hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Colin Patrick McCabe (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HDFS-5394) fix race conditions in DN caching and uncaching
Date Tue, 29 Oct 2013 21:49:27 GMT

     [ https://issues.apache.org/jira/browse/HDFS-5394?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Colin Patrick McCabe updated HDFS-5394:
---------------------------------------

    Attachment: HDFS-5394.005.patch

* renamed visibleLength to length

* removed {{ConcurrentHashMap}} in favor of {{HashMap}} plus locking.

* do all I/O in caching task, not in caller.

* print the name of the block file, not just a static string

* unit test fixup: use a delay rather than countdown latch.

* some more debug logs.

> fix race conditions in DN caching and uncaching
> -----------------------------------------------
>
>                 Key: HDFS-5394
>                 URL: https://issues.apache.org/jira/browse/HDFS-5394
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>          Components: datanode, namenode
>    Affects Versions: 3.0.0
>            Reporter: Colin Patrick McCabe
>            Assignee: Colin Patrick McCabe
>         Attachments: HDFS-5394.005.patch, HDFS-5394-caching.001.patch, HDFS-5394-caching.002.patch,
HDFS-5394-caching.003.patch, HDFS-5394-caching.004.patch
>
>
> The DN needs to handle situations where it is asked to cache the same replica more than
once.  (Currently, it can actually do two mmaps and mlocks.)  It also needs to handle the
situation where caching a replica is cancelled before said caching completes.



--
This message was sent by Atlassian JIRA
(v6.1#6144)

Mime
View raw message