CONETP-126100：由于Snapshot自动删除使网络线程饱和、FlexGroup和间接I/O速度变慢
Knowledge Base guidance
This article may include metadata, visibility, validation state, and related support guidance.
问题描述
- Slowness seen with indirect i/o or Flexgroups.
- Snapshots are not being auto deleted.
- Once the snapcoord counters for snapserver_create minus snapserver_delete reaches a difference of around 65000, the node will make all flexgroups unresponsive.
Cluster-node01::*> statistics show -raw -object snapcoord -counter snapserver_create|snapserver_delete
Object: snapcoord Instance: snapcoord Start-time: 2/13/2024 15:14:02 End-time: 2/13/2024 15:14:02 Scope: Cluster-node01Counter Value -------------------------------- -------------------------------- snapserver_create 153332 snapserver_delete 88332