HP Cisco Nexus 5000 Cisco Nexus 5000 Series Release Notes Release 4.1(3)N1(1) - Page 11

Symptom, Workaround, show interface fex

Page 11 highlights

Caveats Send documentation comments to [email protected] • CSCsx60187 Symptom: Configure duplicate IP address for multiple SVI interfaces. Workaround: Do not configure two or more SVIs with the same IP address. • CSCsx80279 Symptom: When traffic is sent at line rate as a single burst, all addresses are not learned when egress interfaces are FEX facing ports. This problem does not occur if sustained traffic is sent for more than 0.4 seconds. Workaround: Resending the unlearned MAC addresses would render them relearned. • CSCsy99816 Symptom: When a Cisco Nexus 2000 Fabric Extender is already online and a fabric interface that is not part of a port channel is configured with a serial number that does not match that of the FEX, the fabric interface will be brought down, and show interface fex does not display the reason for being down. Workaround: None. This is a display issue. • CSCsy02439 Symptom: Under some circumstances, the FC MAC driver displays the following error message: %KERN-3-SYSTEM_MSG: FCP_ERRFCP_PORT: gat_fcp_isr_ip_fcmac_sync_intr@424, jiffies = 0x1c422:Unknown intr src_id 41 - kernel The error message is when an unused interrupt in the MAC fires. The error message does not indicate any functional problem. Workaround: None • CSCta78541 Symptom: VSH crashes and generate a core when you do the following: attach fex 102 show system error-id list The Cisco Nexus 2000 Fabric Extender does not reload and the core does not impact the functionality of the Fabric Extender. Workaround: None • CSCta74119 Symptom: DCBX fails to converge between a Nexus 5000 switch running 4.1(3)N1(1) and another Nexus 5000 switch running a 4.0(1a)N based release. As a result, priority-flow-control does not work on this link. Workaround: None. • CSCta85268 Symptom: When you downgrade from release 4.1(3)N1(1) to 4.0(1a)N-based releases using Fabric Manager or when you upgrade from a 4.0(1a)N-based release to 4.1(3)N1(1) using Fabric Manager error messages such as the following are generated: 2009 Jul 19 15:32:55 N5K-C snmpd: SNMP Operation (GET) timed out. reqId (1448314586) errno (62) on iso.3.6.1.4.1.9.9.360.1.1.4.2.0 2009 Jul 19 15:32:56 N5K-C snmpd: SNMP Operation (GET) failed. Reason:13 reqId (1448314586) errno (0) error index (1) OL-16601-01 Cisco Nexus 5000 Series and Cisco Nexus 2000 Series Release Notes, Release 4.1(3)N1(1) 11

  • 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

Send documentation comments to [email protected]
11
Cisco Nexus 5000 Series and Cisco Nexus 2000 Series Release Notes, Release 4.1(3)N1(1)
OL-16601-01
Caveats
CSCsx60187
Symptom
: Configure duplicate IP address for multiple SVI interfaces.
Workaround:
Do not configure two or more SVIs with the same IP address.
CSCsx80279
Symptom
: When traffic is sent at line rate as a single burst, all addresses are not learned when egress
interfaces are FEX facing ports. This problem does not occur if sustained traffic is sent for more than
0.4 seconds.
Workaround:
Resending the unlearned MAC addresses would render them relearned.
CSCsy99816
Symptom: When a Cisco Nexus 2000 Fabric Extender is already online and a fabric interface that is
not part of a port channel is configured with a serial number that does not match that of the FEX,
the fabric interface will be brought down, and
show interface fex
does not display the reason for
being down.
Workaround: None. This is a display issue.
CSCsy02439
Symptom
: Under some circumstances, the FC MAC driver displays the following error message:
%KERN-3-SYSTEM_MSG: FCP_ERRFCP_PORT: gat_fcp_isr_ip_fcmac_sync_intr@424, jiffies =
0x1c422:Unknown intr src_id 41 - kernel
The error message is when an unused interrupt in the MAC fires. The error message does not
indicate any functional problem.
Workaround:
None
CSCta78541
Symptom
:
VSH crashes and generate a core when you do the following:
attach fex 102
show system error-id list
The Cisco Nexus 2000 Fabric Extender does not reload and the core does not impact the
functionality of the Fabric Extender.
Workaround:
None
CSCta74119
Symptom
: DCBX fails to converge between a Nexus 5000 switch running 4.1(3)N1(1) and another
Nexus 5000 switch running a 4.0(1a)N based release. As a result,
priority-flow-control does not
work on this link.
Workaround:
None.
CSCta85268
Symptom
: When you downgrade from release 4.1(3)N1(1) to 4.0(1a)N-based releases using Fabric
Manager or when you upgrade from a 4.0(1a)N-based release to 4.1(3)N1(1) using Fabric Manager
error messages such as the following are generated:
2009 Jul 19 15:32:55 N5K-C snmpd: SNMP Operation (GET) timed out. reqId
(1448314586) errno (62) on iso.3.6.1.4.1.9.9.360.1.1.4.2.0
2009 Jul 19 15:32:56 N5K-C snmpd: SNMP Operation (GET) failed. Reason:13 reqId
(1448314586) errno (0) error index (1)