I have done some things (years ago) on Amazon with freeswitch that didn’t work out great. Signaling was fine, media streams were a nightmare. I seen some threads on this forum about FreePBX attempts on Amazon that didn’t go well either.
Would definitely like to know more about the outages/disconnects…
We have been working with Cyberlynk at FreePBX for 5 years and have had only a few small outages in 5 years and always related to upstream ISP going down and having to remove them from the BGP routing as they were causing issues.
@tonyclewis I have to say I love Cyberlynk. The issue had had is Lever3’s connection from the southeast connecting to the data center. The other day it took 3-hours for the service to fully come back on line.
I vpn to New York I didn’t have any issues connecting, so I know it’s not Cyberlynk.
So I need a backup plan. Amazon peering is what I’m looking at. But I can’t get the distro on EC2.
The other day we lost connection from Atlanta to Milwaukie from Comcast and One Ring. Traceroute shows both of them use level3 to get to the data center.
When I come from my NYC location I use AT&T and terminate on Cogent at Cyberlynk bypassing level3. That connection has not failed in the past year that I’ve been connected to it
Can someone please tell me why the routers in my local area didn’t re-route the traffic as soon as the preferred route was terminated. Is that what routes are supposed to do?
So then the edge router at both ends should have rebuilt the route to reconnecting using BGP? Did BGP fail at both ends, not set or missed configured? Does it need to be set correctly on both ends to work? I would expect them to reroute the traffic in seconds.