You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@ozone.apache.org by "Attila Doroszlai (Jira)" <ji...@apache.org> on 2022/10/14 12:21:00 UTC

[jira] [Reopened] (HDDS-7123) container scan should hold the container read lock

     [ https://issues.apache.org/jira/browse/HDDS-7123?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Attila Doroszlai reopened HDDS-7123:
------------------------------------

> container scan should hold the container read lock
> --------------------------------------------------
>
>                 Key: HDDS-7123
>                 URL: https://issues.apache.org/jira/browse/HDDS-7123
>             Project: Apache Ozone
>          Issue Type: Bug
>            Reporter: Jie Yao
>            Assignee: Jie Yao
>            Priority: Major
>              Labels: pull-request-available
>
> now, when scanning container, datanode does not hold a lock. when the blockDeletingService works, it will first deletes the block file in container directory and then delete the metadata in rocksdb. when container scanner works , it will first read the metadata from DB, and then check the block file in container directory.  the two background service runs concurrently for now. there may be a case that when blockDeletingService deletes the block file from container directory but not delete the metadata from rocksdb, container scanner read the metadata from rocksdb and then fail to find the block file in container directory, and then mark the container replica unhealthy



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@ozone.apache.org
For additional commands, e-mail: issues-help@ozone.apache.org