CX6カードを使用したクラスタポートダウン
環境
- ONTAP 9
- FAS/AFF システム
- CX6 PSIDカード
問題
- クラスタポートがダウンしました
e4a MAC Address: a0:b1:c2:d3:e4:f5 (auto-unknown-fd-down)
QSFP Vendor: AVAGO
QSFP Part Number: 332-00389
QSFP Serial Number: QM060143
- イベントログから、ポートe4aがフラッピングしていました
[Node-01: kernel: netif.linkDown:info]: Ethernet e4a: Link down, check cable.
[Node-01: vifmgr: vifmgr.lifmoved.linkdown:notice]: LIF Node-s01_clus1 (on virtual server 4294967293), IP address 169.254.79.159, is being moved to node Node-s01, port e4a.
[Node-01: vifmgr: vifmgr.reach.noreach:notice]: Network port e4a on node Node-s01 cannot reach its expected broadcast domain Cluster:Cluster. No other broadcast domains appear to be reachable from this port.
[Node-01: vifmgr: vifmgr.reach.skipped:notice]: Network port e4a on node Node-s01 was not scanned for reachability because it was administratively or operationally down at the time of the scan.
[Node-01: client_common_RPC: csm.updateBladeIpv4Mapping:debug]: CSM updated blade to IPv4 mapping for cluster 00000000-0000-0000-0000-000000000000, blade baf9c7b3-fc35-11ee-98fe-d039ea589875 to IP address 169.254.231.221, Device ID: e4a, Vserver ID -3.
[Node-01: kernel: netif.linkUp:info]: Ethernet e4a: Link up.
[Node-01: vifmgr: vifmgr.portup:notice]: A link up event was received on node Node-s01, port e4a.
- ONTAPは「Link Resetting」と報告しました
[Node-01: processEntry: netif.linkInfo:info]: Ethernet adapter e4a(pci0:78:0:0) has generated a register dump in /mroot/etc/mlx5log : Link Resetting.
[Node-01: kernel: netif.linkInfo:info]: Ethernet adapter e4a(pci0:78:0:0) failed to generate a register dump with error = 17 : Link Resetting.
[Node-01: kernel: netif.linkInfo:info]: Ethernet adapter e4a(pci0:78:0:0) has generated a register dump in /mroot/etc/mlx5log : Link Resetting.
[Node-01: kernel: netif.linkInfo:info]: Ethernet adapter e4a(pci0:78:0:0) failed to generate a register dump with error = 17 : Link Resetting.
[Node-01: processEntry: netif.linkInfo:info]: Ethernet adapter e4a(pci0:78:0:0) has generated a register dump in /mroot/etc/mlx5log : Link Resetting.
[Node-01: kernel: netif.linkInfo:info]: Ethernet adapter e4a(pci0:78:0:0) failed to generate a register dump with error = 17 : Link Resetting.
- 影響を受けたポートのSFPの交換では問題は解決しなかった