[
https://issues.apache.org/jira/browse/HBASE-15378?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15186948#comment-15186948
]
Phil Yang commented on HBASE-15378:
-----------------------------------
Seems each job in https://builds.apache.org/job/PreCommit-HBASE-Build run each
test twice... Is it expected?
> Scanner can not handle a heartbeat message with no results
> ----------------------------------------------------------
>
> Key: HBASE-15378
> URL: https://issues.apache.org/jira/browse/HBASE-15378
> Project: HBase
> Issue Type: Bug
> Components: dataloss, Scanners
> Affects Versions: 1.2.0, 1.1.3
> Reporter: Phil Yang
> Assignee: Phil Yang
> Priority: Critical
> Fix For: 2.0.0, 1.3.0, 1.2.1, 1.1.4, 1.4.0
>
> Attachments: HBASE-15378-v1.txt, HBASE-15378-v2.txt,
> HBASE-15378-v3.txt, HBASE-15378-v4.patch, HBASE-15378-v5.patch
>
>
> When a RS scanner get a TIME_LIMIT_REACHED_MID_ROW state, they will stop
> scanning and send back what it has read to client and mark the message as a
> heartbeat message. If there is no cell has been read, it will be an empty
> response.
> However, ClientScanner only handles the situation that the client gets an
> empty heartbeat and its cache is not empty. If the cache is empty too, it
> will be regarded as end-of-region and open a new scanner for next region.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
