[Devel] Callling add_rr_param from failure_route[x]

[Bogdan-Andrei Iancu]

indeed, I would also like to get rid of this limitation. I cannot 
estimate when it will be fixed - first I will have to come up with a 
solution and than to implement it. It might appear or not in the next 
release.

regards,
Bogdan

Dmitry Isakbayev wrote:

> That is too bad.  Do you have any idea when you will get a chance 
> TODO'it? I will gladly test it for you :)
>
> Thanks,
> Dmitry
>
>
> ----- Original Message ----- From: "Bogdan-Andrei Iancu" 
> <bogdan at voice-system.ro>
> To: "Dmitry Isakbayev" <isakdim at gmail.com>
> Cc: <devel at openser.org>
> Sent: Tuesday, September 13, 2005 11:03 AM
> Subject: Re: [Devel] Callling add_rr_param from failure_route[x]
>
>
>> it will be safe, but only from branch routes called from request 
>> routes. if you do a t_relay from failure route (and getting called a 
>> branch route from failure routes), you will have the same problem as 
>> before :(.
>>
>> regards,
>> bogdan
>>
>> Dmitry Isakbayev wrote:
>>
>>> Hi Bogdan,
>>>
>>> What about calling "add_rr_param" from a per-branch route?
>>>
>>> Thanks,
>>> Dmitry
>>>
>>> ----- Original Message ----- From: "Bogdan-Andrei Iancu" 
>>> <bogdan at voice-system.ro>
>>> To: "Dmitry Isakbayev" <isakdim at gmail.com>
>>> Cc: <devel at openser.org>
>>> Sent: Tuesday, September 13, 2005 10:36 AM
>>> Subject: Re: [Devel] Callling add_rr_param from failure_route[x]
>>>
>>>
>>>> Hi Dimitry,
>>>>
>>>> for the moment "add_rr_param" cannot be called from failure route 
>>>> due implementation limitations - to be able to add params after the 
>>>> RR was inserted, a pointer to the lump is kept in a static 
>>>> variable; when adding params, the lump body is modified.
>>>> In failure route there are tow problems:
>>>>    1) you cannot use anymore the static variable to identify the RR 
>>>> lump; maybe a solution will be to set type for the RR lumps and try 
>>>> to find them;
>>>>    2) altering a lump from shm mem is very problematic - all of 
>>>> them are allocated in a single mem chunk; no idea how to go around 
>>>> this.....for the moment.....
>>>>
>>>> but this is on my TODO list :)
>>>>
>>>> regards,
>>>> bogdan
>>>>
>>>> Dmitry Isakbayev wrote:
>>>>
>>>>> Hi,
>>>>>
>>>>> What is the reason for not allowing to call "add_rr_param" from a 
>>>>> failure_route block?
>>>>>
>>>>> Thanks,
>>>>> Dmitry
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Devel mailing list
>>>>> Devel at openser.org
>>>>> http://openser.org/cgi-bin/mailman/listinfo/devel
>>>>>
>>>>
>>>
>>
>
>