HP Fabric OS 6.2.2f Release Notes (5697-1756, February 2012)

Table 13 Fabric OS 6.2.2b closed defects (continued)
SolutionClosed defect summary
Fixed in Fabric OS 6.2.2bNo cascaded FICON CUP connectivity on 4/256 Director with 8Gb blade
when E_Port and F_Port are on different ASICs on the blade.
Fixed in Fabric OS 6.2.2b (New
parameter added to Fabric OS 6.2.2b to
Under heavy CPU load, SFP laser fault condition may cause links to reset.
allow adjustment of SFP polling interval
to avoid laser faults.)
Fixed in Fabric OS 6.2.2bIn a very rare circumstance, a configuration file for dynamic area mapping
gets corrupted after firmwarecommit, leading to both CPs failing and
switch not booting up.
Workaround prior to update: Remove the
port2Bound file in /etc/fobos/
port2Bound in both active and standby
CPs, and in both partitions on both CPs,
then reboot the switch.
Fixed in Fabric OS 6.2.2bSwitch can panic or hafailover when detecting termination of daemons
due to bad EOF frames being forwarded to the CPU for processing.
Fixed in Fabric OS 6.2.2b
Workaround prior to update:Reboot the
switch again.
CP fails to boot on DC SAN director, DC04 SAN director, and 8/80 SAN
switch on power up or during hafailover and CF access error: HDA:
drive not ready message may be seen on the console.
Fixed in Fabric OS 6.2.2bPerforming multiple configdownload operations through DCFM or CLI
can trigger memory issues with webd leading to restart or switch panic.
Fixed in Fabric OS 6.2.2bAfter code upgrade to Fabric OS 6.2.x on 4Gb embedded switches, there
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
SolutionClosed defect summary
Fixed in Fabric OS 6.2.2c.
Workaround prior to update:
8 Gb Directors with an 8 Gb blade in slot 1 or 8 Gb
switches can see a frame drop in the fabric when
Avoid configuring an EX_Port or a VEx_Port near an F/E_Port
with an EGID of 0. To identify a port with an EGID of 0, login
Ex/VEx ports are configured in the same ASIC as an
F/E port with EGID 0, leading to either the host not
seeing a target, or poor performance in the fabric.
as root and run bladeportmap [1]. The first port where Upt
This issue can be identified by running sloterrshow
is not -1 uses EGID 0. Once the problem occurs, make the
1 on directors, or sloterrshow on the switches,
F/E_Port into an EX_Port and then configure it back to an
and the type6_miss counter should continuously
increment during traffic load on backend port/E_Port.
F/E_Port to fix the problem port by executing the following
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.
Fixed in Fabric OS 6.2.2c.
Workaround prior to update:
After upgrading Fabric OS from 6.1.x to 6.2.x on 4
Gb switches or director, the switch or director may
Manually ensure that http.enabled matches with SSL certificate
by configshow http.enabled as admin user. If it's
get into a perpetual reboot condition if the http
configuration is invalid (inconsistency between SSL
certificate and http configuration).
inconsistent with SSL Certificate requirement, execute
46