You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@wicket.apache.org by Antoine Angenieux <aa...@clinigrid.com> on 2008/10/31 11:39:07 UTC

Clustering Wicket application for Failover

Hi guys,

I need to setup a production environment with failover support. From 
what I have read in the mailing list, and what I understand of the 
diskpagestore, the only drawback of setting session replication between 
2 tomcat nodes when one of them fails and users are redirected to the 
second one is that the back button support no longer works.

I encountered another message thread in which people said that they 
switched to the old HttpSessionStore in order to have session 
replication effective, but as I understand it, the back button support 
is then not guaranteed...

So, to get to the point, can I stick with the DiskPageStore if I want 
session replication between a main and a failover node, losing 
backbutton support only when the main instance fails ? Or should I use 
the HttpSessionStore ?

Thanks all for your helps !

Cheers,

Antoine.

-- 
Antoine Angénieux
Associé

Clinigrid
5, avenue Mozart
75016 Paris, France
+336 60 21 09 18
aangenieux@clinigrid.com



---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
For additional commands, e-mail: users-help@wicket.apache.org


Re: Clustering Wicket application for Failover

Posted by Antoine Angenieux <aa...@clinigrid.com>.
I guess that's all I wanted to hear ;)

Thanks alot Igor for your quick reply as usual !

Cheers,

Antoine.

Igor Vaynberg wrote:
> wicket supports clustering with failover transparently with or without
> the diskstore.
> 
> 
> if you use httpsessionstore then nothing needs to happen. if you use
> diskstore what happens is that the latest page is replicated using
> httpsession and each node stores it into its own diskstore when it is
> replicated.
> 
> so everything works and everything is transparent.
> 
> -igor
> 
> On Fri, Oct 31, 2008 at 3:39 AM, Antoine Angenieux
> <aa...@clinigrid.com> wrote:
>> Hi guys,
>>
>> I need to setup a production environment with failover support. From what I
>> have read in the mailing list, and what I understand of the diskpagestore,
>> the only drawback of setting session replication between 2 tomcat nodes when
>> one of them fails and users are redirected to the second one is that the
>> back button support no longer works.
>>
>> I encountered another message thread in which people said that they switched
>> to the old HttpSessionStore in order to have session replication effective,
>> but as I understand it, the back button support is then not guaranteed...
>>
>> So, to get to the point, can I stick with the DiskPageStore if I want
>> session replication between a main and a failover node, losing backbutton
>> support only when the main instance fails ? Or should I use the
>> HttpSessionStore ?
>>
>> Thanks all for your helps !
>>
>> Cheers,
>>
>> Antoine.
>>
>> --
>> Antoine Angénieux
>> Associé
>>
>> Clinigrid
>> 5, avenue Mozart
>> 75016 Paris, France
>> +336 60 21 09 18
>> aangenieux@clinigrid.com
>>
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
>> For additional commands, e-mail: users-help@wicket.apache.org
>>
>>
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
> For additional commands, e-mail: users-help@wicket.apache.org
> 

-- 
Antoine Angénieux
Associé

Clinigrid
5, avenue Mozart
75016 Paris, France
+336 60 21 09 18
aangenieux@clinigrid.com



---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
For additional commands, e-mail: users-help@wicket.apache.org


Re: Clustering Wicket application for Failover

Posted by Igor Vaynberg <ig...@gmail.com>.
wicket supports clustering with failover transparently with or without
the diskstore.


if you use httpsessionstore then nothing needs to happen. if you use
diskstore what happens is that the latest page is replicated using
httpsession and each node stores it into its own diskstore when it is
replicated.

so everything works and everything is transparent.

-igor

On Fri, Oct 31, 2008 at 3:39 AM, Antoine Angenieux
<aa...@clinigrid.com> wrote:
> Hi guys,
>
> I need to setup a production environment with failover support. From what I
> have read in the mailing list, and what I understand of the diskpagestore,
> the only drawback of setting session replication between 2 tomcat nodes when
> one of them fails and users are redirected to the second one is that the
> back button support no longer works.
>
> I encountered another message thread in which people said that they switched
> to the old HttpSessionStore in order to have session replication effective,
> but as I understand it, the back button support is then not guaranteed...
>
> So, to get to the point, can I stick with the DiskPageStore if I want
> session replication between a main and a failover node, losing backbutton
> support only when the main instance fails ? Or should I use the
> HttpSessionStore ?
>
> Thanks all for your helps !
>
> Cheers,
>
> Antoine.
>
> --
> Antoine Angénieux
> Associé
>
> Clinigrid
> 5, avenue Mozart
> 75016 Paris, France
> +336 60 21 09 18
> aangenieux@clinigrid.com
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
> For additional commands, e-mail: users-help@wicket.apache.org
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@wicket.apache.org
For additional commands, e-mail: users-help@wicket.apache.org