HP 4400 HP StorageWorks Fabric OS 6.4.1 Release Notes (5697-0760, November 201 - Page 77
and an HP DC SAN Director Multiprotocol Extension Blade., Management Applications DCFM
View all HP 4400 manuals
Add to My Manuals
Save this manual to your list of manuals |
Page 77 highlights
Closed defect summary Solution RPCD panic and subsequent switch cold recovery occurs due to internal memory synchronization and corruption issues when changing security settings through Web Tools while there is an active RPC connection from SMI-A. Fixed in Fabric OS 6.4.1. Workaround prior to upgrade: You can avoid this problem by not using your SMI based application until the code upgrade completes. Switch panics occur when simultaneously managed by both DCFM and Fixed in Fabric OS 6.4.1. SMI-A management applications. Web Linker causes active CP to reboot (initiate a cold recovery) when Fixed in Fabric OS 6.4.1. the standby CP is not responsive (stuck in a reboot loop). Due to a timing issue, hosts are unable to access targets for any opera- Fixed in Fabric OS 6.4.1. tional purposes when the number of FCIP circuit configurations on one side of the FCIP Tunnel does not match the other side. After a firmware download, host and target are unable to establish connectivity across a VEX link on an HP 1606 SAN Extension Switch and an HP DC SAN Director Multiprotocol Extension Blade. Fixed in Fabric OS 6.4.1. A Host times out its login and issues an ABTS after it is brought online when multiple PLOGIns to well known addresses are received back-toback before routes are established and there is no ACCept sent back by the management server. This is typcially seen with hosts that issue fast logins, such as in mainframe FICON environments. Fixed in Fabric OS 6.4.1. During a host reboot, the host may not come up when connected to HP Fixed in Fabric OS 6.4.1. StorageWorks 4/8 and 4/16 SAN Switches and 4Gb embedded switches. Compact flash failures are not always reported; only critical daemon failures are visible, which then trigger a failover (occasionally no failures are seen at all). The switch continues operating, and does not produce any logs. Fixed in Fabric OS 6.4.1. Web Linker processes terminate (core dump) and then restart when Management Applications (DCFM, etc.) query Top Talker information using CAL requests over HTTP. Fixed in Fabric OS 6.4.1. Host ports can experience slow performance after enabling and then disabling Top Talker. This is accompanied by class 3 frame rejects showing in the portlogdump. Fixed in Fabric OS 6.4.1. The switch is unable to reboot due to core files exceeding the size of the Compact Flash (CF), which corrupt the Fabric OS system files. Fixed in Fabric OS 6.4.1. A directional route problem in the fabric causes the switch to go into a Fixed in Fabric OS 6.4.1. disabled state without logging a reason. However, messages indicating that RTWR has reached max retries appear in the RASLOG. Switches in Access Gateway mode are not handling multi-sequence frames properly, which causes memory corruption and switch panic. Fixed in Fabric OS 6.4.1. On 4Gb switches, switch panics occur after an Information Unit (IU) allocation failure. Fixed in Fabric OS 6.4.1. HP StorageWorks Fabric OS 6.4.1 Release Notes 77