HP 4400 HP StorageWorks Fabric OS 6.4.0c Release Notes (5697-0703, September 2 - Page 55

ities with XRC Emulation IPLing systems with SDM LPARs.

Page 55 highlights

Closed defect summary Solution During a SupportSave, "Software VERIFY - Failed expression: index < Fixed in Fabric OS 6.4.0 DP_TABLE_MAX_ENTRIES, file" message is seen. Due to slow read track performance when XRC emulation is enabled, MVS I/O errors are generated when an invalid selective reset frame is sent to the controller, causing Abort Sequences later on. Fixed in Fabric OS 6.4.0 Workaround prior to upgrade: Disable FICON XRC Emulation. RNID frame type field from the host is corrupted internally when the host is connected over FCIP, and is rejected by the switch. This results in the switch being unable to get FICON CHPID online to VSM (Virtual Tape Subsystem) over an emulated or non-emulated VE tunnel. The E_Port between switches functions correctly. Fixed in Fabric OS 6.4.0 Workaround prior to upgrade: Disable all zoning (allow all) - this would avoid passing the frame up to be zone checked. IFCCs encountered when processing Read Channel Extender Capabilities with XRC Emulation (IPLing systems with SDM LPARs). Fixed in Fabric OS 6.4.0 Workaround prior to upgrade: Disable FICON XRC Emulation. Restore Master Key from recovery smart cards does not put EE Online Fixed in Fabric OS 6.4.0 when all EEs in the EG were zeroized. You must remove trial licenses before installing permanent licenses for the same features or switch disruption can occur. The fix will automatically remove/replace the trial license before adding the permanent license, avoiding any disruption. Fixed in Fabric OS 6.4.0 Negative value in core file causes invalid memory access. The console will report the detection of termination of the Weblinker process. Weblinker will automatically restart, but user may see a disruption in GUI access to the switch. Fixed in Fabric OS 6.4.0 For some reason, if the first tunnel, the first circuit is not numbered 0, Fixed in Fabric OS 6.4.0 portshow fciptunnel all -c will add the line with "16 0 unknown Disable ----- 0s 0.00 0.00 0 0/0 0". Bad zone config error message is given, when ICL ports displayed in the zone admin module are included in the TI zone while creating it, for an IM2 logical switch. Fixed in Fabric OS 6.4.0 Rekey sessions supposed to be pending cannot be persisted when the EE hosting node fails over. If HA Cluster failover occurs before keys are successfully created for all LUNs under rekey, the customer will have to restart the rekey individually for those LUNs for which the rekey sessions could not be automatically started. Fixed in Fabric OS 6.4.0 Workaround prior to upgrade: Start the rekey sessions for LUNs on individual basis. In a FICON environment, CUP incorrectly handles problem when sent Unsol Alert Status 0x82, and receives a Selective reset in response. Continues to retry the Unsol Alert on the same LP that received the Sel Reset. After Fabric OS Upgrade from 6.2.0e to 6.3.0b user started seeing non stop FICU-1015 error messages on the switches and CUP appears to not be working. User did hafailover one of the switches, and error messages stopped. Fixed in Fabric OS 6.4.0 After removal and insertion of the HP StorageWorks Encryption blade, the lscfg --show command will show ASIC not ready errors and FW5002 will be observed in the internal raslog. Fixed in Fabric OS 6.4.0 HP StorageWorks Fabric OS 6.4.0c Release Notes 55

  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18
  • 19
  • 20
  • 21
  • 22
  • 23
  • 24
  • 25
  • 26
  • 27
  • 28
  • 29
  • 30
  • 31
  • 32
  • 33
  • 34
  • 35
  • 36
  • 37
  • 38
  • 39
  • 40
  • 41
  • 42
  • 43
  • 44
  • 45
  • 46
  • 47
  • 48
  • 49
  • 50
  • 51
  • 52
  • 53
  • 54
  • 55
  • 56
  • 57
  • 58
  • 59
  • 60
  • 61
  • 62
  • 63
  • 64
  • 65
  • 66
  • 67
  • 68
  • 69
  • 70
  • 71
  • 72
  • 73
  • 74

Solution
Closed defect summary
Fixed in Fabric OS 6.4.0
During a SupportSave,
Software VERIFY - Failed expression: index <
DP_TABLE_MAX_ENTRIES, file" message is seen.
Fixed in Fabric OS 6.4.0
Workaround prior to upgrade:
Disable FICON XRC Emulation.
Due to slow read track performance when XRC emulation is enabled,
MVS I/O errors are generated when an invalid selective reset frame is
sent to the controller, causing Abort Sequences later on.
Fixed in Fabric OS 6.4.0
Workaround prior to upgrade:
Disable all zoning (allow all) - this
would avoid passing the frame up
to be zone checked.
RNID frame type field from the host is corrupted internally when the
host is connected over FCIP, and is rejected by the switch. This results
in the switch being unable to get FICON CHPID online to VSM (Virtual
Tape Subsystem) over an emulated or non-emulated VE tunnel. The
E_Port between switches functions correctly.
Fixed in Fabric OS 6.4.0
Workaround prior to upgrade:
Disable FICON XRC Emulation.
IFCCs encountered when processing Read Channel Extender Capabil-
ities with XRC Emulation (IPLing systems with SDM LPARs).
Fixed in Fabric OS 6.4.0
Restore Master Key from recovery smart cards does not put EE Online
when all EEs in the EG were zeroized.
Fixed in Fabric OS 6.4.0
You must remove trial licenses before installing permanent licenses for
the same features or switch disruption can occur. The fix will automatic-
ally remove/replace the trial license before adding the permanent li-
cense, avoiding any disruption.
Fixed in Fabric OS 6.4.0
Negative value in core file causes invalid memory access. The console
will report the detection of termination of the Weblinker process.
Weblinker will automatically restart, but user may see a disruption in
GUI access to the switch.
Fixed in Fabric OS 6.4.0
For some reason, if the first tunnel, the first circuit is not numbered 0,
portshow fciptunnel all -c will add the line with "16 0 unknown Disable
----- 0s 0.00 0.00 0 0/0 0".
Fixed in Fabric OS 6.4.0
Bad zone config error message is given, when ICL ports displayed in
the zone admin module are included in the TI zone while creating it,
for an IM2 logical switch.
Fixed in Fabric OS 6.4.0
Workaround prior to upgrade:
Start the rekey sessions for LUNs on
individual basis.
Rekey sessions supposed to be pending cannot be persisted when the
EE hosting node fails over. If HA Cluster failover occurs before keys are
successfully created for all LUNs under rekey, the customer will have to
restart the rekey individually for those LUNs for which the rekey sessions
could not be automatically started.
Fixed in Fabric OS 6.4.0
In a FICON environment, CUP incorrectly handles problem when sent
Unsol Alert Status 0x82, and receives a Selective reset in response.
Continues to retry the Unsol Alert on the same LP that received the Sel
Reset. After Fabric OS Upgrade from 6.2.0e to 6.3.0b user started
seeing non stop FICU-1015 error messages on the switches and CUP
appears to not be working. User did hafailover one of the switches,
and error messages stopped.
Fixed in Fabric OS 6.4.0
After removal and insertion of the HP StorageWorks Encryption blade,
the lscfg --show command will show ASIC not ready errors and FW-
5002 will be observed in the internal raslog.
HP StorageWorks Fabric OS 6.4.0c Release Notes
55