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 2023/02/17 16:53:00 UTC

[jira] [Resolved] (HDDS-7178) [Multi-Tenant] Use optimistic read in Ranger background sync

     [ https://issues.apache.org/jira/browse/HDDS-7178?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Attila Doroszlai resolved HDDS-7178.
------------------------------------
    Fix Version/s: 1.4.0
       Resolution: Implemented

> [Multi-Tenant] Use optimistic read in Ranger background sync
> ------------------------------------------------------------
>
>                 Key: HDDS-7178
>                 URL: https://issues.apache.org/jira/browse/HDDS-7178
>             Project: Apache Ozone
>          Issue Type: Sub-task
>          Components: OM
>            Reporter: Ethan Rose
>            Assignee: Ethan Rose
>            Priority: Major
>              Labels: pull-request-available
>             Fix For: 1.4.0
>
>
> Coordination between the background sync and tenant modification requests is currently done through a stamped lock. The background sync currently holds a read lock for a short period every time it runs to fetch a consistent view of OM and Ranger state, however, tenant modifications will be blocked during this time since they need the write lock.
> Since this is already a stamped lock, we can use optimistic read with retries instead of a read lock in the background sync. This way it will not conflict with user tenant modifications as long as there is no divergence between OM and Ranger. If there is divergence, the background sync will still need to push updates under a write lock, blocking tenant modifications while it does so.



--
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