[
https://issues.apache.org/jira/browse/HDFS-1826?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Matt Foley updated HDFS-1826:
-----------------------------
Attachment: 1826_FSImageWrite_concur_upgrade_v3.patch
Agreed. I originally used multiple try/catch blocks to get a different log
message for each possible failure, but just including the Exception object in
the log is sufficient.
Here's an update, with unit test.
> NameNode should save image to name directories in parallel during upgrade
> -------------------------------------------------------------------------
>
> Key: HDFS-1826
> URL: https://issues.apache.org/jira/browse/HDFS-1826
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: name-node
> Affects Versions: 0.23.0
> Reporter: Hairong Kuang
> Assignee: Matt Foley
> Fix For: 0.23.0
>
> Attachments: 1826_FSImageWrite_concur_upgrade_v1.patch,
> 1826_FSImageWrite_concur_upgrade_v3.patch
>
>
> Currently namenode saves a new image to all its name directories in sequence
> during upgrade. We should make it to do in parallel to speedup namenode
> upgrade. Also it should not save its image to edits directories.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
