[SR-Users] Question about setbflag and isbflagset usage

[Володимир Іванець]

Hello again,

I've just noticed that if new branch is created after using *t_suspend* and
*t_continue*, branch flag can not be used. Not sure if it can not be set or
checked. Doues anyone know about potential problems with such configuration?

Thanks!

вт, 28 трав. 2019 о 19:43 Володимир Іванець <volodyaivanets at gmail.com> пише:

> Hello Alex,
>
> Thank you for the clarification. Now I understand this behavior.
>
> пт, 24 трав. 2019 о 22:07 Alex Balashov <abalashov at evaristesys.com> пише:
>
>> Branches are a creature of transactions. It follows that “branch flags”
>> are transaction-persistent flags at the branch level, rather than scoped to
>> the transaction as a whole.
>>
>> ACKs constitute a separate transaction.
>>
>> —
>> Sent from mobile, with due apologies for brevity and errors.
>>
>> On May 24, 2019, at 2:45 PM, Sergiu Pojoga <pojogas at gmail.com> wrote:
>>
>> Then.. I don't know, lol
>>
>> But the fact that ACKs belong or not to even different transactions
>> (depending on whether positive or negative reply) must be playing a role
>> here.
>>
>> Good luck.
>>
>> On Fri, May 24, 2019, 12:11 PM Володимир Іванець, <
>> volodyaivanets at gmail.com> wrote:
>>
>>> Hello Sergiu,
>>>
>>> We allow multiple contacts to register on single account. They can have
>>> different properties.  On step 1 I'm looking for something specific which
>>> can be different for different contacts. If I understood correctly
>>> *setbflag* and *isbflagset* can set and check flags for different
>>> branches independently.
>>>
>>> Thank you!
>>>
>>> пт, 24 трав. 2019 о 18:14 Sergiu Pojoga <pojogas at gmail.com> пише:
>>>
>>>> I might be wrong, but isn't ACK part of a transaction? If so, you'd
>>>> need to use *setflag*/*isflagset*
>>>>
>>>> Cheers.
>>>>
>>>> On Fri, May 24, 2019 at 10:03 AM Володимир Іванець <
>>>> volodyaivanets at gmail.com> wrote:
>>>>
>>>>> Hello everyone!
>>>>>
>>>>> I'm not sure if I correctly understood the description of functions
>>>>> *isbflagset* and *isbflagset* so once again I'm asking for help :)
>>>>>
>>>>> Here is what I do:
>>>>> 1. Asterisk sends INVITE to Kamailio. Kamailio performs lookup of
>>>>> available contacts and then few checks in branch_route. I'm adding *setbflag("17",
>>>>> $T_branch_idx);* there.
>>>>> 2. Later in route[NATMANAGE] I added *if (isbflagset("17",
>>>>> $T_branch_idx)) {* and additional logic. *isbflagset* results in
>>>>> *true* (for same INVITE).
>>>>> 3. Finally one of UACs responds with 200 OK. Asterisk generates ACK
>>>>> packet that gets in route[NATMANAGE]. *isbflagset* results in *false*.
>>>>>
>>>>> So my question is: shouldn't *isbflagset* be *true* for ACK too since
>>>>> it's the same branch or I interpreted these fenctions incorrectly?
>>>>>
>>>>> Thanks a lot!
>>>>> _______________________________________________
>>>>> Kamailio (SER) - Users Mailing List
>>>>> sr-users at lists.kamailio.org
>>>>> https://lists.kamailio.org/cgi-bin/mailman/listinfo/sr-users
>>>>>
>>>> _______________________________________________
>>>> Kamailio (SER) - Users Mailing List
>>>> sr-users at lists.kamailio.org
>>>> https://lists.kamailio.org/cgi-bin/mailman/listinfo/sr-users
>>>>
>>> _______________________________________________
>>> Kamailio (SER) - Users Mailing List
>>> sr-users at lists.kamailio.org
>>> https://lists.kamailio.org/cgi-bin/mailman/listinfo/sr-users
>>>
>> _______________________________________________
>> Kamailio (SER) - Users Mailing List
>> sr-users at lists.kamailio.org
>> https://lists.kamailio.org/cgi-bin/mailman/listinfo/sr-users
>>
>> _______________________________________________
>> Kamailio (SER) - Users Mailing List
>> sr-users at lists.kamailio.org
>> https://lists.kamailio.org/cgi-bin/mailman/listinfo/sr-users
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.kamailio.org/pipermail/sr-users/attachments/20190607/5ffaed3e/attachment.html>