Mailing-List: contact dev-help@harmony.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@harmony.apache.org
Received-SPF: pass (herse.apache.org: domain of t.p.ellison@gmail.com
 designates 66.249.92.173 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=beta;
        h=received:message-id:date:from:user-agent:mime-version:to:subject:references:in-reply-to:x-enigmail-version:content-type:content-transfer-encoding;
        b=F8256cpcLX+6xcGKUtzuTxa/xAWt8GTEv0/ElxXCiyTrsCepUnPkhqhpz+SI2YK+31BtSLW6Z6Tw+MhqYxQPPGxsDzYEozP6VBiflcJUJIsw4nKWUFyAkRrAIp7xnE7MOwYeXRcIaflxU77c9nZCqmRaE2by4W8+zej/KeRrKOs=
Message-ID: <4625D7F2.3080603@gmail.com>
Date: Wed, 18 Apr 2007 09:33:54 +0100
From: Tim Ellison <t.p.ellison@gmail.com>
User-Agent: Thunderbird 1.5.0.10 (Windows/20070221)
MIME-Version: 1.0
To: dev@harmony.apache.org
Subject: Re: [classlib][luni] new test in tests.api.java.net.SocketTest failed
 on my Windows XP sp2
References: <7273946b0704170150u10fe9f13vd430f0a3aa0399e2@mail.gmail.com>
	 <46248CBA.7010702@gmail.com>
	 <7273946b0704170216k43e3165cx30a19eb0ce27ea92@mail.gmail.com>
	 <462498F7.6050300@gmail.com>
 <4624B462.1030607@gmail.com>	 <3b3f27c60704171505g21278641gf528578f6da3608@mail.gmail.com>
	 <211709bc0704172011v633a552bubab887221a30a680@mail.gmail.com>
	 <5c8e69f0704172025r773eba76w40f068d060026a1@mail.gmail.com>
 <4d0b24970704180044y40e6d391q7d0e6ad4586e20ec@mail.gmail.com>
In-Reply-To: <4d0b24970704180044y40e6d391q7d0e6ad4586e20ec@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit

Andrew Zhang wrote:
> On 4/18/07, LvJimmy,Jing <firepure@gmail.com> wrote:
>>
>> This method is good, but not perfect, for
>> 1. getNextPortsForUDP get a free port by a mechanism "get and
>> release", but release the port may not be successful (e.g, in linux)
>>
> 2. even it releases successfully, during the period between utility
>> method "release" and serversocket reuse it, there may be some other
>> program reuse the port, which cause the test fail again.
> 
> 
> Open and close is not safe in theory, as we discussed before.
> Let's stop getNextPort, and stick to use port 0. :) I think we have agreed
> that a long time ago.

Yes.

> 3. this may cost much more time to get a port.
>>
> The perfect solution for this is re-write the testcases, create
>> ServerSocket with port 0, and then get the local port for socket to
>> connect. However I guess it may cause a lot of affect.
> 
> 
> yes, but we can do it lazily and gradually. The important thing is that we
> should not use the buggy getNextPort any more....

I agree, how about we mark it @deprecated to help remind people?

Regards,
Tim