Might need consider cloud telephony hosting providers reachability, if branch come from different countries; eg the cloud vendors POPs location to be considered as well, best to use cloud vendors POP to POP connections rather than the internet 🙂   However, the best approach will be talk to Cisco and Meraki account team, they will happy to discuss on more 🙂 The account team will be able to reach out to DE or engineering team to verify designed solution;   you can reference cisco CVD for most of design case as well www.cisco.com/go/cvd       ... View more

Meraki MX works with Broadsoft, Broadsoft is hosted telephony system; MX inspection just need passing the voip traffic with access switch and MX QOS enabled as well;   the voip just a applications running over MX WAN networks either autovpn or direct internet would works fine as long as get design right :).   The key is get firewall rule right :).     ... View more

It would be easy, if take the MPLS connection as a circuit, like the normal internet; Leave the routing/ packet forwording /pfr things to Meraki MX :);     ... View more

Hi Babine,   you mean below traffic flow design:   Refer BRKCRS-2142 (ciscolive.com)   However you might need a support case to confirm.       ... View more

Hi Babine,   In that case, and based on what you mentioned in the post;   The issue might be the failover did not work, and the challenge is it is not working even you failback to primary connection; you may need baseline your normal primary connection telephone traffic then comparing it with the failover one.   It could be same issue on below post in 2019, and it is stateful firewall related; it could be already be addressed in the new code, not sure yet; for the time being, you can try 'disable/enable BACKUP interface' 🙂  VoIP problem with stack SIP with the MX WAN connectivity failover - The Meraki Community   My understanding is you may need a support ticket when the failback reconverge did not work.   Happy to help!   Best Regards,       ... View more

You can using more than one SIP trunk if it is possible, depending the PBX you are using; troubleshooting to find out why it is not working even the link failback, the phone or phone system may have log;   As for SBC you can enabled Active-Active I think, there is a call routing control, most of time, the Telco can re-route the incoming call, however, for outgoing call, it need enable Active-Active call control or Active-standby call control.     ... View more

Try these scrips in postman: GitHub - dexterlabora/Meraki-Dashboard-API-Postman-Collection: The complete Meraki Dashboard API in a Postman Collection   ... View more

It seems there is easy way to solve this, and Microsoft has introduced Router Server as a control plane, fyi below: https://docs.microsoft.com/en-us/azure/route-server/overview it is still in public preview stage, if anyone like to try it out, there following is github has more details;   it is routing control plane in Azure which could be used via  BGP to peering with vMX and peering with MS VNET as well.    ... View more

Agreed that there is a misleading on this picture.   Make sense, two MXs in front of the MX concentrator HA pair, then the ebgp can be utilized to Big office internal network or Datacenter internal core, and autovpn ibgp as wan fibric network :).   Let say, the front MX working as firewalls, inspect traffic for DIA and internet underlay path resilience, and this will allow to use full feature sets of the firewalls functions, and the concentrator HA pair designed for terminating the autovpn fibric connections only.   However, if the routed mode MX can keep all feature sets, then converge the front end MX and concentrator HA pair into one pair will save two front end MX or firewalls :).         ... View more

🙂   This is make sense solution, as it is support ospf and static with route mode :);   the ebgp support will make some network migration more smooth as so much feature sets, and easy to scale out; like the idea on talking to support get the bgp enabled for testing only, not production until it is officially supported and CVD verified it.    just some thought, it would be nice if the MX able to converge these Routed, NATing, security and routing protocol ebgp, regardless the fate-sharing drawback;   However, the modular design from  CVD is nice!   anyway, thanks a lot for these info.   Cheers, ... View more