Cisco Cluster Network Switch が再起動ループに陥っています

最後の更新
PDFとして保存

Views:

Visibility:: Public

Votes:: 0

Category:: ontap-9

Specialty:: hw

Last Updated:

環境

FAS/AFF
ONTAP 9

問題

両ノードからスイッチへ接続しているクラスタポートが同時にダウンします。

EMS

[node1: kernel: netif.linkDown:info]: Ethernet e0a: Link down, check cable.
[node2: kernel: netif.linkDown:info]: Ethernet e0a: Link down, check cable.

Sysconfig -a

   NetApp Release 9.15.1P14: Tue Aug 5 11:53:54 EDT 2025
   System ID: 05XXXXX994 (node1); partner ID: 05XXXXX944 (node2)
   System Serial Number: 95XXXXXX2341 (node1)

   slot 0: Dual 10/25G Ethernet Controller E823L-SFP+
    e0b MAC Address:   d0:39:ea:c8:2a:36 (auto-25g_cr-fd-up)
      SFP Vendor:     CISCO-BIZLINK
      SFP Part Number:   L45593-D278-D20
      SFP Serial Number: LCC2813GWAA-CH1
   e0a MAC Address:   d0:39:ea:c8:2a:37 (auto-unknown-fd-down)
      SFP Vendor:     CISCO-BIZLINK
      SFP Part Number:   L45593-D278-D20
      SFP Serial Number: LCC2813GWAP-CH1

   NetApp Release 9.15.1P14: Tue Aug 5 11:53:54 EDT 2025
   System ID: 05XXXXX944 (node2); partner ID: 05XXXXX994 (node1)
   System Serial Number: 95XXXXXX2374 (node2)
   System Rev: A9

   slot 0: Dual 10/25G Ethernet Controller E823L-SFP+
    e0b MAC Address:   d0:39:ea:c8:26:bd (auto-25g_cr-fd-up)
      SFP Vendor:     CISCO-BIZLINK
      SFP Part Number:   L45593-D278-D20
      SFP Serial Number: LCC2813GWAA-CH2
   e0a MAC Address:   d0:39:ea:c8:26:be (auto-unknown-fd-down)
      SFP Vendor:     CISCO-BIZLINK
      SFP Part Number:   L45593-D278-D20
      SFP Serial Number: LCC2813GWAP-CH2

クラスターは次のヘルスアラートを報告します：
Cluster1::> system health alert show
        Node: node2
      Alert ID: ClusterSwitchConnectionDegraded_Alert
      Resource: node1
      Severity: Major
   Indication Time: Sat Jan 03 14:39:23 2026
      Suppress: false
     Acknowledge: false
   Probable Cause: Only one cluster switch, "switch2(FDXXXXXX3H)",
           has been discovered to be connected to the cluster
           ports of node "node1". If a cluster
           connection was not discovered, but the expected
           cluster switch is listed in the output of "system
           switch ethernet show", it indicates that the node is
           not currently receiving discovery information about
           the missing switches.
   Possible Effect: If the remaining cluster switch,
           "switch2(FDXXXXXX3H)", that is connected to
           node "node1" fails, the node may lose access
           to the cluster.
Corrective Actions: 1) Ensure node "node1" is connected to multiple cluster switches. Inspect cables for damage or loose connections, check port status and health, verify VLAN configurations, ensure matching port speed and duplex settings, and review logs for errors.
           2) Enable the corresponding link layer discovery protocol on the affected switch(es) if previously disabled. As a reminder for cluster switches, Broadcom switches use ISDP, Cisco switches use CDP, and Nvidia switches use LLDP. The discovery protocol should be enabled once configured with the correct Reference Configuration File (RCF).
           Refer to your switch documentation for specific
           instructions on enabling the protocol manually.

        Node: node2
      Alert ID: SwitchIslIfDownWarn_Alert
      Resource: Ethernet1/35
      Severity: Major
   Indication Time: Sat Jan 03 13:42:39 2026
      Suppress: false
     Acknowledge: false
   Probable Cause: The cable attached to the ISL port
           "switch2(FDXXXXXX3H)/Ethernet1/35" with the
           description of "Intra-Cluster Switch ISL Port 1/35
           (port channel)" might be faulty. For the port-channel,
           please check each individual link.
   Possible Effect: Inter- or Intra- cluster redundancy might be lost.
Corrective Actions: 1) Check whether the cable is fully inserted into the interface port on both ends.
           2) Reconnect the ISL port using another cable. After
           reconnecting it, verify whether the link of the ISL
           port is up by executing the commands per the switch
           configuration guide or by checking whether the Link
           status LED is on.

        Node: node2
      Alert ID: SwitchIslIfDownWarn_Alert
      Resource: Ethernet1/36
      Severity: Major
   Indication Time: Sat Jan 03 13:42:39 2026
      Suppress: false
     Acknowledge: false
   Probable Cause: The cable attached to the ISL port
           "switch2(FDXXXXXX3H)/Ethernet1/36" with the
           description of "Intra-Cluster Switch ISL Port 1/36
           (port channel)" might be faulty. For the port-channel,
           please check each individual link.
   Possible Effect: Inter- or Intra- cluster redundancy might be lost.
Corrective Actions: 1) Check whether the cable is fully inserted into the interface port on both ends.
           2) Reconnect the ISL port using another cable. After
           reconnecting it, verify whether the link of the ISL
           port is up by executing the commands per the switch
           configuration guide or by checking whether the Link
           status LED is on.

スイッチが3～4分ごとに再起動するため、スイッチログを取得できません。