Mailing-List: contact hdfs-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Date: Wed, 21 Jun 2017 15:42:00 +0000 (UTC)
From: "Sean Mackrory (JIRA)" <jira@apache.org>
To: hdfs-issues@hadoop.apache.org
Message-ID: <JIRA.12993387.1469745277000.71023.1498059720537@Atlassian.JIRA>
In-Reply-To: <JIRA.12993387.1469745277000@Atlassian.JIRA>
References: <JIRA.12993387.1469745277000@Atlassian.JIRA> <JIRA.12993387.1469745277753@jira-lw-us.apache.org>
Subject: [jira] [Updated] (HDFS-10702) Add a Client API and Proxy Provider
 to enable stale read from Standby
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
archived-at: Wed, 21 Jun 2017 15:42:06 -0000


     [ https://issues.apache.org/jira/browse/HDFS-10702?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Sean Mackrory updated HDFS-10702:
---------------------------------
    Attachment: HDFS-10702.008.patch

Rebasing patch on more recent changes. I had also discovered that one of the tests was failing - if you disable stale reads client-side and a fail-over happens before a write-operation, you still might hit a standby namenode and the server might still return a response. I did a quick fix by also setting the minimum txId to Long.MAX_VALUE. Not sure that's how I want to fix it: it means anyone would have to reset the txId when enabling (although in practice I doubt that's a problem, as long as they know they have to, and I doubt that will be a common use case).

I'm going to look at how best to address the potential getBlockLocations issue as well.

> Add a Client API and Proxy Provider to enable stale read from Standby
> ---------------------------------------------------------------------
>
>                 Key: HDFS-10702
>                 URL: https://issues.apache.org/jira/browse/HDFS-10702
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>            Reporter: Jiayi Zhou
>            Assignee: Sean Mackrory
>            Priority: Minor
>         Attachments: HDFS-10702.001.patch, HDFS-10702.002.patch, HDFS-10702.003.patch, HDFS-10702.004.patch, HDFS-10702.005.patch, HDFS-10702.006.patch, HDFS-10702.007.patch, HDFS-10702.008.patch, StaleReadfromStandbyNN.pdf
>
>
> Currently, clients must always talk to the active NameNode when performing any metadata operation, which means active NameNode could be a bottleneck for scalability. One way to solve this problem is to send read-only operations to Standby NameNode. The disadvantage is that it might be a stale read. 
> Here, I'm thinking of adding a Client API to enable/disable stale read from Standby which gives Client the power to set the staleness restriction.


--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-help@hadoop.apache.org