[jira] [Commented] (FLINK-30461) Some rocksdb sst files will remain forever

["Rui Fan (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/FLINK-30461?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17649696#comment-17649696 ] 

Rui Fan commented on FLINK-30461:
---------------------------------

Hi [~pnowojski] , could you please help take a look in your free time? thanks :)

And I have a POC commit to fix this bug.

[https://github.com/1996fanrui/flink/commit/86a456b2bbdad6c032bf8e0bff71c4824abb3ce1]

 

BTW, FLINK-28984 is also a bug caused some checkpoint files remain forever. I have fixed these 2 bugs in my test job, no files remain with these fixes.

 

 

> Some rocksdb sst files will remain forever
> ------------------------------------------
>
>                 Key: FLINK-30461
>                 URL: https://issues.apache.org/jira/browse/FLINK-30461
>             Project: Flink
>          Issue Type: Bug
>          Components: Runtime / Checkpointing
>    Affects Versions: 1.16.0, 1.17.0, 1.15.3
>            Reporter: Rui Fan
>            Priority: Major
>         Attachments: image-2022-12-20-18-45-32-948.png, image-2022-12-20-18-47-42-385.png
>
>
> In rocksdb incremental checkpoint mode, during file upload, if some files have been uploaded and some files have not been uploaded, the checkpoint timeout is canceled at this time, and the uploaded files will remain.
>  
> h2. Impact: 
> The shared directory of a flink job has more than 1 million files. It exceeded the hdfs upper limit, causing new files not to be written.
> However only 50k files are available, the other 950k files should be cleaned up.
> !https://user-images.githubusercontent.com/38427477/207588272-dda7ba69-c84c-4372-aeb4-c54657b9b956.png|width=1962,height=364!
> h2. Root cause:
> If an exception is thrown during the checkpoint async phase, flink will clean up metaStateHandle, miscFiles and sstFiles.
> However, when all sst files are uploaded, they are added together to sstFiles. If some sst files have been uploaded and some sst files are still being uploaded, and the checkpoint timeout is canceled at this time, all sst files will not be added to sstFiles. The uploaded sst will remain on hdfs.
> [code link|https://github.com/apache/flink/blob/49146cdec41467445de5fc81f100585142728bdf/flink-state-backends/flink-statebackend-rocksdb/src/main/java/org/apache/flink/contrib/streaming/state/snapshot/RocksIncrementalSnapshotStrategy.java#L328]
> h2. Solution:
> Using the CloseableRegistry as the tmpResourcesRegistry. If the async phase is failed, the tmpResourcesRegistry will cleanup these temporary resources.
>  
> POC code:
> https://github.com/1996fanrui/flink/commit/86a456b2bbdad6c032bf8e0bff71c4824abb3ce1
>  
>  
> !image-2022-12-20-18-45-32-948.png|width=1114,height=442!
> !image-2022-12-20-18-47-42-385.png|width=1332,height=552!
>  



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