11-23-2009 09:19 AM
We are facing some problems using the Brocade 8gb switch (5470) which is inside the IBM Bladecenter chassis.
We have a chassis with 4 of these Brocade switches configured as Access Gateway mode with the latest firmware (6.2.0c equivalent for bladecenter). Each of the 14 blades have 4 HBA ports connecting to the 4 access gateways.
We have 2 uplinks trunked using N_port trunking to Brocade DCX. The F_port to N_Port mapping is set to Auto Balancing.
We find that when we reboot a blade the "enc out", "link fail" and "link loss" counter in porterrshow increase on MANY of the internal 14 ports. The OS running on the blade reports a "loss in path" and the ports are reported as "Marginal Ports" on the switch logs.
We are trying to find out why a reboot on one blade is causing the errors on the many of the other ports?
Where do we need to look next?