HP 4400 HP StorageWorks Fabric OS 6.4.1 Release Notes (5697-0760, November 201 - Page 54

Closed defect summary Solution Common Access Library Daemon (CALD) detects a large memory increase on the HP StorageWorks 2408 FCoE switch when running multiple SMIA clients over FCoE. Other Fabric OS-based products do not exhibit the same behavior when running the same scripts. Fixed in Fabric OS 6.4.0 HBA is seen in switchshow but not nsshow. When plugged two ports into a switch, both ports show logged in and online in switchshow, but the HBA device is missing from nsshow because the FLOGI response did not update PID properly. Fixed in Fabric OS 6.4.0 When small read requests (low I/O rate traffic) sent via FCIP tunnel, a 2 - 4 ms delay occurs between the time data arrives on the GE port until it is sent out from the FC port, leading to >10 ms response time. Fixed in Fabric OS 6.4.0 Workaround prior to upgrade: Do not run low I/O rate traffic.This occurs only when low I/O rate traffic is passing through the FCIP Tunnel. With a 2408 FCoE switch in Access Gateway mode, traffic stops on all host ports within a trunk when one of the hosts is disabled using the fcoe --disable command, or shutting that interface, or by pulling the cable on it. Fixed in Fabric OS 6.4.0 After upgrade to Fabric OS 6.2.x, the 400 MP router, or multiprotocol router blade observes a credit problem on virtual channel 0, leading to frame drop on FCIP and FC FastWrite, and the host being unable to see the target. Fixed in Fabric OS 6.4.0 Workaround prior to upgrade: Power On Reset of the router; or slotpower off/on of router blade, or Power On Reset the chassis. After a target port bounce, the FC FastWrite host cannot recognize the Fixed in Fabric OS 6.4.0 target due to cam setup problem on the backend port used by FC FastWrite. WWN Based PID Assignment assigning wrong PID to port in first posi- Fixed in Fabric OS 6.4.0 tion in wwnAddress -show PID binding list. Problem running traffic over ports after FCIP bounce. FCIP link recovers, Fixed in Fabric OS 6.4.0 but routes to FC ports do not. With SNMP severity set to 3, the DC SAN Backbone director sends out Fixed in Fabric OS 6.4.0 traps when a core blade is inserted, but not when it is removed. RTWR (Reliable Transport Write and Read) messages are logged when more than 16 dynamic paths are set up to remote domains on HP 8Gb switches. You may experience path loss and traffic interruption. Fixed in Fabric OS 6.4.0 Workaround prior to upgrade: Limit the number of dynamic paths on these platforms. If there are more than 16 dynamic paths already on these platforms, upgrade to the fixed release could be a cold recovery. When an HP 10/24 FCoE blade is installed in the highest number slot in a DC SAN Director or DC04 SAN Director, it will not send out any UAs (unsolicited advertisements), so FCoE traffic will not occur. Fixed in Fabric OS 6.4.0 Workaround prior to upgrade: Install FCOE10-24 to lower slot. 54