Hey all,
We’re seeing a weird spike in 405 Method Not Allowed errors on our primary trunk, US-Central-02, specifically during the SIP OPTIONS keep-alive attempts. It’s not constant - it’ll be clear for hours, then a burst of 405s for about 10-15 minutes, then back to normal. The dialer isn’t completely choking, but definitely a lot more dropped calls than usual during those bursts. It’s not impacting inbound at all, which is odd.
The carrier is pointing fingers at us, saying the SIP options aren’t being sent correctly. We’ve checked the trunk config in Genesys Cloud several times - the SIP profile looks normal, the outbound SIP headers are the standard stuff. We’ve got the “Allow” header set to OPTIONS, and the codec list is matching what the carrier expects. They’re insistent it’s a malformed OPTIONS request.
I’ve dug into the CDRs and the failing OPTIONS requests all look identical. Here’s a snippet from a recent error log:
1678886400.000000 UTC 1234567890 SIP/2.0 405 Method Not Allowed
Via: SIP/2.0/UDP 192.0.2.1:5060;branch=z9hG4bK776asdhfs
From: "Trunk" <sip:trunk@ourdomain.com>
To: <sip:carrier@theirdomain.com>
Call-ID: abcdef0123456789
CSeq: 1 OPTIONS
Allow: OPTIONS
Contact: <sip:trunk@192.0.2.1:5060>
We’ve got a workaround in place - bumping the SIP keep-alive interval from 60 seconds to 120 seconds seems to reduce the 405s, but it feels like a band-aid. Anyone else hit this? Feels like something’s up with how Genesys is handling the OPTIONS requests under load.