Mailing-List: contact issues-help@geode.incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@geode.incubator.apache.org
Date: Mon, 1 Feb 2016 18:08:40 +0000 (UTC)
From: "ASF subversion and git services (JIRA)" <jira@apache.org>
To: issues@geode.incubator.apache.org
Message-ID: <JIRA.12934637.1453940300000.266394.1454350120156@Atlassian.JIRA>
In-Reply-To: <JIRA.12934637.1453940300000@Atlassian.JIRA>
References: <JIRA.12934637.1453940300000@Atlassian.JIRA>
 <JIRA.12934637.1453940300020@arcas>
Subject: [jira] [Commented] (GEODE-871) Frequent use of executeOnAllServers
 causes build-up of sockets in TIME_WAIT on locator machine
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/GEODE-871?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15126687#comment-15126687 ] 

ASF subversion and git services commented on GEODE-871:
-------------------------------------------------------

Commit d22176bb1632dc5593a1755a1355aacd0360c5e8 in incubator-geode's branch refs/heads/feature/GEODE-805 from [~bschuchardt]
[ https://git-wip-us.apache.org/repos/asf?p=incubator-geode.git;h=d22176b ]

GEODE-871: Frequient use of executeOnAllServers causes build-up of TIME_WAIT sockets

Modified the fix to not abort the connection if a response wasn't read.
I also propagated this pattern to GMSHealthMonitor so that the "final
check" sockets are quickly released.


> Frequent use of executeOnAllServers causes build-up of sockets in TIME_WAIT on locator machine
> ----------------------------------------------------------------------------------------------
>
>                 Key: GEODE-871
>                 URL: https://issues.apache.org/jira/browse/GEODE-871
>             Project: Geode
>          Issue Type: Bug
>          Components: client/server
>            Reporter: Bruce Schuchardt
>            Assignee: Bruce Schuchardt
>             Fix For: 1.0.0-incubating.M2
>
>
> The executeOnAllServers implementation contacts a locator in order to get the current set of servers.  This opens a TCP/IP connection to the locator, interacts with the locator and then closes the TCP/IP connection.  The connection is ending up in TIME_WAIT state though, indicating that the locator is usually initiating the close of the connection instead of the client.
> These TIME_WAIT TCP/IP connections can build up rapidly and cause the locator to run out of ephemeral ports.  That puts the locator out of commission until the connections are timed out by the OS.


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