HP 4400 HP StorageWorks Fabric OS 6.4.1 Release Notes (5697-0760, November 201 - Page 56

ities with XRC Emulation IPLing systems with SDM LPARs.

Page 56 highlights

Closed defect summary When auditing is on for security class, a race condition can occur in raslogd, which leads to a switch panic or hafailover. Solution Fixed in Fabric OS 6.4.0 Workaround prior to upgrade: Disable auditing via: Auditcfg --disable. During a supportsave, Software VERIFY - Failed expres- Fixed in Fabric OS 6.4.0 sion: index < 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 reports the detection of termination of the Web Linker process. Web Linker automatically restarts, 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 adds 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 user 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. 56

  • 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
  • 75
  • 76
  • 77
  • 78
  • 79
  • 80

Solution
Closed defect summary
Fixed in Fabric OS 6.4.0
Workaround prior to upgrade:
Disable auditing via:
Auditcfg
--disable
.
When auditing is on for security class, a race condition can occur in
raslogd, which leads to a switch panic or
hafailover
.
Fixed in Fabric OS 6.4.0
During a
supportsave
,
Software VERIFY - Failed expres-
sion: 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
reports the detection of termination of the Web Linker process. Web
Linker automatically restarts, 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
adds 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 user will have to restart
the rekey individually for those LUNs for which the rekey sessions could
not be automatically started.
56