HP 1606 HP StorageWorks Fabric OS 6.3.1c Release Notes (5697-0509 September 2 - Page 84

When power cycling an HP StorageWorks 2408 FCoE switch, C25464/5465 errors can be observed due to logical port to FCoE virtual Fixed in Fabric OS 6.3.1 user port mapping error. Configuration back up on Access Gateway switch does not have the bottleneck configurations. Fixed in Fabric OS 6.3.1 ARB(FF) usage as fill words for 8 Gb connectivity does not abide by the Fibre Channel standards. Implement two new modes in addition to Fixed in Fabric OS 6.3.1 current mode 0 (idle) and mode 1 (arbff) to revise behavior. When portcfgdefault is issued on VE port, error messages are posted on DP (Data Plane/port blade) and CP: portcfgdefault 4/22: Port 182 NPIV Limit Set to 126! Unknown Error Fixed in Fabric OS 6.3.1 Workaround prior to upgrade: Do not issue the portCfgDefault command on 1606 extension SAN switch or multi-protocol extension blade VE ports. When Access Gateway switch is connecting to a secondary port on an 8 Gb or 4 Gb 48-port blade, if the primary port goes down, some frames may be wrongly routed to the Access Gateway switch's secondary port, and AG will respond to these frames with ACC rather than drop. Fixed in Fabric OS 6.3.1 Workaround prior to upgrade: Do not put AG switch at the secondary port on a 48-port blade. NPIV tape devices may not be seen by hosts due to name server not sending subsequent PLOGI ACC following a sequence where NPIV device sends LOGO before link down. Fixed in Fabric OS 6.3.1 After hafailover, the portlog parameters debug and Portlog events enable are not persistent and go back to default, or setting in previous reboot rather than the new setting before hafailover. Fixed in Fabric OS 6.3.1 After reboot of 2408 FCoE switch, a flood of internal RASLOG messages Fixed in Fabric OS 6.3.1 occur on FCoE/NPIV ports. No other functional impact. After upgrade from Fabric OS 6.2.x to 6.3.0b/c, then downgrade back to 6.2.0x, critical message BKSW-5002" swd_drv.c, line: Fixed in Fabric OS 6.3.1 1323 is observed. FCIP tunnel bounce observed during impairment test. Fixed in Fabric OS 6.3.1 Workaround prior to upgrade: Identify and fix cause of TCP network retransmissions. Reduce tunnel maximum data rate to a level that does not experience packet loss or corruption on the network. Some storage arrays support SPC-3 but indicate support for SPC-2 in standard inquiry data. This leads to the LUN to show up with a different LUN number on a different path, which causes the encryption LUN serial number (LSN) to be incorrect. Fixed in Fabric OS 6.3.1. This fix enables multipath support for encryption with the full set of supported HP storage arrays. During fabric initialization, it can take a long time (15 to 20 seconds) Fixed in Fabric OS 6.3.1 for a loop port to come up in interopmode 2/interopmode 3. Observed VE tunnel bounce after a hareboot on edge switch EX/VEXport (Integrated Routing) under heavy load. This is due to fspftimeout during hafailover resulting from neighbor switch disabling the port caused by the switch taking too long to recover. Fixed in Fabric OS 6.3.1 84