In my script I send all “100 Trying” responses manually* and use the 0x01 flag
when calling t_relay so that it will not send its own “100 Giving a try”
response, as per the documentation [1].
Unfortunately, in 1.11.7 and 1.11.8 this flag seems to have stopped working
which results in multiple
Hi.
In addiction to my last message, the same problem also exists to the following
configuration:
modparam("uri", "use_uri_table", 0)
modparam("uri", "db_url", "sqlite:///usr/local/opensips/db/sisc.sqlite") #
CUSTOMIZE ME
If use_uri_table is equal to zero, we must comment the line that
Hi.
Thanks Daniel Fússia, a coworker in my office, now I'm sending more details
about the memory leak we saw in OpenSIPS 2.2 (newest commit from today):
The following configuration doesn't causes memory leaks:
modparam("dialog", "db_mode", 1)
modparam("dialog", "db_url",
Hi.
We discovered another memory leak in OpenSIPS 2.2, even using newest SQLite.
Now the issue doesn't relate to the data base. There is a issue related with a
parser.
In few minutes I will post here more details, with valgrind log.
Regards.
RODRIGO PIMENTA CARVALHO
Inatel Competence
Hi Eric.
Probably not. Because I still don't know what is a federated-sip. And I didn't
have to take control of RTPs in opensips script.
However, a coworker in my office will check these details and help us to
conclude more things about it.
Is there a quick way to check if someone is using
Hey Rodrigo,
Are you running https://github.com/etamme/federated-sip by chance? Your
use of the PCRE module made me think you might be. I run federated-sip
and I do use sqlite3 with opensips - my current sqlite version is:
sqlite-3.7.17-4.el7.x86_64
I do not know that I have memory leaks
Thank you Ionut.
We will try it so.
Today morning, we noticed that OpenSIPS 2.2 while running and using SQLite,
without online clients, without registers and without calls, causes a memory
leak. That is, OpenSIPS even without any SIP request causes a memory leak due
to the use of SQLite.
Hi Rodrigo,
Pushed a fix both into 2.2[0] and master[1] branches. If you still
think sqlite leaks even with this fix,
please feel free to open an issue on github.
[0]
https://github.com/OpenSIPS/opensips/commit/c1aa55e73f7c56ea4894997aeb7f25a03cb289cf
[1]
Hi Liviu.
Very good.
We will see the resolution process.
Thank you very much!
Regards.
RODRIGO PIMENTA CARVALHO
Inatel Competence Center
Software
Ph: +55 35 3471 9200 RAMAL 979
De: users-boun...@lists.opensips.org em
Hi Rodrigo!
A GitHub issue [1] regarding this leak was just reported today by Eric,
so you can track the resolution process over there! You can even
subscribe to that ticket if you have an account, in order to receive emails.
[1]: https://github.com/OpenSIPS/opensips/issues/911
Liviu Chircu
Hi.
People from my team is investigating a memory leak related to OpenSIPS 2.2.
As I had commented in another discussion in the past, it seems that the problem
comes from SQLite we are using as the Registrar for our OpenSIPS 2.2.
For example, a script opensips.cfg that doesn't use SQLite
Hi Rahul,
Even if it is a reply, use $rm - it will extract the method from the
cseq hdr if reply.
Regards,
Bogdan-Andrei Iancu
OpenSIPS Founder and Developer
http://www.opensips-solutions.com
On 16.06.2016 19:24, Gupta, Rahul wrote:
Nevermind, I can just use is_method(). If there is
Topic moved on devel list:
http://lists.opensips.org/pipermail/devel/2016-June/020379.html
Bogdan-Andrei Iancu
OpenSIPS Founder and Developer
http://www.opensips-solutions.com
On 16.06.2016 16:55, Arsen wrote:
Hi guys,
We have upgraded to 2.2 but seems it's crashing occasionally.
Please take
13 matches
Mail list logo