[HDFS-14362] HDFS cluster with EC enabled crashes without any workload

• Type: Bug
• Status: Open
• Priority: Major
• Resolution: Unresolved
• Affects Version/s: 3.1.1
• Component/s: erasure-coding
• Assignee: Unassigned
• Reporter: Aihua Xu
• Created: 12/Mar/19 17:38
• Updated: 12/Mar/19 17:42

Description

We have a small test cluster on 3.1.1 with erasure coding enabled. We loaded some data but doesn't have active data access. Then we are seeing the namenode, datanode crash. Use this parent jira to track the issues. Right now we are not sure if it's related to EC and if it has been fixed in the later version.

Sub-Tasks

• Namenode crashes without any workload: Open
• DataNode crashes without any workload: Open