[ 
https://issues.apache.org/jira/browse/HBASE-25431?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17257906#comment-17257906
 ] 

Baiqiang Zhao commented on HBASE-25431:
---------------------------------------

Yes, the original logic is use maxFileSizeLowerLimit (2MB) instead if < 0. But 
the real value in TableDescriptor is still < 0, and we can see negative values 
​​on the UI.

> MAX_FILESIZE and MEMSTORE_FLUSHSIZE should not be set negative number
> ---------------------------------------------------------------------
>
>                 Key: HBASE-25431
>                 URL: https://issues.apache.org/jira/browse/HBASE-25431
>             Project: HBase
>          Issue Type: Improvement
>    Affects Versions: 3.0.0-alpha-1, 2.4.0
>            Reporter: Baiqiang Zhao
>            Assignee: Baiqiang Zhao
>            Priority: Major
>         Attachments: image-2020-12-22-11-46-38-967.png
>
>
> Before create or alter table, we will do sanityCheck in 
> TableDescriptorChecker. If MAX_FILESIZE or MEMSTORE_FLUSHSIZE < 0, use 
> maxFileSizeLowerLimit or flushSizeLowerLimit instead to pass check. But the 
> real value in TableDescriptor is still < 0, and we can see negative values 
> ​​on the UI.
> However in flush and split logic, MAX_FILESIZE and MEMSTORE_FLUSHSIZE will 
> judge whether it's value <= 0 , if true, change to default value. This does 
> not affect flush and split.
> !image-2020-12-22-11-46-38-967.png!



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

Reply via email to