[HDFS-15391] Standby NameNode due loads the corruption edit log, the service exits and cannot be restarted

• Type: Bug
• Status: Resolved
• Priority: Critical
• Resolution: Duplicate
• Affects Version/s: 3.2.0
• Component/s: namenode
• Assignee: Unassigned
• Reporter: Haiyang Hu
• Due: 05/Jun/20
• Created: 05/Jun/20 12:10
• Updated: 15/Jun/20 10:08
• Resolved: 15/Jun/20 09:40

Description

In the cluster version 3.2.0 production environment，
We found that due to edit log corruption, Standby NameNode could not properly load the Ediltog log, result in abnormal exit of the service and failure to restart

The specific scenario is that Flink writes to HDFS(replication file), and in the case of an exception to the write file, the following operations are performed :
1.close file
2.open file
3.truncate file
4.append file

Issue Links

• is duplicated by: HDFS-15175 Multiple CloseOp shared block instance causes the standby namenode to crash when rolling editlog