On Mon, Dec 14, 2009 at 11:57 AM, Ruwan Linton <[email protected]> wrote:
> Supun,
>
> What is the issue that we are trying to solve here? I am sorry I am
> confused :-( .
>
> ESB (to be more specific the nhttp transport) uses a connection pool to
> support keep-alive, if you want each and every request going out from
> ESB lookup the DNS and create a new connection, you may disable
> keep-alive by putting a property named http.connection.disable.keepalive
> into the nhttp.properties file.
>
> Yet again if a DNS change has been occurred while trying to reuse an
> existing connection, it will fail and the connection will be reset and
> there is a retry on the endpoint which is configurable and will connect
> to the correct endpoint.
>
> I am sorry, I don't see any issue here :-(
>

Yes I don't see an issue here as well. I just wanted to make sure as I
said in my first mail :).

Supun..

> Thanks,
> Ruwan
>
> Supun Kamburugamuwa wrote:
>> Actually this is not the server side of ESB that I'm talking about. It
>> is the client side of the ESB.
>>
>> Anyway if we can test a scenario like this it will be great.
>>
>> >From code here is what should happen.
>>
>> We have a connection pool with host name and ip address. So even
>> though the DNS server is configured to point to a new IP address for a
>> host name, the connection pool doesn't change. For these connections
>> to become un-usable the back end server which hosted the services
>> should be shut down. Until that ESB can still send messages to the old
>> server because it is taking connections from a connection pool.
>>
>> When ESB creates new connections after the DNS change, it will do it 
>> correctly.
>>
>> Thanks,
>> Supun..
>>
>> On Sat, Dec 12, 2009 at 12:15 PM, Senaka Fernando <[email protected]> wrote:
>>
>>> Hi Supun,
>>>
>>> If you have configured the server to use the host name instead of the ip
>>> address, it should work without any issues. If you did use the ip address as
>>> the host name or didn't specify it, the change in ip address could cause
>>> some aspects not to work, such as calculated EPRs in EventBrokers.
>>>
>>> Thanks,
>>> Senaka.
>>>
>>> On Sat, Dec 12, 2009 at 9:35 AM, Afkham Azeez <[email protected]> wrote:
>>>
>>>> Yeah, not only for ESB, but for any generic client-server system this
>>>> should not cause any problems as long as the first machine is kept online
>>>> until all requests it received have been served. If that machine is taken
>>>> offline immediately, the request which were in flight and which were being
>>>> processed will fail. So there should be a graceful mechanism to do this,
>>>> like our graceful shutdown procedure. Subsequent requests should be
>>>> successfully forwarded to the new machine though.
>>>> Azeez
>>>>
>>>> On Sat, Dec 12, 2009 at 5:23 AM, Supun Kamburugamuwa <[email protected]>
>>>> wrote:
>>>>
>>>>> Hi,
>>>>>
>>>>> Suppose we have several back end services and ESB refer to them using
>>>>> a URL. These URLs are DNS resolvable URLs not IP address. Suppose we
>>>>> substitute a new physical machine with a different IP and configure
>>>>> the DNS server to point to the new IP for the same old URL. The
>>>>> question is what will happen?
>>>>>
>>>>> From the code it seems ESB will work correctly. I thought of bringing
>>>>> this question up to make sure that it will work.
>>>>>
>>>>> Thanks,
>>>>> Supun..
>>>>>
>>>>> _______________________________________________
>>>>> Carbon-dev mailing list
>>>>> [email protected]
>>>>> https://mail.wso2.org/cgi-bin/mailman/listinfo/carbon-dev
>>>>>
>>>>
>>>> --
>>>> --
>>>> Afkham Azeez
>>>> [email protected]
>>>> WSO2 Inc. http://wso2.com
>>>> Blog: http://afkham.org
>>>>
>>>> _______________________________________________
>>>> Carbon-dev mailing list
>>>> [email protected]
>>>> https://mail.wso2.org/cgi-bin/mailman/listinfo/carbon-dev
>>>>
>>>>
>>>
>>> --
>>> Senaka Fernando
>>> Software Engineer
>>> WSO2 Inc.
>>> E-mail: [email protected];  Mobile: +94 77 322 1818
>>>
>>> http://www.wso2.com/ - "The Open Source SOA Company"
>>>
>>> _______________________________________________
>>> Carbon-dev mailing list
>>> [email protected]
>>> https://mail.wso2.org/cgi-bin/mailman/listinfo/carbon-dev
>>>
>>>
>>>
>>
>> _______________________________________________
>> Carbon-dev mailing list
>> [email protected]
>> https://mail.wso2.org/cgi-bin/mailman/listinfo/carbon-dev
>>
>>
>
>
> --
> Ruwan Linton
> Technical Lead & Product Manager; WSO2 ESB; http://wso2.org/esb
> WSO2 Inc.; http://wso2.org
> email: [email protected]; cell: +94 77 341 3097
> blog: http://blog.ruwan.org
>
>
>
> _______________________________________________
> Carbon-dev mailing list
> [email protected]
> https://mail.wso2.org/cgi-bin/mailman/listinfo/carbon-dev
>

_______________________________________________
Carbon-dev mailing list
[email protected]
https://mail.wso2.org/cgi-bin/mailman/listinfo/carbon-dev

Reply via email to