[ 
https://issues.apache.org/jira/browse/HDDS-7715?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

ASF GitHub Bot updated HDDS-7715:
---------------------------------
    Labels: pull-request-available  (was: )

> keyManagerImpl#listStatus exceeds the maximum RPC length
> --------------------------------------------------------
>
>                 Key: HDDS-7715
>                 URL: https://issues.apache.org/jira/browse/HDDS-7715
>             Project: Apache Ozone
>          Issue Type: Bug
>          Components: OM, Ozone Manager
>    Affects Versions: 1.3.0
>         Environment: Version: Our forked version based ozone-1.3.0 branch. 
> commit hash is 
> [https://github.com/apache/ozone/commit/9c61a8aa497ab96c014ad3bb7b1ee4f731ebfaf8]
> FSO-enabled bucket
>            Reporter: Kohei Sugihara
>            Priority: Major
>              Labels: pull-request-available
>         Attachments: large-protos.png
>
>
> keyManagerImpl#listStatus for FSO buckets serves a KeyLocationList but might 
> be large size of response (> 100 MB) and exceed the limit. The problem is 
> KeyLocation field is large enough (> 1MB) so just listing some multi-part 
> uploaded keys exceeds the limit. A possible solution is to reduce KeyLocation 
> size by omitting pipeline fields.
> {quote}% ozone sh key ls /s3v/bucket --prefix=path/to/prefix
> RPC response exceeds maximum data length
> {quote}
>  
> We checked a List<OzoneFileStatus> in our environment and it turned out that 
> protobufs corresponding to some multi-part keys are larger than 1MB. AFAIK, 
> the maximum IPC length is 100MB by default so it easily exceeds the limit.
> !large-protos.png!



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

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to