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

Table 13 Fabric OS 7.0.1 closed defects (continued)
SolutionClosed fixes summary
license, non-10 Gbps ports do require the Extended Fabric license for this
operation.
On the HP SN6000B 16 Gbps Fabric Switch, when using the CLI for 10
Gbps ports, it is not possible to configure anything but the default Extended
Fabric modes without an Extended Fabric license present
On the HP SN8000B Directors, when using the CLI, the Extended Fabric
configuration behaves as expected for licensed 10 Gbps ports (no Extended
Fabric license is required for any extended distance modes).
Workaround prior to upgrade: Toggle the
port.
There may be D_Port failures when more than 4 ports are initiated
simultaneously.
Fixed in Fabric OS 7.0.1.
Fixed in Fabric OS 7.0.1.A switch reboot or an HA Failover may occur when the Admin Domain
(AD) database is not in sync and there is a port offline event on local/remote
switch.
Workaround prior to upgrade: Avoid out
of band GXX_ID queries (such as
In an SMI configuration, multiple name server panics may occur in
conjunction with applications that perform out of band GXX_ID queries
(such as GCS_ID). GCS_ID). Do not perform queries with the
ID of the device that is part of a frame
redirection via RPC.
Fixed in Fabric OS 7.0.1.
Fixed in Fabric OS 7.0.1.The early warning message for loss credits does not display for EX_Ports.
Performance problems may occur due to loss of credits without notification.
Fixed in Fabric OS 7.0.1.IDLE is not being sent after an NOS. This problem is manifested by one
side of the ISL becoming a G_Port while the peer remains an E_Port.
Fixed in Fabric OS 7.0.1.Third party applications may report a migration failure between servers
because Access Gateway does not properly clean up the frame exchange
if FLOGI/FDISC comes in from different ports together with the same
exchange ID (OX-ID).
Workaround prior to upgrade: Remove
zone files from the switch and reboot.
Users are unable to remove configured Admin Domain when defzone -
allacess is set since it is an invalid configuration.
Fixed in Fabric OS 7.0.1.
Fixed in Fabric OS 7.0.1.Directors may experience HA Failover and switches may reboot during a
code upgrade. This is caused by a kernel panic while doing zone check
when an F_Port trunk goes down and stays offline.
Fixed in Fabric OS 7.0.1.FabricWatch Port Class is not monitoring ICL ports. ICL Ports do not report
any threshold warnings and are not fenced for Port Class attributes.
Fixed in Fabric OS 7.0.1.The Compact Flash may become full on an encryption switch after a crash
and reboot. This is caused by a large corefile being saved to the Compact
Flash after the crash.
Workaround prior to upgrade: Disable
and then Enable all ICL ports.
After downgrading from Fabric OS 7.0.0a or 7.0.0b to Fabric OS 6.4.x
on an HP DC SAN Backbone Director or a DC04 SAN Director, a route
Fixed in Fabric OS 7.0.1.
rebalance or recalculation causes some ICL paths to be blocked. Connectivity
between devices is lost after a route change.
Fixed in Fabric OS 7.0.1.It is not possible to create more than three tunnels with crossport circuit
configuration on an HP DC SAN Director Multiprotocol Extension Blade in
a chassis with multiple HP DC SAN Director Multiprotocol Extension Blades.
Workaround prior to upgrade: Reseating
the QSFPs may correct the issue.
ICL Ports on the HP SN8000B SAN Directors may show Mod_Val on 1 out
of 4 ports on a QSFP. The event log indicates a bad QSFP requiring reseat
or replacement.
Fixed in Fabric OS 7.0.1.
50