[rudder-users] cf-agent aborted on defined class "could_not_download_uuid

[Olivier Desport]

Le 02/12/2013 12:49, Matthieu CERDA a écrit :
> Le 02/12/2013 11:31, Matthieu CERDA a écrit :
>>> Hello Olivier,
>>>
>>> According to curl's documentation, it seems that he gets an HTTP 4XX 
>>> error with Rudder.
>>>
>>> Can you please tell me which OS are you using, and try to run 
>>> "/usr/bin/curl --proxy '' -o "/var/rudder/tmp/uuid.txt" 
>>> http://172.31.0.61/uuid" manually ?
>> The OS is debian.
>>
>> Output for /usr/bin/curl --proxy '' -o "/var/rudder/tmp/uuid.txt" 
>> http://172.31.0.61/uuid :
>>
>> ################################################
>> % Total    % Received % Xferd  Average Speed   Time    Time Time  
>> Current
>>                                  Dload  Upload   Total   Spent Left  
>> Speed
>> 100  3159  100  3159    0     0   206k      0 --:--:-- --:--:-- 
>> --:--:--  220k
>> ###########################################
>>>
>>> Also, do you have a proxy in your environment ? (especially a 
>>> transparent one)
>> I have a no-transparent proxy which is defined in $http_proxy and 
>> $no_proxy. The client and the server are on the same subnet, so the 
>> proxy shouldn't be used by rudder.
> OK, so the invocation looks good manually too.
>
> Can you please send me the output of "grep <NODE IP ADDRESS> 
> /var/log/rudder/apache2/access.log", to be executed on the Rudder 
> server ? (replacing the node ip address by the IP of the node unable 
> to be registered)
The output of the grep is empty.
>
> I'm especially searching for the actual HTTP error code sent back to cURL.
>
> Regards,
> -- 
> Matthieu CERDA

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.rudder-project.org/pipermail/rudder-users/attachments/20131202/70747708/attachment.html>