[
https://issues.apache.org/jira/browse/HBASE-19977?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16362606#comment-16362606
]
Hudson commented on HBASE-19977:
--------------------------------
FAILURE: Integrated in Jenkins build HBase-Trunk_matrix #4578 (See
[https://builds.apache.org/job/HBase-Trunk_matrix/4578/])
HBASE-19977 FileMmapEngine allocation of byte buffers should be
(ramkrishna.s.vasudevan: rev 16f1f5b49424fcabc9b5c10882dab4f5bf7fa84b)
* (edit)
hbase-server/src/main/java/org/apache/hadoop/hbase/io/hfile/bucket/FileMmapEngine.java
> FileMmapEngine allocation of byte buffers should be synchronized
> ----------------------------------------------------------------
>
> Key: HBASE-19977
> URL: https://issues.apache.org/jira/browse/HBASE-19977
> Project: HBase
> Issue Type: Bug
> Affects Versions: 2.0.0-alpha-4
> Reporter: ramkrishna.s.vasudevan
> Assignee: ramkrishna.s.vasudevan
> Priority: Major
> Fix For: 2.0.0-beta-2
>
> Attachments: HBASE-19977.patch
>
>
> Recently we have been testing bucket cache with mmap mode. We found that
> after the multi threading way of allocating the byte buffers for offheap
> bucket cache, the creation of the mmapped byte buffers needs to be
> synchronized for allocating the right sized mmapped buffers.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
