hc-httpclient-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Todd W Lainhart <lainh...@us.ibm.com>
Subject Re: HttpClient with CircularRedirectException
Date Tue, 28 Oct 2014 12:58:32 GMT
> Is it better i go ahead and raise a jira for this one.

I wouldn't open a bug unless you can point to the code at fault.



Todd Lainhart
Senior Software Engineer - ExperienceOne, Digital Analytics

Phone: 1-978-899-4705 | Mobile: 1-508-331-1110
E-mail: lainhart@us.ibm.com
550 King St
Littleton, MA 01460-1250
United States




From:   srihari na <namasrihari@gmail.com>
To:     HttpClient User Discussion <httpclient-users@hc.apache.org>
Date:   10/28/2014 05:47 AM
Subject:        Re: HttpClient with CircularRedirectException



I am still not convinced that it is a server side problem because other
clients are able to handle the redirects to some sort of solution. I
believe there might be some issue in the redirect strategy of the HTTP
client. Is it better i go ahead and raise a jira for this one.

On Wed, Oct 15, 2014 at 6:58 PM, Todd W Lainhart <lainhart@us.ibm.com>
wrote:

> srihari na <namasrihari@gmail.com> wrote on 10/15/2014 04:38:23 AM:
>
> > My concern is why the Http Client library is hitting Circular redirect
> or
> > maximum redirect issue where as normal rest client or simple java
> program
> > does not have any issue while invoking the URL.
> >
>
> An obvious guess is that the other clients are following the redirects 
to
> some sort of resolution.  Take a look at network traffic between the
> client and the server to confirm this.  An excessive number of redirects
> (circular or otherwise) might be an indication of some sort of problem 
on
> the server, which the circularRedirects property can alert you to.  From
> RequestConfig:
>
>     /**
>      * Determines whether circular redirects (redirects to the same
> location) should
>      * be allowed. The HTTP spec is not sufficiently clear whether
> circular redirects
>      * are permitted, therefore optionally they can be enabled
>      * <p/>
>      * Default: <code>false</code>
>      */




-- 
Regards,
Srihari NA


Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message