[OpenSIPS-Users] Does OpenSIPS use SQLite configuration to hold on queries when database is locked?

Rodrigo Pimenta Carvalho pimenta at inatel.br
Wed Jun 29 16:49:09 CEST 2016


Dear Ionut.


Thank you very much for the precise explanation!

I got the point.


Best regards.


RODRIGO PIMENTA CARVALHO
Inatel Competence Center
Software
Ph: +55 35 3471 9200 RAMAL 979


________________________________
De: users-bounces at lists.opensips.org <users-bounces at lists.opensips.org> em nome de Ionut Ionita <ionutionita at opensips.org>
Enviado: quarta-feira, 29 de junho de 2016 11:02
Para: OpenSIPS users mailling list
Assunto: Re: [OpenSIPS-Users] Does OpenSIPS use SQLite configuration to hold on queries when database is locked?

Hi Rodrigo,

    I am the one who implemented SQLITE module. First of all I wasn't aware of the BUSY_TIMEOUT option, but still if I was I wouldn't use it and I'll explain why. Setting such a parameter will cause the OpenSIPS processes, except the one that has the lock on the database, to sleep for X ms. After those X ms the database may be locked again and you have to sleep again for that amout of seconds, even though at some point in that sleeping interval you may have had the change to get the lock. In the current implementation, all processes keep trying to get the lock, avoiding dead times, when they could have had the lock but they were sleeping. We can't do anything else while the database is locked since we need to process the current message in order to get to the next one. SQLITE has limitations based on the fact that for each query the whole database is locked, and to explain that I would like to quote official documentation:

"However, client/server database engines (such as PostgreSQL, MySQL, or Oracle) usually support a higher level of concurrency and allow multiple processes to be writing to the same database at the same time. This is possible in a client/server database because there is always a single well-controlled server process available to coordinate access. If your application has a need for a lot of concurrency, then you should consider using a client/server database. But experience suggests that most applications need much less concurrency than their designers imagine."[0]

    We would be glad to implement such a mechanism if our software would benefit from it, but in my humble opinion it would bring nothing useful for our module.

[0] http://www.sqlite.org/faq.html#q5

Regards,
Ionut Ionita
OpenSIPS Developer

On 06/28/2016 11:52 PM, Rodrigo Pimenta Carvalho wrote:


Dear OpenSIPS users,


In my software, programmed in QT (framework for C++) and handling data with a SQLite database, I have used this:


pDb.setConnectOptions("QSQLITE_BUSY_TIMEOUT=6000");


That is, if SQLite complains that "database is locked" sometime when my software tries to register some datum there, such database keeps the query paused (hold on), and then after 6 seconds let the query execute. This mechanism is transparent for my software and certify that the query will be tried every 6 seconds until it complete successfully.


Is there something similar to it in OpenSIPS ? That is, does OpenSIPS uses some kind of configuration provided by SQLite?

If not, why the developers team decided not to use such mechanism?


Any comment will be very helpful!


Thanks a lot!



RODRIGO PIMENTA CARVALHO
Inatel Competence Center
Software
Ph: +55 35 3471 9200 RAMAL 979



_______________________________________________
Users mailing list
Users at lists.opensips.org<mailto:Users at lists.opensips.org>
http://lists.opensips.org/cgi-bin/mailman/listinfo/users


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opensips.org/pipermail/users/attachments/20160629/460ba658/attachment-0001.htm>


More information about the Users mailing list