HP Brocade 8/12c HP Fabric OS 6.2.2f Release Notes (5697-1756, February 2012) - Page 46

Table 13 Fabric OS 6.2.2b closed defects (continued) Closed defect summary Solution No cascaded FICON CUP connectivity on 4/256 Director with 8Gb blade Fixed in Fabric OS 6.2.2b when E_Port and F_Port are on different ASICs on the blade. Under heavy CPU load, SFP laser fault condition may cause links to reset. Fixed in Fabric OS 6.2.2b (New parameter added to Fabric OS 6.2.2b to allow adjustment of SFP polling interval to avoid laser faults.) In a very rare circumstance, a configuration file for dynamic area mapping Fixed in Fabric OS 6.2.2b gets corrupted after firmwarecommit, leading to both CPs failing and Workaround prior to update: Remove the switch not booting up. port2Bound file in /etc/fobos/ port2Bound in both active and standby CPs, and in both partitions on both CPs, then reboot the switch. Switch can panic or hafailover when detecting termination of daemons Fixed in Fabric OS 6.2.2b due to bad EOF frames being forwarded to the CPU for processing. CP fails to boot on DC SAN director, DC04 SAN director, and 8/80 SAN Fixed in Fabric OS 6.2.2b switch on power up or during hafailover and CF access error: HDA: Workaround prior to update:Reboot the drive not ready message may be seen on the console. switch again. Performing multiple configdownload operations through DCFM or CLI Fixed in Fabric OS 6.2.2b can trigger memory issues with webd leading to restart or switch panic. After code upgrade to Fabric OS 6.2.x on 4Gb embedded switches, there Fixed in Fabric OS 6.2.2b is not enough room left in compact flash for the switch to operate adequately. If there is a core dump, switch compact flash could get full, resulting in switch not being able to boot up due to a corrupted file system, etc. Fabric OS 6.2.2c fixes Table 14 (page 46) lists defects closed in the Fabric OS 6.2.2c firmware release. Table 14 Fabric OS 6.2.2c closed defects Closed defect summary Solution 8 Gb Directors with an 8 Gb blade in slot 1 or 8 Gb Fixed in Fabric OS 6.2.2c. switches can see a frame drop in the fabric when Workaround prior to update: Ex/VEx ports are configured in the same ASIC as an Avoid configuring an EX_Port or a VEx_Port near an F/E_Port F/E port with EGID 0, leading to either the host not seeing a target, or poor performance in the fabric. This issue can be identified by running sloterrshow 1 on directors, or sloterrshow on the switches, with an EGID of 0. To identify a port with an EGID of 0, login as root and run bladeportmap [1]. The first port where Upt is not -1 uses EGID 0. Once the problem occurs, make the F/E_Port into an EX_Port and then configure it back to an and the type6_miss counter should continuously F/E_Port to fix the problem port by executing the following increment during traffic load on backend port/E_Port. instructions: 1. Disable the F/E_Port. 2. Configure the F/E_Port as an EX_Port for a valid FID. 3. Connect a link between the FID and the EX_Port (IFL). 4. Enable the port. 5. Verify that the link is online, then disable the port. 6. Disable the EX_Port configuration. 7. Connect the host/ISL back to the F/E_Port. 8. Enable the port. After upgrading Fabric OS from 6.1.x to 6.2.x on 4 Gb switches or director, the switch or director may get into a perpetual reboot condition if the http configuration is invalid (inconsistency between SSL certificate and http configuration). Fixed in Fabric OS 6.2.2c. Workaround prior to update: Manually ensure that http.enabled matches with SSL certificate by configshow http.enabled as admin user. If it's inconsistent with SSL Certificate requirement, execute 46