Hello, On 9/30/13 11:39 AM, Charles Chance wrote: Hi, Not sure whether it is appropriate to get this in before the upcoming freeze or wait until after. all new code should go on git before freeze -- unless there is a major change that will require a lot of coding in other parts to get it coherent, there is no rule of stopping new features from getting merged. I have the additions to htable module nearly ready for review. What is the most convenient - patch or private branch? Typically I review better patches, if they are not big. But I can look at the patches on the gitweb as well if it is more convenient for you. Cheers, Daniel Just completing tests but should be available tomorrow or day after. I have kept disruption to existing code to a minimum and added all functions into a separate file. New functionality is disabled by default and must be enabled on a per table basis. What are everyone's thoughts? Maybe it is too close to deadline for such a new feature? Very happy to try though :) Cheers, Charles www.sipcentric.com Follow us on twitter @sipcentric <http://twitter.com/sipcentric> Sipcentric Ltd. Company registered in England & Wales no. 7365592. Registered office: Unit 10 iBIC, Birmingham Science Park, Holt Court South, Birmingham B7 4EJ. _______________________________________________ sr-dev mailing listsr-dev@lists.sip-router.orghttp://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-dev -- Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda Kamailio Advanced Trainings - Berlin, Nov 25-28; Miami, Nov 18-20, 2013 - more details about Kamailio trainings at http://www.asipto.com - _______________________________________________ sr-dev mailing list sr-dev(a)lists.sip-router.org http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-dev

Hi Charles, thanks for the patch. Few remarks: - for strict C compatibility, would be good to declare variables at the beginning of the functions or block, not in between statements - i think the next variable has to be initialized to NULL in modules/htable/ht_dmq.c, like: dmq_peer_t* ht_dmq_peer = NULL; Otherwise it can happen that enable_dmq is not set, but dmqreplicate per hash table is set and could cause a crash. Eventually, when enable_dmq is not 1, then make dmqreplicate per hash table 0, no matter what value is given -- I guess this will require that enable_dmq parameter has to be set before defining a hash table. Also, in such case, could be useful to have a log message to warn about setting replication in hash table definition, but the module-level parameter is not enabled. If you redo the entire patch, the line with $Id$ at the beginning of files can be removed, it is a reminiscent from SVN/CVS times, with no real usage now - this is just beautification. The rest seems ok, so can be committed. Cheers, Daniel On 10/3/13 6:58 PM, Charles Chance wrote: Hi Daniel, Please see attached - if there is a better way of presenting it please let me know (still getting to grips with git!). There are some improvements planned such as fetching values from other nodes at startup and some kind of buffer/queue to ensure updates are always applied in order, even if some messages are delayed slightly. But it works well as a starting point, and of course the functionality is optional. Please let me have your feedback and I can modify/improve as required. Cheers, Charles On 30 September 2013 10:46, Daniel-Constantin Mierla &lt;miconda(a)gmail.com&gt;wrote;wrote: www.sipcentric.com Follow us on twitter @sipcentric <http://twitter.com/sipcentric> Sipcentric Ltd. Company registered in England & Wales no. 7365592. Registered office: Unit 10 iBIC, Birmingham Science Park, Holt Court South, Birmingham B7 4EJ. -- Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda Kamailio Advanced Trainings - Berlin, Nov 25-28; Miami, Nov 18-20, 2013 - more details about Kamailio trainings at http://www.asipto.com -