hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Todd Lipcon (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HDFS-3438) BootstrapStandby should not require a rollEdits on active node
Date Fri, 18 May 2012 22:01:09 GMT

     [ https://issues.apache.org/jira/browse/HDFS-3438?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Todd Lipcon updated HDFS-3438:
------------------------------

    Attachment: hdfs-3438.txt

Attached patch makes the following changes:
- bootstrapStandby no longer rolls the edit log on the other node. So, the other node may
be in STANDBY state and/or in safemode.
- it now uses a new RPC {{getMostRecentCheckpointTxId}} to determine which image to transfer.
I looked at using the {{txid=latest}} feature in TransferFsImage, but unfortunately the APIs
would have had to change around a fair bit in order to use this and also know _which_ txid
was fetched, etc.
- changed {{getTransactionId}} RPC to return either the most recent written ID (if active)
or the most recent applied ID (if in standby). This was necessary so we can make sure that
the new standby has access to the most up-to-date edits.
                
> BootstrapStandby should not require a rollEdits on active node
> --------------------------------------------------------------
>
>                 Key: HDFS-3438
>                 URL: https://issues.apache.org/jira/browse/HDFS-3438
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: ha
>    Affects Versions: 2.0.0
>            Reporter: Todd Lipcon
>            Assignee: Todd Lipcon
>         Attachments: hdfs-3438.txt
>
>
> Currently, BootstrapStandby uses a rollEdits() call on the active NN in order to determine
the most recent checkpoint and transaction ID. However, this means that you cannot bootstrap
a standby when the active NN is in safe mode -- i.e you have to start the whole cluster before
you can bootstrap a new standby. This makes the workflow to convert an existing cluster to
HA more complicated. We should allow BootstrapStandby to work even when the NN is in safe
mode.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message