As is usual for issues I post on here about, this one really has me scratching my head.
We were sold PBXact 13 with high availability after it was already being sunsetted - management does not want to lose the redundancy after paying so much for the module. Frankly, I don’t want to lose the HA either, only for technical reasons. (I can reboot our servers in the middle of the day and as long as I’m careful, nobody notices. 
) We’re effectively stuck on 13 as 14+ do not support HA. Our phones are mostly Polycom 400 phones with a few 500 models thrown in. A very few people wanted to keep their Sangoma s405 / s500 / s700 phones, but these equate to 3 out of ~80 phones we have deployed. All are running chan_sip. Our servers are housed on-site behind a Meraki edge router and are on their own VLAN. Things were working exactly as one would expect from an office phone system… until mid-January.
I did nothing significant to our system at the time and our SIP trunk provider says they changed nothing. My users will dial a number and there is no ringing or any indication whatsoever that the call is progressing - you dial a number, there’s silence, and then either the person says hello or their voice mail picks up. It doesn’t hamper functionality, but is confusing for my users. It does not appear to be specific to any particular model phone and began out of the blue.
The PBX recordings apparently do not begin until the call is connected, so I am unable to hear whether there is a ring tone there or not. Our provider ran a packet capture on their gateway and found that we are being sent rings from them for a call this happened on, at least according to the captured RTP streams. Later, a user reported it happened once when she called a direct extension, so I have to conclude it’s something inside the PBX.
Unfortunately, I’m really at a loss - we’ve never had this issue before, nothing has changed at our provider, and I’ve done nothing but reset the name/password on extensions when people come & go for several months. The problem is very intermittent, happening only a few times a day. I’ve asked if this behavior happens on every call to specific numbers (thinking it might be something with the destination’s phone) and am waiting on a response.
I found this post from 2017 about what sounds like the same exact issue, but there are no responses.
Another topic here seems to be similar, though affecting all calls instead of only a handful. It suggested adding prematuremedia=no to sip_general_custom.conf and progressinband=yes to the trunk configuration, but this proved ineffective. There is a suggestion to add r to the trunk configuration, though with drawbacks; I’d rather find and fix the real cause if I can before making such a change. (And I’m not entirely certain that this would solve the issue, given that a user reported the same behavior on an internal call.)
A post on another forum also describes similar behavior. The person there said that “adding bridge_early_media=true followed by ring_ready into the outbound routes dialplan after the destination_number instruction” solved the problem, though with behavior that sounds strikingly similar to what is described by adding ‘r’ to the trunk options. This post is not for FreePBX, so I can understand the ‘r’ being added as a shortcut to enable the same configuration as a convenience to make things easier. A post linked to that one says to set “instant_ringback=true” & “ringback=$${nl-ring}” in the trunk configuration. I have tried neither of these as I do not understand what they do & don’t want to break our PBX while trying to resolve a relatively minor glitch.
I appreciate any help or direction in tracking down & fixing whatever cause this.