Cluster Network Degraded alerts reported multiple times due to errors on cluster ports

Applies to

• ONTAP 9
• Cluster Switch

Issue

• Following cluster network degraded autosupport notification is received multiple times:

HA Group Notification (CLUSTER NETWORK DEGRADED) ALERT

• Multiple errors are reported on cluster ports of all the nodes except for one cluster port as seen in RECEIVE section of IFSTAT output.
• You may see the cluster switch health as degraded in the system health alert show command output
• A performance impact could be seen if there is a high amount of CRCs
• Cluster ports with multiple errors (may see a combination of CRC err, Error symbol, Illegal symbol, etc):

clustershell::>system node run -node <nodename> -command ifstat -a

-- interface e0b (219 days, 5 hours, 16 minutes, 2 seconds) --
RECEIVE
 Frames/second:    2301  | Bytes/second:      3785k  | Errors/minute:          0
 Discards/minute:     0  | Total frames:       154g  | Total bytes:         176t
 Total errors:    43918  | Total discards:       65  | Multi/broadcast:    4452k
 No buffers:          0  | Non-primary u/c:       0  | L2 terminate:       14908
 Tag drop:            0  | Vlan tag drop:         0  | Vlan untag drop:        0
 Vlan forwards:       0  | CRC errors:        29328  | Runt frames:            0
 Fragment:            0  | Long frames:          65  | Jabber:                 0
 Error symbol:    29328  | Illegal symbol:    14590  | Bus overruns:           0
 Queue drop:          0  | Xon:                   0  | Xoff:                   0
 Jumbo:           5634k  | JMBuf RxFrames:     162g  | JMBuf DrvCopy:      27146

• Single cluster port without errors:

clustershell::>system node run -node <nodename> -command ifstat -a

-- interface e0b (219 days, 7 hours, 2 minutes, 24 seconds) --
RECEIVE
 Frames/second:    1092  | Bytes/second:       950k  | Errors/minute:          0
 Discards/minute:     0  | Total frames:     47631m  | Total bytes:         107t
 Total errors:        0  | Total discards:     1159  | Multi/broadcast:    4473k
 No buffers:       1087  | Non-primary u/c:       0  | L2 terminate:         302
 Tag drop:            0  | Vlan tag drop:         0  | Vlan untag drop:        0
 Vlan forwards:       0  | CRC errors:            0  | Runt frames:            0
 Fragment:            0  | Long frames:          50  | Jabber:                 0
 Error symbol:        0  | Illegal symbol:        0  | Bus overruns:          22
 Queue drop:          0  | Xon:                   0  | Xoff:                   0
 Jumbo:           2769m  | JMBuf RxFrames:        0  | JMBuf DrvCopy:          0

• Single cluster port reports low mW in ifconfig -vvv output: 

::> system node run -node <nodename> -command ifconfig -vvv
…
e0b: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 9000
      uuid: 0320a80b-caa3-11eb-b14a-d039ea306760
      ...
      RX: 0.06 mW (-12.13 dBm) TX: 0.55 mW (-2.59 dBm)

• You may see the follwing messages in EMS logs and health alerts generated

netif.linkErrors: Excessive link errors on network interface e2c. Might indicate a bad cable, switch port, or NIC, or that a
cable connector is not fully inserted in a socket. On a 10/100 port, might indicate a duplex mismatch.

• The following alert may be generated against the connected switch port of the only node port without receive CRC errors:

[?] Tue Nov 01 18:13:10 -0700 [node-01: mgwd: callhome.hm.alert.major:alert]: Call home for Health Monitor process cshm: ClusterIfInErrorsWarn_Alert[switch01(FOC123456789)/Ethernet1/9].