Hello everyone!
I’m fighting with my setup for nextcloud talk and hope someone here can point me in the right direction. My setup in a nutshell:
Nextcloud (20.0.4) and coturn (4.5.1.3) are both hosted on the same server. Server is a dedicated server hosted in a data center with static IP, directly connected to the internet. Configuration of coturn is pretty much according to the manual:
listening-port=3478
listening-ip=SERVER_IP
relay-ip=SERVER_IP
verbose
fingerprint
use-auth-secret
static-auth-secret=theSecret
realm=cloud.example.com
total-quota=0
bps-capacity=0
no-tls
no-dtls
stale-nonce=600
no-stdout-log
log-file=/var/log/turnserver/turn.log
simple-log
no-multicast-peers
What I’m seeing is that I make a call with Nextcloud Talk, the call connects and I have video and audio for about 2-3 minutes. Then the call drops and tries to reconnect fail.
This is what I’m seeing in the coturn logs when it happens:
170: session 007000000000000002: usage: realm=<cloud-example-com>, username=foo_bar, rp=417, rb=352832, sp=1631, sb=1425612
170: session 007000000000000002: peer usage: realm=<cloud-example-com>, username=foo_bar, rp=1631, rb=1298659, sp=417, sb=333841
172: session 002000000000000003: refreshed, realm=<cloud-example-com>, username=foo_bar, lifetime=0
172: session 002000000000000003: realm <cloud-example-com> user foo_bar: incoming packet REFRESH processed, success
172: session 007000000000000002: refreshed, realm=<cloud-example-com>, username=foo_bar, lifetime=0
172: session 007000000000000002: realm <cloud-example-com> user foo_bar: incoming packet REFRESH processed, success
172: session 014000000000000001: refreshed, realm=<cloud-example-com>, username=foo_bar, lifetime=0
172: session 014000000000000001: realm <cloud-example-com> user foo_bar: incoming packet REFRESH processed, success
172: session 014000000000000001: TCP socket closed remotely PUBLIC_IP_1:39715
172: session 014000000000000001: usage: realm=<cloud-example-com>, username=foo_bar, rp=55, rb=7604, sp=6, sb=584
172: session 014000000000000001: peer usage: realm=<cloud-example-com>, username=foo_bar, rp=0, rb=0, sp=49, sb=4704
172: session 013000000000000001: refreshed, realm=<cloud-example-com>, username=foo_bar, lifetime=0
172: session 014000000000000001: closed (2nd stage), user foo_bar realm <cloud-example-com> origin <>, local SERVER_IP:3478, remote PUBLIC_IP_1:39715, reason: TCP connection closed by client (callback)
172: session 013000000000000001: realm <cloud-example-com> user foo_bar: incoming packet REFRESH processed, success
172: session 014000000000000001: delete: realm=<cloud-example-com>, username=foo_bar
172: session 014000000000000001: peer PUBLIC_IP_2 deleted
172: session 014000000000000001: peer SERVER_IP deleted
172: session 014000000000000001: peer PUBLIC_IP_3 deleted
172: session 013000000000000001: TCP socket closed remotely PUBLIC_IP_1:35829
172: session 013000000000000001: usage: realm=<cloud-example-com>, username=foo_bar, rp=55, rb=7604, sp=6, sb=584
172: session 013000000000000001: peer usage: realm=<cloud-example-com>, username=foo_bar, rp=0, rb=0, sp=49, sb=4704
172: session 013000000000000001: closed (2nd stage), user foo_bar realm <cloud-example-com> origin <>, local SERVER_IP:3478, remote PUBLIC_IP_1:35829, reason: TCP connection closed by client (callback)
172: session 013000000000000001: delete: realm=<cloud-example-com>, username=foo_bar
172: session 013000000000000001: peer PUBLIC_IP_2 deleted
172: session 013000000000000001: peer SERVER_IP deleted
172: session 013000000000000001: peer PUBLIC_IP_3 deleted
173: session 002000000000000003: usage: realm=<cloud-example-com>, username=foo_bar, rp=55, rb=7604, sp=6, sb=584
173: session 002000000000000003: peer usage: realm=<cloud-example-com>, username=foo_bar, rp=0, rb=0, sp=49, sb=4704
173: session 002000000000000003: closed (2nd stage), user foo_bar realm <cloud-example-com> origin <>, local SERVER_IP:3478, remote PUBLIC_IP_1:40572, reason: allocation timeout
173: session 002000000000000003: delete: realm=<cloud-example-com>, username=foo_bar
173: session 002000000000000003: peer PUBLIC_IP_2 deleted
173: session 002000000000000003: peer SERVER_IP deleted
173: session 002000000000000003: peer PUBLIC_IP_3 deleted
173: session 007000000000000002: usage: realm=<cloud-example-com>, username=foo_bar, rp=177, rb=105736, sp=1641, sb=1372756
173: session 007000000000000002: peer usage: realm=<cloud-example-com>, username=foo_bar, rp=1640, rb=1245034, sp=176, sb=97594
173: session 007000000000000002: closed (2nd stage), user foo_bar realm <cloud-example-com> origin <>, local SERVER_IP:3478, remote PUBLIC_IP_1:40181, reason: allocation timeout
173: session 007000000000000002: delete: realm=<cloud-example-com>, username=foo_bar
173: session 007000000000000002: peer PUBLIC_IP_2 deleted
173: session 007000000000000002: peer SERVER_IP deleted
173: session 007000000000000002: peer PUBLIC_IP_3 deleted
Reconnection attempts pretty much follow the same picture, I see
ALLOCATE processed, success
CREATE_PERMISSION processed, success
REFRESH processed, success
TCP socket closed remotely
I don’t really know what to look for here 
I dropped my firewall rules and tried without any to see if it’s a firewall issue, but the issue remained so I doubt is has anything to do with my firewall settings.
Any pointers here would be much appreciated.