HP B-series Fabric OS 7.0.2d Release Notes (5697-2822, August 2013--includes all 7.0.x versions)

1
All ports from a 48 port card must be assigned to a logical switch with zero-based addressing configured. They must
be in a logical switch other than the default switch.
2
Supported only in a logical switch that will not have any FICON traffic. This is for FCP traffic only.
Any blade not listed in Table 8 (page 37) above is not supported in a chassis that carries FICON
traffic.
Fabric OS 7.0.0 fixes
Table 9 (page 38) lists defects closed in the Fabric OS 7.0.0 firmware release.
Table 9 Fabric OS 7.0.0 closed defects
SolutionClosed fixes summary
Workaround prior to upgrade: Before HA
failover starts, make sure the response trap
Removing an active CP from a fully populated HP StorageWorks 4/256
SAN Director switch (which is 384 F_Ports) causes some FC_Port blades to
reinitialize and possible frame drops. installed on the ports have been cleaned
up.
Fixed in Fabric OS 7.0.0.
Fixed in Fabric OS 7.0.0Stress on the fabric involving slot power cycling of all the blades in sequence
with CP HA Failover results in CP panic and termination of process fwd.
Workaround prior to upgrade: Remove
the Native-VLAN from the tagged list
before changing it.
After changing native VLAN on a converged mode port, the HP 2408 FCoE
Switch reports VLAN has changed, but traffic capture shows that both old
and new VLANs are now native VLANs on that port.
Fixed in Fabric OS 7.0.0.
Fixed in Fabric OS 7.0.0.Scalability stress may cause NSD to panic and crash on a HP StorageWorks
4/32 SAN Switch when enabling zoning with Top Talkers enabled.
Fixed in Fabric OS 7.0.0.In a FICON environment, Port faults are logged continuously when
switchenable is followed directly by fmsenable where fmsenable is being
invoked before the F_Ports have had a chance to come online.
Fixed in Fabric OS 7.0.0.Random errors occur while performing Access Gateway F-N port mapping
operations from WebTools.
Workaround prior to upgrade: Create
logical switches with SMIA disabled.
EVMD panics due to message queue full on standby CP during logical
switch creation. Customers may experience HA Failover, but there is no
disruption in functionality when this occurs.
Fixed in Fabric OS 7.0.0.
Workaround prior to upgrade: Perform a
switch disable/enable on switches where
the targets are attached.
In a large scale fabric with multiple VEX_Ports or non-trunked EX_Ports to
the same edge fabric, an FCR device import processing logic is duplicated
for each port and takes a long time to complete for all devices. This may
Fixed in Fabric OS 7.0.0.
cause delays in hosts discovering all targets over the IFLs and devices could
be stuck in the initialization state after the switchdisable/enable transition.
Fixed in Fabric OS 7.0.0.When many devices are attempting to login simultaneously, FDISCs on
NPIV devices may be rejected for a period of time due to lack of internal
resources.
Fixed in Fabric OS 7.0.0.In an encryption environment, boot over SAN may fail when there is no
connection to the key vault.
Workaround prior to upgrade: Reboot the
switch.
Proxy device showing as imported in fcrproxydevshow but absent from
nsallshow.
Fixed in Fabric OS 7.0.0.
Workaround prior to upgrade: Switch
offline/online or restart AG.
Access Gateway is not displayed by the DCFM client when utilizing an
EOS seed switch.
Fixed in Fabric OS 7.0.0.
Fixed in Fabric OS 7.0.0.Adding a new ISL or performing other events that trigger route change may
cause connectivity between servers and devices to bel ost.
38