888 Too many Fibre Channel logins between nodes.

Explanation
The system determined that the user zoned the fabric such that this node received more than 16
unmasked logins that originated from another node or node canister. The other node or node canister
can be any non-service-mode node or canister in the local clustered system or in a remote clustered
system with a partnership. An unmasked login is from a port whose corresponding bit in the FC port
mask is '1'. If the error is raised against a node in the local clustered system, then it is the
local FC port mask that is applied. If the error is raised against a node in a remote clustered
system, then it is the partner FC port masks from both clustered systems that apply. 
More than 16 logins is not a supported configuration as it increases internode communication and
can affect bandwidth and performance. For example, if node A has eight ports and node B has eight
ports where the nodes are in different clustered systems, if node A has a partner FC port mask of
00000011 and node B has a partner FC port mask of 11000000 there are four unmasked logins possible
(1,7 1,8 2,7 2,8). Fabric zoning can be used to reduce this amount further, that is, if node B port
8 is removed from the zone there are only 2 (1,7 and 2,7). The combination of masks and zoning must
leave 16 or fewer possible logins.
Note: This count includes both FC and Fibre Channel over Ethernet (FCoE) logins. The login count
does not include masked ports.
When this event is logged, the clustered system ID and node ID of the first node whose logins
exceed this limit on the local node are reported, as well as the WWNN of said node. If logins
change, the error is automatically fixed and another error is logged if appropriate (this error
might or might not choose the same node to report in the sense data if the same node is still over
the maximum allowed). 
A text string is displayed that reports the following data: 

WWNN of the other node

Clustered system ID of other node

Arbitrary node ID of one other node that is logged into this node.
(node ID as it appears in lsnode) 

User response
The error is resolved by either reconfiguring the system to change which type of connection is
allowed on a port, or by changing the SAN fabric configuration so ports are not in the same zone. A
combination of both options can be used.
The system
reconfiguration is to change the Fibre Channel ports mask to reduce
which ports can be used for internode communication.
Modify the local Fibre Channel port mask if the clustered system ID reported matches the
clustered system ID of the node that logged the error.
Modify the partner Fibre Channel port mask if the clustered system ID reported does not match the
clustered system ID of the node that logged the error. The partner Fibre Channel port mask might
need to be changed for one or both clustered systems.
SAN fabric configuration is set by using the switch configuration utilities.
Use the lsfabric command to view
the current number of logins between nodes.
Possible Cause-FRUs or other cause: 
None

Service error code: 1801