farmmamba created HDFS-17054:
--------------------------------
Summary: Erasure coding: optimize checkReplicaOnStorage method to
avoid regarding all replicas on one datanode as corrupt repeatly.
Key: HDFS-17054
URL: https://issues.apache.org/jira/browse/HDFS-17054
Project: Hadoop HDFS
Issue Type: Improvement
Reporter: farmmamba
Currently, when we execute countNodes method for a striped block, if a datanode
has two ec blocks with different block id (because of some special cases and I
will find and fix it laterly.). One block is LIVE, the other is CORRUPT.
Current logic will compute two CORRUPT replicas because corruptReplicas
contains the datanode. We should fix it.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
