I don't know why the code is working on the 2000 machine. you are going
through port 443 so it should be secure and it does not verify the security.
I have not been able to force mine to do that. have you tried adding a
security certificate file?

----- Original Message -----
From: "Nate Nielsen" <nnielsen_at_cfgod.com>
To: "curl with PHP" <curl-and-php_at_cool.haxx.se>
Sent: Saturday, November 26, 2005 2:26 PM
Subject: Re: site login problem with PHP + cURL

> I'm adding additional bounty to my problem for anyone who can help me fix
> it.
>
> I'll send you a Rio Cali Sport 256 MB MP3 player - new in box. Comes with
> the sport ear buds, arm strap, few other accessories. You can google it
> for the details of it. Great for a stocking stuffer or as an extra gadget
> for kicks (as if we all dont have enough). =D
>
> To restate the problem for those that have forgotten, one server is
> working (win 2k, other is not win 2k3) - same code, same versions of
> everything. I've created a page for you to view the code / the verbose
> data / the phpInfo() for the two servers. Its a script to auto-login to
> a site - not particularly yahoo, the problem occurs with all cookie based
> logins, so using this yahoo script as an example. Again, the win2k server
> ==IS== working with that code - so its dosent appear to be a code issue.
>
> You can view the debug information here : http://70.84.198.254/index.html
>
> thanks again,
>
> Nate Nielsen
> nnielsen_at_cfgod.com
>
> ----- Original Message -----
> From: "Daniel Stenberg" <daniel_at_haxx.se>
> To: "curl with PHP" <curl-and-php_at_cool.haxx.se>
> Sent: Wednesday, November 23, 2005 3:23 AM
> Subject: Re: site login problem with PHP + cURL
>
>
>> On Tue, 22 Nov 2005, Nate Nielsen wrote:
>>
>>> that is a great question, i have no idea :)
>>>
>>> as you can see the code is literally identical except a drive letter for
>>> the cookie file. so, i am totally confused why it is responding in that
>>> way as well. =S
>>>
>>> the working version is getting a LOT more data back in the verbose
>>> screen, including a much much larger cookie that gets set when the login
>>> is a success.
>>
>> Well, the code obviously does a GET first so the 'working' verbose trace
>> is not complete.
>>
>> Also, since the traces don't reveal how the cookies end up in the cookie
>> jar file, we can't see why/how the 'broken' one gets a much shorter
>> cookie set.
>>
>> --
>> Commercial curl and libcurl Technical Support: http://haxx.se/curl.html
>> _______________________________________________
>> http://cool.haxx.se/cgi-bin/mailman/listinfo/curl-and-php
>
> _______________________________________________
> http://cool.haxx.se/cgi-bin/mailman/listinfo/curl-and-php
>
_______________________________________________
http://cool.haxx.se/cgi-bin/mailman/listinfo/curl-and-php
Received on 2005-11-27