[jira] [Resolved] (HDDS-3926) Ozone Manager Token Identifier table should use in-house serialization rather than rely on proto serialization for key.

["Xiaoyu Yao (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HDDS-3926?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Xiaoyu Yao resolved HDDS-3926.
------------------------------
    Fix Version/s: 0.6.0
       Resolution: Fixed

Thanks [~ppogde] for the contribution and all for the reviews. Change has been merged and cherry-picked to ozone-0.6.0

> Ozone Manager Token Identifier table should use in-house serialization rather than rely on proto serialization for key.
> -----------------------------------------------------------------------------------------------------------------------
>
>                 Key: HDDS-3926
>                 URL: https://issues.apache.org/jira/browse/HDDS-3926
>             Project: Hadoop Distributed Data Store
>          Issue Type: Sub-task
>          Components: SCM
>            Reporter: Aravindan Vijayan
>            Assignee: Prashant Pogde
>            Priority: Major
>              Labels: pull-request-available, upgrade-p0
>             Fix For: 0.6.0
>
>
> Relying on Protobuf serialization for exact match is unreliable according to the docs. Hence, we have to move away from using proto.toByteArray() for on disk RocksDB keys. For more details, check parent JIRA.
> In the case of the Ozone token identifier, it can be uniquely identified using the following fields.
> * Issue Date
> * Master Key ID
> * Sequence Number
> We can provide a simplified serialization method using the above 3 fields (in that order) to be used in the Token Identifier codec.
> cc [~xyao]



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

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