集群端口关闭导致 CLAM 崩溃

最后更新
另存为PDF

Knowledge Base guidance

This article may include metadata, visibility, validation state, and related support guidance.

适用于

ONTAP 9
连接性、活跃性和可用性监视器 (CLAM)

问题描述

由于节点的两个集群端口在很短的时间内关闭而导致 CLAM 死机

Sat Aug 10 22:02:37 -0400 [node2: kernel: netif.linkDown:info]: Ethernet e3a: Link down, check cable.

Sat Aug 10 22:02:55 -0400 [node2: kernel: netif.linkDown:info]: Ethernet e3b: Link down, check cable.

Sat Aug 10 22:03:54 -0400 [node1: kltp: clam.received.quorum:debug]: Local node received a quorum update from Cluster node (name=node3, ID=1002). Sat Aug 10 22:03:54 -0400 [node1: kltp: callhome.clam.node.ooq:EMERGENCY]: Call home for NODE(S) OUT OF CLUSTER QUORUM. Sat Aug 10 22:03:54 -0400 [node1: kltp: clam.node.ooq:EMERGENCY]: Node (name=node2, ID=1005) is out of "CLAM quorum" (reason=quorum update). Sat Aug 10 22:03:54 -0400 [node1: kltp: clam.node.avail.change:debug]: The availability status of node (name=node2, ID=1005) changed from Available to Unavailable. Sat Aug 10 22:03:54 -0400 [node1: ThreadHandlerun: clam.update.partner.state:info]: CLAM on node (ID=1004) updated failover state of partner (ID=1005) to nohb. Sat Aug 10 22:03:54 -0400 [node1: clam_bg: clam.quorum.epoch:debug]: CLAM set the quorum epoch on local node to 223. Sat Aug 10 22:03:54 -0400 [node1: ThreadHandlerun: clam.takeover:info]: Local node (name=node1, ID=1004) initiated takeover of partner node (name=node2, ID=1005, state=193) result=successful, type=panic.