HP StorageWorks 8/80 HP StorageWorks Fabric OS 6.1.2a release notes (5697-0350 - Page 30

Fabric OS 6.1.2a fixes Table 15 lists defects closed in the Fabric OS 6.1.2a firmware release. Table 15 Fabric OS 6.1.2a closed defects Closed defect summary Solution Name Server daemon (nsd) panic occurs after enabling frame redirection with redirection (RD) zone present in the fabric. A cfgsave causes zoning to push the effective zone database to the Name Server daemon, and cfgenable causes the push to recur. If the cfgenable follows the cfgsave too closely, it is possible the two database pushes may trigger an nsd panic. Workaround prior to upgrade: Delay executing cfgenable after cfgsave when there is a frame RD zone present. Fixed in Fabric OS 6.1.2a Host with Fibre Channel FastWrite (FCFW) enabled has problems with port disable/port enable of HBA ports, causing a server to Fixed in Fabric OS 6.1.2a hang after 400 MP Router reboot. After upgrade to Fabric OS 6.1.2, event manager daemon (evmd) segmentation fault on standby CP occurs on DC SAN Director when using Fabric Manager. Workaround prior to upgrade: Disable Fabric Manager and use DCFM. Fixed in Fabric OS 6.1.2a MP router blade fails to load firmware, indicating a firmware download failure in the DC SAN Director. Workaround prior to upgrade: Perform slot power off/slot power on on the blade that timed out and failed the download the firmware. Fixed in Fabric OS 6.1.2a FCFW configuration is not properly installed after repeated cable pulls and reinsertions. This can cause servers to receive responses for an Fixed in Fabric OS 6.1.2a OXID before it has granted transfer sequence initiative (TSI). FCFW corrupts buffer pool due to a Turbo Write Block (TWB) allocation failure on Proxy Initiator. Fixed in Fabric OS 6.1.2a The sfpshow command displays incorrect values for current, RX power, and TX power for any port on the switch as a result of SFP polling Fixed in Fabric OS 6.1.2a hang when attempting to poll a disabled port during hafailover. The supportsave was updated with Fabric OS 6.1.2 to collect more data. When data is read from disabled ports, a parity error may be logged for those ports. No errors will occur on enabled ports or after a disabled port is enabled. Fixed in Fabric OS 6.1.2a When QoS is enabled by default and is not utilized, SAN performance drops. QoS may be activated automatically on an ISL after a bounce of the link, which can cause removal of an ISL link from a trunk group or failure of a long-distance link. Fixed in Fabric OS 6.1.2a 30