Hi Matevz,
I had a deeper look at the issue. Actually the 'host' field is used
but not set by the low-level
protocol, and there is no caching of protocol objects, hence of
Entity objects.
I have tried to reproduce it with the simplest setup (one server,
two clients connecting with
proxies generated with the same certificate from different
machines) but I did not manage,
the Entity content is consistent with the expectations.
Can you say more on the setup where you see this?
Cheers, Gerri
On 07/03/2014 17:59, Matevz Tadel wrote:
> Ping? :)
>
> Matevz
>
> On 02/26/14 23:57, Gerardo Ganis wrote:
>>
>> Hi Matevz,
>>
>> It smells like a bug.
>> I will have a look ASAP, but likely not before tomorrow.
>>
>> Cheers, Gerri
>>
>> On 27 Feb 2014, at 00:56, Matevz Tadel <[log in to unmask]> wrote:
>>
>>> Hi Gerri,
>>>
>>> I just noticed that if one comes to the same server from different
>>> hosts (with different proxies but with the same DN), the host part
>>> of the entity remains what it was on the first connection. I use the
>>> entity.host as preferred client host source for monitoring (as it
>>> does reverse name lookup).
>>>
>>> Does this make any sense? I never observed this before so I figured
>>> I'd ask before I try to read this out of the code ;)
>>>
>>> Cheers,
>>> Matevz
>
--
+-----------------------------------------------------------------+
Gerardo GANIS CERN, PH Dept, SFT group, CH 1211 Geneve 23
room: 2-R-023, email: [log in to unmask]
tel/cell: (+41) 22 7676439 / 76 4871128
+-----------------------------------------------------------------+
########################################################################
Use REPLY-ALL to reply to list
To unsubscribe from the XROOTD-DEV list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=XROOTD-DEV&A=1
|