hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Suresh Srinivas (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-5709) Improve upgrade with existing files and directories named ".snapshot"
Date Tue, 21 Jan 2014 19:45:20 GMT

    [ https://issues.apache.org/jira/browse/HDFS-5709?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13877783#comment-13877783

Suresh Srinivas commented on HDFS-5709:

bq. By "silently" I meant without logging anything, not that we shouldn't necessarily do the
rename by default. 
By silently I mean without the active participation of the user. Logging into namenode logs,
I can see many people not realizing the rename has occurred until some applications that depend
on the path fail.

bq. Why would they prefer to have to manually downgrade the software, start up the cluster,
manually find all reserved names, rename them, then re-attempt the upgrade? 
Where is this being proposed? Why is downgrade required in any of the proposed solutions?

bq. Suresh - what about the other part of my proposal, i.e. by default perform the rename
to the name you suggested ("<original file name><layout_version>.reserved_renamed_after_ugprade"),
but optionally rename to something else?
Doing the same thing in multiple ways seems unnecessary and confusing. I do not see any significant
advantage to it. Perhaps I do not understand this solution.

> Improve upgrade with existing files and directories named ".snapshot"
> ---------------------------------------------------------------------
>                 Key: HDFS-5709
>                 URL: https://issues.apache.org/jira/browse/HDFS-5709
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: namenode
>    Affects Versions: 3.0.0, 2.2.0
>            Reporter: Andrew Wang
>            Assignee: Andrew Wang
>              Labels: snapshots, upgrade
>         Attachments: hdfs-5709-1.patch, hdfs-5709-2.patch, hdfs-5709-3.patch, hdfs-5709-4.patch,
> Right now in trunk, upgrade fails messily if the old fsimage or edits refer to a directory
named ".snapshot". We should at least print a better error message (which I believe was the
original intention in HDFS-4666), and [~atm] proposed automatically renaming these files and

This message was sent by Atlassian JIRA

View raw message