[
https://issues.apache.org/jira/browse/HDDS-890?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Shashikant Banerjee updated HDDS-890:
-------------------------------------
Description: With raft log entries getting truncated, the corresponding
writeStateMachineData futures doing actual chunk writes can not aborted. The
next subsequent log entry corresponding to the same client request may end up
writing the same chunk file concurrently hitting OverlappingFileLockException.
This Jira aims to address this issue by appending logIndex and term to every
tmp chunkFile name created during writeStateMachineData call, (was: With raft
log entries getting truncated, the corresponding writeStateMachineData futures
doing actual chunk writes can not aborted. The next subsequent log entry
coresponding to the same client request may end up writing the same chunk file
concurrently hitting )
> Handle OverlappingFileLockException during writeStateMachineData in
> ContainerStateMachine
> -------------------------------------------------------------------------------------------
>
> Key: HDDS-890
> URL: https://issues.apache.org/jira/browse/HDDS-890
> Project: Hadoop Distributed Data Store
> Issue Type: Improvement
> Components: Ozone Datanode
> Affects Versions: 0.4.0
> Reporter: Shashikant Banerjee
> Assignee: Shashikant Banerjee
> Priority: Major
>
> With raft log entries getting truncated, the corresponding
> writeStateMachineData futures doing actual chunk writes can not aborted. The
> next subsequent log entry corresponding to the same client request may end up
> writing the same chunk file concurrently hittingÂ
> OverlappingFileLockException. This Jira aims to address this issue by
> appending logIndex and term to every tmp chunkFile name created during
> writeStateMachineData call,
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
