hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jerry He <jerry...@gmail.com>
Subject Re: hbase.server.scanner.max.result.size
Date Wed, 17 Aug 2016 02:59:51 GMT
  >>  - *hbase.client.scanner.max.result.size
   <http://hbase.apache.org/book.html#hbase.client.scanner.max.result.size>*

As documented, the client scanner tries with a max size limit per fetch.
See a related: hbase.client.scanner.caching

 >> - *hbase.server.scanner.max.result.size: **(HBase 13362
   <https://issues.apache.org/jira/browse/HBASE-13362>v1.1)*

As documented, the server tries to enforce a max size limit per fetch on
the server side to protect itself.

>> You can retieve rows > 10 Mb

Yes, if the row boundary is such.

Thanks.

Jerry


On Sun, Aug 14, 2016 at 3:36 AM, Alberto Ramón <a.ramonportoles@gmail.com>
wrote:

> Hi,
>
> We have two properties:
>
>    - *hbase.client.scanner.max.result.size
>    <http://hbase.apache.org/book.html#hbase.client.scanner.max.result.size
> >*
>
>
>    - *hbase.server.scanner.max.result.size: **(HBase 13362
>    <https://issues.apache.org/jira/browse/HBASE-13362>v1.1)*
>
> Cloudera blog:
> <http://blog.cloudera.com/blog/2015/05/scan-improvements-in-apache-hbase-
> 1-1-0/>
> "The max result size limit is only applied at row boundaries. "
>
> I don't understand because, if you put  client max.result.size = 10 MB
>
> You can retieve rows > 10 Mb
>
>
> BR, Alberto
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message