Mailing-List: contact hdfs-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: hdfs-issues@hadoop.apache.org
Date: Tue, 15 Mar 2016 08:15:37 +0000 (UTC)
From: "dragon (JIRA)" <jira@apache.org>
To: hdfs-issues@hadoop.apache.org
Message-ID: <JIRA.12950302.1458029724000.18452.1458029737226@Atlassian.JIRA>
In-Reply-To: <JIRA.12950302.1458029724000@Atlassian.JIRA>
References: <JIRA.12950302.1458029724000@Atlassian.JIRA>
 <JIRA.12950302.1458029724962@arcas>
Subject: [jira] [Created] (HDFS-10038) CLONE -
 ClientProtocol#createErasureCodingZone API was wrongly annotated as
 Idempotent
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit

dragon created HDFS-10038:
-----------------------------

             Summary: CLONE - ClientProtocol#createErasureCodingZone API was wrongly annotated as Idempotent
                 Key: HDFS-10038
                 URL: https://issues.apache.org/jira/browse/HDFS-10038
             Project: Hadoop HDFS
          Issue Type: Sub-task
          Components: namenode
            Reporter: dragon
            Assignee: Vinayakumar B
             Fix For: HDFS-7285


Currently createErasureCodingZone was annotated as Idempotent
But it should be annotated as @AtMostOnce as we handle retries via retryCache.

{code}
 @Idempotent
  public void createErasureCodingZone(String src, ECSchema schema)
      throws IOException;
{code}

It will fail to create Zone if its already a zone. So, simply we can not retry by ignoring previous call success. So, we were using retryCache already for handling this situation.

{code}
 if (getECSchema(srcIIP) != null) {
      throw new IOException("Directory " + src + " is already in an " +
          "erasure coding zone.");
    }
{code}


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