Mailing-List: contact hdfs-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Date: Thu, 1 Sep 2016 06:43:22 +0000 (UTC)
From: "Zhe Zhang (JIRA)" <jira@apache.org>
To: hdfs-issues@hadoop.apache.org
Message-ID: <JIRA.13001378.1472574356000.466433.1472712202322@Atlassian.JIRA>
In-Reply-To: <JIRA.13001378.1472574356000@Atlassian.JIRA>
References: <JIRA.13001378.1472574356000@Atlassian.JIRA> <JIRA.13001378.1472574356498@arcas>
Subject: [jira] [Updated] (HDFS-10817) Add Logging for Long-held NN Read
 Locks
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
archived-at: Thu, 01 Sep 2016 06:43:25 -0000


     [ https://issues.apache.org/jira/browse/HDFS-10817?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Zhe Zhang updated HDFS-10817:
-----------------------------
    Attachment: HDFS-10817.addendum.001.patch

Thanks Erik. The addendum fix does solve the problem. Somehow the {{LogCapturer}} prints thread IDs in trunk but not in branch-2. But after the fix we are no longer relying on the behavior of {{LogCapturer}}.

But the "Spin up a bunch of threads" part no longer makes much sense now. Because we are checking by method name, and 25 threads share the same method name. I removed that part in the test (based on the original trunk test we have already verified the behavior). We should think of a proper way to test with multiple parallel threads.

> Add Logging for Long-held NN Read Locks
> ---------------------------------------
>
>                 Key: HDFS-10817
>                 URL: https://issues.apache.org/jira/browse/HDFS-10817
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: logging, namenode
>            Reporter: Erik Krogen
>            Assignee: Erik Krogen
>         Attachments: HDFS-10817.000.patch, HDFS-10817.001.patch, HDFS-10817.002.patch, HDFS-10817.003.patch, HDFS-10817.004.patch, HDFS-10817.addendum.000.patch, HDFS-10817.addendum.001.patch
>
>
> Right now the Namenode will log when a write lock is held for a long time to help tracks methods which are causing expensive delays. Let's do the same for read locks since these operations may also be expensive/long and cause delays. 


--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

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