Mailing-List: contact users-help@cxf.apache.org; run by ezmlm
Precedence: bulk
Reply-To: users@cxf.apache.org
Received-SPF: pass (athena.apache.org: domain of diego.vera@capgemini-sdm.com
 designates 192.76.162.230 as permitted sender)
Content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Subject: AW: Web Service Synchronization
Date: Tue, 25 Aug 2009 14:11:00 +0200
Message-ID: <D4C7AEF079ED754EA6182024DE2DB0780425A10D@mucmail4.sdm.de>
In-Reply-To: <61b5d9410908250456p5556755gd9f697f04872678c@mail.gmail.com>
Thread-Topic: Web Service Synchronization
Thread-Index: Acole3tQ1G2pNqwITMeyVQMdMGfNVgAAFkxA
References: <D4C7AEF079ED754EA6182024DE2DB07804259FCC@mucmail4.sdm.de>
 <25132477.post@talk.nabble.com>
 <61b5d9410908250456p5556755gd9f697f04872678c@mail.gmail.com>
From: <diego.vera@capgemini-sdm.com>
To: <users@cxf.apache.org>

Hi all,

What I exactly mean is the concurrency control in the web service. I =
have a web service where users write information that they store in a =
list, and this list could be very large, that is why I see if I do not =
control the concurrence to this web method, some information could be =
stored in the profile of one users and the other in the profile of the =
other user, but the list should be stored in the profile of the right =
user, without concurrence control it could happen that the information =
is stored for the incorrect user. The solution proposed by Glen could =
work (locking the data base), however every entry in the list is =
independent that is why I was thinking that controlling the concurrency =
in the web method could be a better solution. Is there a better =
approach?

Thanks in advance,

-Diego Vera

-----Urspr=FCngliche Nachricht-----
Von: Benson Margulies [mailto:bimargulies@gmail.com]=20
Gesendet: Dienstag, 25. August 2009 13:57
An: users@cxf.apache.org
Betreff: Re: Web Service Synchronization

Let's be a bit more specific. Are you concerned with the reentrancy of =
your implementation or of the client stubs?

On Tue, Aug 25, 2009 at 7:17 AM, Glen Mazza<glen.mazza@gmail.com> wrote:
>
> Synchronous/Asynchronous has a different meaning with web services,=20
> but using your standard definition of synchronous, I can't imagine web =

> methods being synchronized by default, because that would slow down=20
> read-only web methods such as getCapitalOfBritain that don't require=20
> synchronization. =A0I'm unsure how the lack of synchronization would=20
> prevent you from logging who accesses your web service, and for=20
> writable web services that access a database, I would first check to=20
> make sure that the database would not take care of the synchronization =

> for you anyway (i.e., some databases lock during=20
> insert/updates/deletes, not allowing other calls to proceed until=20
> finished.)
>
> HTH,
> Glen
>
>
> diego.vera wrote:
>>
>> Hi all,
>>
>> I got a question about cxf web services. I require synchronization in =

>> the web methods since the users that are calling them should be=20
>> logged and of couse is important that a web method call fnish before=20
>> another user calls the same method. The question is: are by default=20
>> the call of the web methods synchonized? Or should I synchronize the =
access?
>>
>> Thanks a lot,
>>
>> -Diego Vera
>>
>
> --
> View this message in context:=20
> http://www.nabble.com/Web-Service-Synchronization-tp25131659p25132477.
> html Sent from the cxf-user mailing list archive at Nabble.com.
>
>