You are viewing a plain text version of this content. The canonical link for it is here.
Posted to httpclient-users@hc.apache.org by Derick Fernando <de...@gmail.com> on 2012/03/29 00:18:05 UTC

ConnManagerPNames.html#TIMEOUT vs. CoreConnectionPNames.html#CONNECTION_TIMEOUT

Hello,

I was attempting to set the timeout on a connection manager in HTTP Client
4.1.

The constant ConnManagerPNames.html#TIMEOUT is deprecated in favor of
CoreConnectionPNames.html#CONNECTION_TIMEOUT.

However the underlying value for
CoreConnectionPNames.html#CONNECTION_TIMEOUT is "http.connection.timeout"
where as CoreConnectionPNames.html#CONNECTION_TIMEOUT corresponds to
"http.conn-man.timeout.

I wanted to make sure this is correct and intentional. It makes sense
logically since the connection timeout means that something will eventually
be returned to an exhausted pool.

Thanks,
Derick

Re: ConnManagerPNames.html#TIMEOUT vs. CoreConnectionPNames.html#CONNECTION_TIMEOUT

Posted by Derick Fernando <de...@gmail.com>.
Hi Oleg,

Thanks a lot for looking into this.

Derick

On Thu, Mar 29, 2012 at 3:10 PM, Oleg Kalnichevski <ol...@apache.org> wrote:

> On Wed, 2012-03-28 at 15:18 -0700, Derick Fernando wrote:
> > Hello,
> >
> > I was attempting to set the timeout on a connection manager in HTTP
> Client
> > 4.1.
> >
> > The constant ConnManagerPNames.html#TIMEOUT is deprecated in favor of
> > CoreConnectionPNames.html#CONNECTION_TIMEOUT.
> >
> > However the underlying value for
> > CoreConnectionPNames.html#CONNECTION_TIMEOUT is "http.connection.timeout"
> > where as CoreConnectionPNames.html#CONNECTION_TIMEOUT corresponds to
> > "http.conn-man.timeout.
> >
> > I wanted to make sure this is correct and intentional. It makes sense
> > logically since the connection timeout means that something will
> eventually
> > be returned to an exhausted pool.
> >
> > Thanks,
> > Derick
>
> Hi Derick
>
> Yes, this is intentional. I personally think there is no good reason to
> differentiate connect timeout and timeout for connection lease
> operation. Having said that in HttpClient 4.2 one override the default
> timeout value by using ClientPNames.CONN_MANAGER_TIMEOUT parameter.
>
> Hope this helps
>
> Oleg
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: httpclient-users-unsubscribe@hc.apache.org
> For additional commands, e-mail: httpclient-users-help@hc.apache.org
>
>

Re: ConnManagerPNames.html#TIMEOUT vs. CoreConnectionPNames.html#CONNECTION_TIMEOUT

Posted by Oleg Kalnichevski <ol...@apache.org>.
On Wed, 2012-03-28 at 15:18 -0700, Derick Fernando wrote:
> Hello,
> 
> I was attempting to set the timeout on a connection manager in HTTP Client
> 4.1.
> 
> The constant ConnManagerPNames.html#TIMEOUT is deprecated in favor of
> CoreConnectionPNames.html#CONNECTION_TIMEOUT.
> 
> However the underlying value for
> CoreConnectionPNames.html#CONNECTION_TIMEOUT is "http.connection.timeout"
> where as CoreConnectionPNames.html#CONNECTION_TIMEOUT corresponds to
> "http.conn-man.timeout.
> 
> I wanted to make sure this is correct and intentional. It makes sense
> logically since the connection timeout means that something will eventually
> be returned to an exhausted pool.
> 
> Thanks,
> Derick

Hi Derick

Yes, this is intentional. I personally think there is no good reason to
differentiate connect timeout and timeout for connection lease
operation. Having said that in HttpClient 4.2 one override the default
timeout value by using ClientPNames.CONN_MANAGER_TIMEOUT parameter.

Hope this helps

Oleg 


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


Re: ConnManagerPNames.html#TIMEOUT vs. CoreConnectionPNames.html#CONNECTION_TIMEOUT

Posted by Derick Fernando <de...@gmail.com>.
Sorry, there is a typo in the original, it should read:

However the underlying value for

ConnManagerPNames.html#TIMEOUT is "http.conn-man.timeout" where as
CoreConnectionPNames.html#CONNECTION_TIMEOUT corresponds to
"http.connection.timeout".

Thanks

On Wed, Mar 28, 2012 at 3:18 PM, Derick Fernando <de...@gmail.com> wrote:

> Hello,
>
> I was attempting to set the timeout on a connection manager in HTTP Client
> 4.1.
>
> The constant ConnManagerPNames.html#TIMEOUT is deprecated in favor of
> CoreConnectionPNames.html#CONNECTION_TIMEOUT.
>
> However the underlying value for
> CoreConnectionPNames.html#CONNECTION_TIMEOUT is "http.connection.timeout"
> where as CoreConnectionPNames.html#CONNECTION_TIMEOUT corresponds to
> "http.conn-man.timeout.
>
> I wanted to make sure this is correct and intentional. It makes sense
> logically since the connection timeout means that something will eventually
> be returned to an exhausted pool.
>
> Thanks,
> Derick
>
>
>
>
>
>
>