hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Himanshu Vashishtha (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-9539) Handle post namespace snapshot files when checking for HFile V1
Date Mon, 16 Sep 2013 18:38:53 GMT

    [ https://issues.apache.org/jira/browse/HBASE-9539?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13768594#comment-13768594
] 

Himanshu Vashishtha commented on HBASE-9539:
--------------------------------------------

Thanks for the review Stack.

Re: naming: These DOT_XXX naming is borrowed from existing code, NameSpaceUpgrade class. But
no problem changing them.

Re: +if (upgrade) {
Yes, that is to add mutual exclusive behavior in the upgrade script in the two modes (check
and upgrade). I think it is good to have.

Re: more jave doc
Sure, will add more of it.

bq. Does this make it so this tool only works against pre-0.96 snapshot layout? 
Not really. This patch adds more paths to look for when we want to open the FileLink (the
FileLink takes an array of path to look when opening the link). So, we just add pre-96 snapshot
layout in those options, and make it work on a pre-96 layout (we not removing post-96 related
paths).


                
> Handle post namespace snapshot files  when checking for HFile V1 
> -----------------------------------------------------------------
>
>                 Key: HBASE-9539
>                 URL: https://issues.apache.org/jira/browse/HBASE-9539
>             Project: HBase
>          Issue Type: Bug
>          Components: migration
>    Affects Versions: 0.95.2
>            Reporter: Himanshu Vashishtha
>             Fix For: 0.98.0, 0.96.0
>
>         Attachments: HBase-9539.patch
>
>
> When checking for HFileV1 before upgrading to 96, the snapshot file links tries to read
from post-namespace locations. The migration script needs to be run on 94 cluster, and it
requires reading the old (94) layout to check for HFileV1.
> {code}
> Got exception while reading trailer for file: hdfs://xxx:41020/cops/cluster_collection_events_snapshot/2086db948c484be62dcd76c170fe0b17/meta/cluster_collection_event=42037b88dbc34abff6cbfbb1fde2c900-c24b358ddd2f4429a7287258142841a2
> java.io.FileNotFoundException: Unable to open link: org.apache.hadoop.hbase.io.HFileLink
locations=[hdfs://xxx:41020/hbase-96/data/default/cluster_collection_event/42037b88dbc34abff6cbfbb1fde2c900/meta/c24b358ddd2f4429a7287258142841a2,
hdfs://xxx:41020/hbase-96/.tmp/data/default/cluster_collection_event/42037b88dbc34abff6cbfbb1fde2c900/meta/c24b358ddd2f4429a7287258142841a2,
hdfs://xxx:41020/hbase-96/archive/data/default/cluster_collection_event/42037b88dbc34abff6cbfbb1fde2c900/meta/c24b358ddd2f4429a7287258142841a2]
> {code}

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message