A800 interconnect down with e0a/e0b Fatal parity error (0x10)

Applies to

• AFF A800
• Dual 40/100G Ethernet T62100-MEZZ
• ONTAP 9

Issue

• After node power cycle, reboot, or upgrade, the system is in partial giveback with Interconnect status: "RDMA Interconnect is down"
• The storage failover status is: "Storage failover interconnect error. NVRAM log not synchronized. Disk inventory not exchanged"
• The console logs show: e0a/e0b:Fatal parity error (0x10)
• The ONTAP OS and the BMC, BIOS and T62100 firmware are updated and running in both nodes

EMS logs:

May 02 07:58:09 [node_name:netif.fatal.err:ALERT]: The network device in slot 0 encountered fatal error e0a/e0b.
May 02 07:58:09 [node_name:netif.fatal.err:ALERT]: The network device in slot 0 encountered fatal error e0a/e0b.
May 02 22:49:05 [node_name: kernel: netif.linkDown:info]: Ethernet e0a: Link down, check cable.
May 02 22:49:05 [node_name: kernel: netif.linkDown:info]: Ethernet e0b: Link down, check cable.
May 02 22:49:05 [node_name: intr: rlib.ifconfig.linkEvent:notice]: params: {'ifname': 'e0b', 'eventType': 'DOWN'}
May 02 22:49:05 -0800 [node_name: vifmgr: vifmgr.portdown:notice]: A link down event was received on node node_name, port e0a.
May 02 22:49:05 -0800 [node_name: nvmm_error: nvmm.mirror.offlined:debug]: params: {'mirror': 'HA_PARTNER'}
May 02 22:49:05 -0800 [node_name: vifmgr: vifmgr.clus.linkdown:EMERGENCY]: The cluster port e0a on node node_name has gone down unexpectedly.
May 02 23:00:00 -0800 [node_name: statd: ic.HAInterconnectDown:error]: HA interconnect: Interconnect down for 10 minutes: link0 down
May 02 23:00:00 -0800 [node_name: statd: callhome.hainterconnect.down:alert]: Call home for HA INTERCONNECT DOWN due to link0 down.