If SIP trunk is possible, will any of the users via the different product (Asterisk and 3CX ) generate independent report showing if the call was serviced without relying on the other partner report ?
Note, the way SIP TRUNK works, once the call moves from one trunk to another, the originating trunk will not be able to determine the end result of the other calls hence, the report will show you only when calls came in and was transferred to another trunk.
I believe I was directing my comments more to people who had been left without an easy upgrade option from Elastix 2.5. Issabel provides a painless upgrade in such cases. You have a choice between asterisk 11 and 13 at install time if you need or want the few new features introduced.
In my humble opinion the telephone system does not belong on the UTM gateway. All core features should be highly available with failover in seconds (via carp or similar).
Those same users would not likely need documentation as the organization and features remains about the same. Wazo offers some very interesting features for those more interested in cutting edge telephony.
Just my two cents as someone who has been heavily involved with VOIP since 2004.
I believe that ship has sailed. Anyone still sitting on Elastix or similar will be forced to migrate by recent events.
Of course but not in a highly available manner. Haast could be used here but that still leaves database replication to solve at least in the non commercial version. Gulera extensions could be an answer to that issue but cluster recovery would have to be added to Nethserver.
Wazo is highly available out of the box but offers a ālimpā mode rather than handling database replication. I plan on studying it over the next few months with an eye to replace my current voip setup.
I am involved with managing several Perconna clusters (gulera) and I like them. The limitations probably wonāt affect a pbx but I havenāt used the experimental MyISAM replication. The database engine used by Freepbx is MyISAM but that decision is probably left over from the days when Mysql didnāt support full text indexing in innodb.
ETA- changed search to indexing (still on my first cup of coffee)
HAAst supports PostgreSQL (which is used by Wazo) - so you can build a full cluster using HAAst+Wazo. Of course HAAst supports Issabel too, which is why I recommend. We migrated our Elastix PBXās to Issabel with HAAst to create HA clusters.
Most people seem to think mirror the DB is the biggest issue, itās not! Properly detecting various failures, proper recovery & fall back, etc. are just as important. We started with corosync + kamailio and tried different heartbeats etc. In the end is the DIY solution didnāt work 100% right, 100% of the timeā¦not worth the amount of effort it was taking too. So we implemented HAAst (see www.telium.ca).