 FC Array: Failure to Do a Full Fabric Login to a Switch  


Related Items
  • Sun Storage 3510 FC Array
  •  
Related Categories
  • GCS>Sun Microsystems>Storage - Disk>Modular Disk - 3xxx Arrays
  •  

PreviouslyPublishedAs
210588


Symptoms
In the fabric Storage Area Network (SAN) configuration, the
switches communicate with the Sun StorEdge[TM] 3510 FC array host
ports using a fabric point-to-point (F-port) mode. During
installation the 3510 fails to login to the switch in full
fabric mode.

Resolution
This problem results from incorrect configuration on the
Sun StorEdge 3510 array Several parameters must be correctly
configured for the array to correctly log in to a fabric.
Point-to-Point Configuration Guidelines
============================================================
Use the following guidelines when connecting to Fabric switches:
* The default mode is "Loop only."  You must change the Fibre Channel
Connection mode to "Point-to-point only" with the firmware application.
Caution:  If you keep the default loop mode and connect to a Fabric
switch,  the array automatically shifts to public loop mode.  As a
result, communication between the array and the switched Fabric
runs in  half duplex  (send or receive) instead of providing the full
duplex (send and receive) performance of point-to-point mode.
* Check the host IDs on all the channels to ensure that there is only one
ID per channel (on the primary controller or on the secondary controller)
for point-to-point mode.
When viewing the host IDs, there should be one primary controller ID
(PID) or one secondary controller ID (SID); the alternate port ID should
display N/A.  Proper point-to-point mode allows only one ID per
channel.
If you change to "Point-to-point only" and attempt to add a second ID to
the same controller, the controller does not allow you to add an ID to
the same controller and channel.  For example, if you have ID 40 on
CH 0 PID, and N/A on CH 0 SID, the controller does not allow you
to add another PID to CH 0.
Additionally, the controller displays a warning if the user is in
point-to-point mode and tries to add an ID to the same channel but on the
other controller.
A port bypass circuit connects the primary and secondary
controllers on each channel and functions as a mini-hub.  For example,
channel 0 of the primary controller is connected to channel 0 of the
secondary controller through this port bypass circuit.   This feature
allows for redundant connections to a logical drive that can, by
design, only belong to one controller at a time.   If you ignore this
warning and add an ID to the other controller, the redundant array
of independent disks (RAID) controller does not allow a login as a
Fabric-Loop (FL) port because this would be illegal in a
point-to-point configuration.
* Do not connect more than one port per channel on an array to a Fabric
switch.
For example, connect one switch port to either the primary or the
secondary channel 0, not to both.
In point-to-point mode, only one switch port is allowed per channel.
Connecting more than one port per channel to a switch can violate the
point-to-point topology of the channel, force two switch ports to
compete for an AL_PA value of 0 (which is reserved for loop to Fabric
attachment), or both.  This situation results from the port bypass
circuit, described previously
* Map the logical unit numbers (LUNs) and cable the channels to the
switches correctly.
With four host channels and four host IDs, you should load-balance the
host ID setup so that half the IDs are on the primary controller and half
the IDs are on the secondary controller. When setting up LUNs, map
each LUN to either two PIDs or two SIDs. The hosts are in turn
dual-pathed to the same two switched Fabrics.
When attaching the cables, for redundancy, make sure that the first cable
is connected to the upper port of one channel and the second cable is
connected to the lower port of another channel.
For example, to provide redundancy, map half the LUNs across Channel 0
(PID 40) and Channel 1 (PID 41), and then the other half of the LUNs
across Channel 4 (SID 50) and Channel 5 (SID 51).
Channel 0: PID 40 (assigned to the primary controller)
Channel 1: PID 41 (assigned to the primary controller)
Channel 4: SID 50 (assigned to the secondary controller)
Channel 5: SID 51 (assigned to the secondary controller)
Map the logical drives on the primary controller to channels 0 and 1 of
the primary controller.  Map the logical drives on the secondary
controller to channels 4 and 5 of the secondary controller.
Connect cables in the following manner:
From channel 0 (PID 40) and channel 4 (PID N/A) to switch 1.
From channel 5 (SID 51) and channel 1 (SID N/A) to switch 2.
In this way you will have full redundancy to your LUNS and fabric login
to the switches.


Additional Information
For a more detailed description, refer to the Sun StorEdge[TM] 3000
Family Installation, Operation, and Service Manual.

Product
Sun StorageTek 3510 FC Array

3510, switch, fabric, login
Previously Published As
75655

Change History
Date: 2004-05-13
User Name: c8840
Action: Approved
* Revised the title for wording and capitalization
* Spelled out the first occurence of accronyms
* Changed the unordered list from a numbered scheme to a bulleted scheme
* Inserted a heading in the Resolution section
* Fixed non-printable characters
* Re-worded some text
* Spelled out full name of product where appropriate
* Added trademark des

Attachments
This solution has no attachment
  Copyright © 2011 Sun Microsystems, Inc.  All rights reserved.
 Feedback