Mailing-List: contact common-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Date: Tue, 1 Aug 2017 18:19:00 +0000 (UTC)
From: "Jason Lowe (JIRA)" <jira@apache.org>
To: common-issues@hadoop.apache.org
Message-ID: <JIRA.13091638.1501611485000.64229.1501611540147@Atlassian.JIRA>
In-Reply-To: <JIRA.13091638.1501611485000@Atlassian.JIRA>
References: <JIRA.13091638.1501611485000@Atlassian.JIRA> <JIRA.13091638.1501611485505@jira-lw-us.apache.org>
Subject: [jira] [Created] (HADOOP-14713) Audit for durations that should be
 measured via Time.monotonicNow
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
archived-at: Tue, 01 Aug 2017 18:19:05 -0000

Jason Lowe created HADOOP-14713:
-----------------------------------

             Summary: Audit for durations that should be measured via Time.monotonicNow
                 Key: HADOOP-14713
                 URL: https://issues.apache.org/jira/browse/HADOOP-14713
             Project: Hadoop Common
          Issue Type: Improvement
            Reporter: Jason Lowe


Whenever we are measuring a time delta or duration in the same process, the timestamps probably should be using Time.monotonicNow rather than Time.now or System.currentTimeMillis.  The latter two are directly reading the system clock which can move faster or slower than actual time if the system is undergoing a time adjustment (e.g.: adjtime or admin sets a new system time).

We should go through the code base and identify places where the code is using the system clock but really should be using monotonic time.


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

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