One of the pbx servers that I help manage is having issues regarding calls being dropped due to “All circuits are busy” message.
Looking at the logs, I realize that the connection to the SIP peer is going up and down randomly. It seems like they are only getting the “busy” message, when the circuit shows down, which makes sense. There is no pattern on when this connection drops, it is very random.
What does not make sense is that I was on a call from softphone to cellphone, letting it sit to see if the call dropped. I was in the call maybe 8 minutes and I saw the connection to the peer go down in the logs. According to the logs, the connection to the peer was lost for about a minute, but my session did not drop. I did not lose audio either way.
I have not been able to re-create this issues testing with a softphone calling to and from my personal cell phone.
I have opened a ticket with the SIP provider as well as the internet circuit provider this is going through and neither of them reported any issues.
The ‘connection’ didn’t necessarily go down but the SIP ‘pings’ failed for some reason.
The SDP media session remained up (the audio remained) but that might well be coming from a different server.
Who is your carrier? and how is your connection to your provider configured?
Lumen is our carrier. Here are screenshots of trunk config
use sngrep to watch for any ‘failures/timeouts’ filter on the VSP’s name/IP
When I am looking for these, will it say failure or timeout here, instead of Notify or Options
I was looking through sngrep and saw this
The SIP provider got back to me and said the 405 not allowed is because we are sending options and that is not supported on our current trunk config so I don’t think that is the problem. I looked at one of my working PBXs and saw the same 405 not allowed come through in the logs so that is not the problem.
I am seeing some rejected invites in the sngrep output. These look to be expected though as the call is going to an extension that is not currently registered.
We set the qualify frequency to 0 on the trunk so essentially no qualify now. Will we still see these messages saying the sip peer is reachable/unreachable with qualify taken off the trunk?
[2022-01-14 19:14:50] VERBOSE res_pjsip/pjsip_options.c: Contact L3_pjsip/sip:x.x.x.x:5060 is now Unreachable. RTT: 0.000 msec
[2022-01-14 19:17:47] VERBOSE res_pjsip/pjsip_options.c: Contact L3_pjsip/sip:x.x.x.x:5060 is now Reachable. RTT: 38.024 msec
[2022-01-14 19:25:06] VERBOSE res_pjsip/pjsip_options.c: Contact L3_pjsip/sip:x.x.x.x:5060 is now NonQualified.
I finally was able to replicate the issue. I saw the peer connection go down and attempted to make a call out from my softphone and got the “All circuits are busy now” message.
I believe setting the trunk to no qualify has fixed this issue. I have not had any reports of not being able to make calls since this change has been made.
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.