On 9 August 2016 at 22:18, Richard Smith <pob...@gmail.com> wrote: > Apologies for top posting all... Mobile device responses and all... Excuse > perceived brevity et al > > My comments and objections, to the general consensus that Asterisk is the > way forward, are based on several years of working with it. > > As I've mentioned before, I generally try not to pass comment on specific > products because a number of my customers use a variety of products to > fulfil a business need/function. To blithely pass judgement on things > without consideration of individual business cases would just be > unprofessional. > > That said... My objections to asterisk are based on direct experience that > using it as the core function of a service is nigh-on impossible to build as > a stable and generally functional high availability platform.
I think this is OT now for the OP as Paul was asking for their own company use, not to use as a base for an ITSP or VoIP Service Offering within an ISP? > Asterisk lacked the capability (this may have changed and if so, someone, > please direct me to the relevant documentation that supports your > assertions) to function as a high availability and performant (IMO > performant = >5k registrations with subscribe and notifies etc + >>50calls/sec) cluster capable of providing type 2/type 4 services which > complies with the mandates for OFCOM General Condition 4 (and subsequently > section 102 and 105 (I think) of the telecommunications act 2003). It's core function, although it is described on asterisk.org as "An open source telephony switching and private branch exchange service for Linux." is mainly as a b2bua and media translation platform. As you know it can be used for many, many other things. Asterisk 13 is now much better due to the option of not use chan_sip and opting for chan_pjsip. pjsip is a lovely OSS project. > I agree, asterisk has a place... It is not however at the core of a > multi-tenant platform marketed as a cloud (urg) replacement for POTS > services. You just wouldn't design a solution with Asterisk doing everything anyway. Whether it's Asterisk, FreeSWITCH, Yate or the many others, they all have their place in the stack and all known limitations can be tested, documented and worked round with SIP Proxies as registrars, load balancers, presence platforms and all the other things SIP Proxies should be used for. It just depends on what scale you are designing for or moving to/from. Again, http://mcfunley.com/choose-boring-technology It's an amazing platform as is FreeSWITCH, OpenSIPS, Kamailio and all the other Open Source and Free Software projects out there that form part of an ITSP/ISP and are mature and have active development. We live in lucky times to have so much choice. Anyway, this is probably better placed on UKNOT list now. -- Kind Regards, Gavin Henry. Managing Director. Winner of the Best Business ITSP (Medium Enterprise) 2016! http://www.surevoip.co.uk/2016-best-provider
