解决 FabricPool 对象存储网络延迟过大的问题

最后更新
另存为PDF

Views:: 80

Visibility:: Public

Votes:: 0

Category:: ontap-9

Specialty:: perf

Last Updated:

适用于

ONTAP 9
StorageGRID FabricPool

问题

已触发 EMS 消息：ARS is triggered when latency is between 5 and 10 seconds detected when using FabricPool..

Fri Sep 06 09:47:54 MDT [node1: OscHighPriThreadPool: wafl.ca.latency.threshold:notice]: Measured latency from cloud (5915 msec) more than threshold (1000 msec) (aggregate: "node1_ssd", aggregate uuid: "563a8dfc-97d1-45c9-921a-5ced93b34f5c", object store: "node_grid")

如果延迟严重，也可能会看到超时错误，导致用户看到 NFS/CIFS 延迟增加，并可能阻止访问对象存储上的数据

[Cluster01-01: OscLowPriThreadPool_4: object.store.unavailable:EMERGENCY]: Unable to connect to the object store "S3_objectstore" from node <uuid>. Reason: Operation Timedout.