Sun Microsystems StorEdge 3000 Series, Installation, Operation And Service Manual

Page: 90 / 318

5-12 Sun StorEdge 3000 Family Installation, Operation, and Service Manual • March 2004
Important rules to remember are:
■
A single logical drive can be mapped to only one controller, either the primary
controller or the secondary controller.
■
In a point-to-point configuration, only one host ID per channel is allowed. The
host ID can be assigned to the primary controller and be a PID, or it can be
assigned to the secondary controller and be a SID.
■
If you have two switches and set up multipathing (to keep all logical drive
connections operational for any switch failure or the removal of any I/O
controller module), ensure that each logical drive is mapped to two ports, one on
each I/O controller module, and on two channels. The cables from the two ports
mapped to each logical drive must be cabled to two separate switches. See
FIGURE 5-1
for an example of this configuration.
FIGURE 5-1
shows the channel numbers (0, 1, 4, and 5) of each host port and the host
ID for each channel. N/A means that the port does not have a second ID
assignment. The primary controller is in the top I/O controller module, and the
secondary controller is in the bottom I/O controller module.
The dashed lines between two ports indicate a port bypass circuit that functions as a
mini-hub and has the following advantages:
■
The port bypass circuit on each channel connects the upper and lower ports on
the same channel and provides access to both controllers at the same time.
■
If there are two host connections to the upper and lower ports on Channel 0,
and one host connection is removed, the other host connection remains
operational.
■
Therefore, if you have a redundant multipathing configuration in which you
have two host connections to each logical drive and one connection fails, the
remaining path maintains a connection to the logical drive.
In
FIGURE 5-1
, with multipathing software to reroute the data paths, each logical
drive remains fully operational when the following conditions occur:
■
One switch fails or is disconnected, and the logical drive is routed to the second
switch. For example, if switch 0 fails, switch 1 automatically accesses logical drive
0 through the cabling to the lower port on PID 41.
■
One I/O controller module fails, and all the host IDs for that controller are
reassigned (moved) to the second 1/O controller module. For example, if the
upper I/O controller module is removed, host IDs 40 and 41 are automatically
moved to the lower module and are managed by the second controller.
■
An I/O controller module fails or one cable is removed from an I/O controller
module, and all I/O traffic to the disconnected channel is rerouted through the
second port/host LUN assigned to the logical drive. For example, if you remove
the cable to channel 4, the data path for logical drive 1 switches to the port on
channel 5.