Mailing-List: contact common-dev-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: common-dev@hadoop.apache.org
Received-SPF: pass (nike.apache.org: domain of mohitsikri@huawei.com
 designates 119.145.14.64 as permitted sender)
Date: Fri, 04 Feb 2011 10:23:19 +0530
From: Mohit <mohitsikri@huawei.com>
Subject: RE: Query on Hadoop-1849 with respect to latest working trunk.
In-reply-to: <5BB95907E2BA499EB9B30F14A5D23420@china.huawei.com>
To: common-dev@hadoop.apache.org
Reply-to: mohitsikri@huawei.com
Message-id: <BC112BFC1E084BDC92571DC58C2A9B56@china.huawei.com>
Organization: Htipl
MIME-version: 1.0
Content-type: multipart/alternative;
 boundary="Boundary_(ID_QyDXfpfVqmi453XaDb4W6w)"
Thread-index: AcvCqLZdXvPfxPkPQkOyTuyD2m5gNwBfc7cg

--Boundary_(ID_QyDXfpfVqmi453XaDb4W6w)
Content-type: text/plain; charset=us-ascii
Content-transfer-encoding: 7BIT

Hello Authors,

 
I suspect there is a problem in there,

 
I configured a property ipc.server.listen.queue.size in core-site.xml to a
negative value say -6.

If this property has actually being considered while calculating queue size,
the result would be a negative value, and hence the startup of NAMENODE
should fail because Server.java <init> tries to instantiate linked blocking
queue with negative fixed capacity; but it actually passed.

 
-Mohit

 
****************************************************************************
***********

This e-mail and attachments contain confidential information from HUAWEI,
which is intended only for the person or entity whose address is listed
above. Any use of the information contained herein in any way (including,
but not limited to, total or partial disclosure, reproduction, or
dissemination) by persons other than the intended recipient's) is
prohibited. If you receive this e-mail in error, please notify the sender by
phone or email immediately and delete it!

 
-----Original Message-----
From: Mohit [mailto:mohitsikri@huawei.com] 
Sent: Wednesday, February 02, 2011 12:44 PM
To: common-dev@hadoop.apache.org
Subject: Query on Hadoop-1849 with respect to latest working trunk.

 
Hello Authors,

 
I have a query, just by seeing the code, didn't tested though, the code

changes with respect to defect Hadoop-1849 against latest

trunk[HADOOP-COMMON].

 
The defect was raised quite a long time back to make call queue size per IPC

handler configurable from core-default.xml.

 
The snippet in Server.java

 
this.maxQueueSize = handlerCount *

conf.getInt(CommonConfigurationKeys.IPC_SERVER_HANDLER_QUEUE_SIZE_KEY,

CommonConfigurationKeys. IPC_SERVER_HANDLER_QUEUE_SIZE_DEFAULT);

 
which calculates the queue size based on input configuration and if the

property not found, 100 is treated as default.

 
The property key in the java code explodes to ipc.server.handler.queue.size

but in core-default.xml it is named as ipc.server.listen.queue.size.

 
Kindly help me clear the confusion on this which is the correct property to

be configured for queue size, or it is really an issue in latest trunk.

 
Note :- I didn't found any deprecation entry for this particular property in

method addDeprecatedKeys() in Configuration.java like I found for

"fs.defaultFS" , "fs.df.interval" etc.

 
-Mohit

 
****************************************************************************

***********

This e-mail and attachments contain confidential information from HUAWEI,

which is intended only for the person or entity whose address is listed

above. Any use of the information contained herein in any way (including,

but not limited to, total or partial disclosure, reproduction, or

dissemination) by persons other than the intended recipient's) is

prohibited. If you receive this e-mail in error, please notify the sender by

phone or email immediately and delete it!

 
--Boundary_(ID_QyDXfpfVqmi453XaDb4W6w)--