[
https://issues.apache.org/jira/browse/HDDS-12323?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17928707#comment-17928707
]
Sammi Chen commented on HDDS-12323:
-----------------------------------
The function will be covered by HDDS-12233.
> Improve container import failure cleanup flow
> ---------------------------------------------
>
> Key: HDDS-12323
> URL: https://issues.apache.org/jira/browse/HDDS-12323
> Project: Apache Ozone
> Issue Type: Sub-task
> Reporter: Sammi Chen
> Assignee: Sammi Chen
> Priority: Major
> Labels: pull-request-available
>
> When a schema V3 container failed to be imported, three resource will be
> cleaned up,
> a. the potential still existed downloaded *.tar file on volume's
> tmp/container-copy/ directory
> b. the imported container metadata in RocksDB
> c. the potential existed $containerID directory on data volume.
> Currently, the clean up order is a -> b -> c.
> Given that during Datanode startup, only a missing or incorrect
> container.yaml file will get the container be skip to be loaded by datanode,
> we should first make sure container.yaml file is inaccessible to datanode
> after the container import is failed.
> This task aims to change the clean up order to a -> c -> b.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
