This message was deleted.

hmmm that error means it’s not able to reach any Egress server, which probably means a configuration error either with livekit-cli or egress. This kind of error is basically impossible to help with without knowing everything about your setup. How are you running the egress service? What’s the command/config you use with livekit-cli?

Thanks @acoustic-engineer-41666 for your reply. Here is my used setup so far • Egress config.yaml in the mounted volume. I did all steps for redis config file as it’s mentioned on the docs.

log_level: debug
api_key: devkey
api_secret: secret
ws_url: <ws://192.168.65.2:7880>
insecure: true
redis:
  address: 192.168.65.2:6379

• Command that I am using to run Egress service in a docker container

docker run --rm \
    --cap-add=SYS_ADMIN \
    -e EGRESS_CONFIG_FILE=/out/config.yaml \
    -v livekit-egress:/out \
    livekit/egress

• No custom config for livekit-cli. Here is the command that I am using to expose Egress service

livekit-cli start-room-composite-egress --api-key devkey --api-secret secret  --request /Users/wakeel/Downloads/room-composite-file.json

• Also the same for livekit-server, I just ran it like docs without any additional config

brew install livekit

livekit-server --dev

What I can confirm is that livekit-cli can communicate with livekit-server without any problem. As I can see the request is received from livekit-cli on the server. Server can’t communicate with Egress service at all. I am not seeing even failing logs. Here are Egress service logs.

+ rm -rf '/home/egress/tmp/*'
+ rm -rf /var/run/pulse /var/lib/pulse /home/egress/.config/pulse /home/egress/.cache/xdgr/pulse
+ pulseaudio -D --verbose --exit-idle-time=-1 --disallow-exit
I: [pulseaudio] main.c: Daemon startup successful.
+ exec egress
2023-11-24T02:08:47.997Z	INFO	egress	redis/redis.go:127	connecting to redis	{"nodeID": "NE_e9JmJrrGt5C2", "clusterID": "", "simple": true, "addr": "172.17.0.2:6379"}
2023-11-24T02:08:48.006Z	INFO	egress	stats/monitor.go:130	cpu available: 4.000000 max cost: 4.000000	{"nodeID": "NE_e9JmJrrGt5C2", "clusterID": ""}
2023-11-24T02:08:48.010Z	DEBUG	egress	service/debug.go:38	debug handler disabled	{"nodeID": "NE_e9JmJrrGt5C2", "clusterID": ""}
2023-11-24T02:08:48.011Z	DEBUG	egress	service/templates_server.go:38	starting template server on address localhost:7980	{"nodeID": "NE_e9JmJrrGt5C2", "clusterID": ""}
2023-11-24T02:08:48.011Z	DEBUG	egress	service/service.go:94	starting service	{"nodeID": "NE_e9JmJrrGt5C2", "clusterID": "", "version": "1.8.0"}
2023-11-24T02:08:48.016Z	INFO	egress	service/service.go:100	service ready	{"nodeID": "NE_e9JmJrrGt5C2", "clusterID": ""}

just checking, you ran

docker run -it --rm alpine nslookup host.docker.internal

and it returned

192.168.65.2

?

Yes, that’s what I am getting

Server:		192.168.65.5
Address:	192.168.65.5:53

Non-authoritative answer:
Name:	host.docker.internal
Address: 192.168.65.2

Non-authoritative answer:

@bulky-finland-78364 Egress is kind of tricky to setup. I'd recommend using a dedicated VM deploy like outlined here.

Hmmm..Thanks @dry-elephant-14928 I am not sure it’s worth going to VM direction or not as we are planning to use LiveKit cloud anyway. I just wanted a local environment to test out recording as its 1 limit only. So, it’s hard to do more tests with different scenarios with limit 1.

Oh, yeah if you plan to use LiveKit Cloud then I highly recommend against spending time figuring out how to deploy Egress. I’ll point out that if you enter a credit card for Cloud, you still get 50GB bandwidth and 1hr egress for free, but the egress limit is raised. (We do this because keeping egress instances warm is expense, so we want to have some indication of the seriousness of users.)