[
https://issues.apache.org/jira/browse/HBASE-15325?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15169454#comment-15169454
]
Phil Yang commented on HBASE-15325:
-----------------------------------
[~cuijianwei] Agree with you, there is another bug.
I am wondering the correct definition of "isPartial".
In comment,
{quote}
Partial results contain a subset of the cells for a row and should be combined
with a result representing the remaining cells in that row to form a complete
(non-partial) result.
{quote}
But in fact if a Result contains the last parts of cells, isPartial will return
false, right?
And if we fix the batching bug as you said, we'll have an interesting situation
that if we set batch to 1, we will get cells one by one and they are all
partial even the last one because we will reach batch limit before the end of
row judgement. Is it acceptable?
> ResultScanner allowing partial result will miss the rest of the row if the
> region is moved between two rpc requests
> -------------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-15325
> URL: https://issues.apache.org/jira/browse/HBASE-15325
> Project: HBase
> Issue Type: Bug
> Affects Versions: 1.2.0, 1.1.3
> Reporter: Phil Yang
> Assignee: Phil Yang
> Priority: Critical
> Attachments: 15325-test.txt, HBASE-15325-v1.txt, HBASE-15325-v2.txt
>
>
> HBASE-11544 allow scan rpc return partial of a row to reduce memory usage for
> one rpc request. And client can setAllowPartial or setBatch to get several
> cells in a row instead of the whole row.
> However, the status of the scanner is saved on server and we need this to get
> the next part if there is a partial result before. If we move the region to
> another RS, client will get a NotServingRegionException and open a new
> scanner to the new RS which will be regarded as a new scan from the end of
> this row. So the rest cells of the row of last result will be missing.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
