OpenViduPro deployment

OpenVidu v2 Issues with deployment
1

Hello ,
I have problems with the deployment of the application pro in local machine, (we have a CE version which is in production), so I want to do some tests before replacing the CE version with the PRO one

NGINX LOGS :
nginx_1 | =======================================
nginx_1 | = START OPENVIDU PROXY =
nginx_1 | =======================================
nginx_1 |
nginx_1 | 2021/01/26 12:08:43 [warn] 67#67: “ssl_stapling” ignored, no OCSP responder URL in the certificate “/etc/letsencrypt/live/127.0.0.1/fullchain.pem”
nginx_1 | nginx: [warn] “ssl_stapling” ignored, no OCSP responder URL in the certificate “/etc/letsencrypt/live/127.0.0.1/fullchain.pem”
nginx_1 | 2021/01/26 12:08:43 [notice] 67#67: signal process started
nginx_1 | 2021/01/26 12:08:43 [warn] 15#15: “ssl_stapling” ignored, no OCSP responder URL in the certificate “/etc/letsencrypt/live/127.0.0.1/fullchain.pem”
nginx_1 | 2021/01/26 12:20:20 [error] 68#68: *74 connect() failed (111: Connection refused) while connecting to upstream, client: 127.0.0.1, server: 127.0.0.1, request: “GET /kibana HTTP/1.1”, upstream: “http ://[::1]:5601/”, host: “127.0.0.1”
nginx_1 | 2021/01/26 12:20:20 [warn] 68#68: *74 upstream server temporarily disabled while connecting to upstream, client: 127.0.0.1, server: 127.0.0.1, request: “GET /kibana HTTP/1.1”, upstream: “http ://[::1]:5601/”, host: “127.0.0.1”
nginx_1 | 2021/01/26 12:20:30 [error] 68#68: *74 connect() failed (111: Connection refused) while connecting to upstream, client: 127.0.0.1, server: 127.0.0.1, request: “GET /inspector HTTP/1.1”, upstream: “http ://[::1]:5443/inspector”, host: “127.0.0.1”
nginx_1 | 2021/01/26 12:20:30 [warn] 68#68: *74 upstream server temporarily disabled while connecting to upstream, client: 127.0.0.1, server: 127.0.0.1, request: “GET /inspector HTTP/1.1”, upstream: “http ://[::1]:5443/inspector”, host: “127.0.0.1”
nginx_1 | 2021/01/26 12:20:30 [error] 68#68: *74 connect() failed (111: Connection refused) while connecting to upstream, client: 127.0.0.1, server: 127.0.0.1, request: “GET /inspector HTTP/1.1”, upstream: “http ://127.0.0.1:5443/inspector”, host: “127.0.0.1”
nginx_1 | 2021/01/26 12:20:30 [warn] 68#68: *74 upstream server temporarily disabled while connecting to upstream, client: 127.0.0.1, server: 127.0.0.1, request: “GET /inspector HTTP/1.1”, upstream: “http ://127.0.0.1:5443/inspector”, host: “127.0.0.1”

OPENVIDU LOGS :

openvidu-server_1 | [INFO] 2021-01-26 13:57:16,045 [main] io.openvidu.server.config.OpenviduConfig - Configuration properties read from file /opt/openvidu/.env
openvidu-server_1 | [INFO] 2021-01-26 13:57:16,307 [main] io.openvidu.server.pro.OpenViduServerPro - OpenVidu Pro is deployed in ‘on_premise’ environment
openvidu-server_1 | [WARN] 2021-01-26 13:57:16,309 [main] io.openvidu.server.pro.OpenViduServerPro - OpenVidu Pro cluster mode is ‘manual’. There will be no automatic instances management
openvidu-server_1 | [INFO] 2021-01-26 13:57:16,457 [main] io.openvidu.server.pro.OpenViduServerPro - OpenVidu Pro cluster mode enabled
openvidu-server_1 | [INFO] 2021-01-26 13:57:16,473 [main] io.openvidu.server.pro.OpenViduServerPro - Cluster identifier got from disk: clu_HHEOUiOa
openvidu-server_1 | [INFO] 2021-01-26 13:57:16,508 [main] io.openvidu.server.pro.OpenViduServerPro - OpenVidu Pro Elasticsearch service is enabled
openvidu-server_1 | [ERROR] 2021-01-26 13:57:19,804 [main] io.openvidu.server.pro.cdr.CDRLoggerElasticSearch - Connection to Elasticsearch failed at http ://127.0.0.1:9200 (Connection is closed)
openvidu-server_1 | [ERROR] 2021-01-26 13:57:19,808 [main] io.openvidu.server.pro.cdr.CDRLoggerElasticSearch - If property ‘OPENVIDU_PRO_ELASTICSEARCH_HOST’ is defined, then it is mandatory that OpenVidu Server Pro is able to connect to it
openvidu-server_1 | [ERROR] 2021-01-26 13:57:19,809 [main] io.openvidu.server.pro.cdr.CDRLoggerElasticSearch - Terminating OpenVidu Server Pro

2

Hello:

The error log is telling you what is missing:

Connection to Elasticsearch failed at http://127.0.0.1:9200 (Connection is closed)
If property ‘OPENVIDU_PRO_ELASTICSEARCH_HOST’ is defined, then it is mandatory that OpenVidu Server Pro is able to connect to it
Terminating OpenVidu Server Pro

There is no Elasticsearch service available at http://127.0.0.1:9200. To test in local environment without OpenVidu Pro Elastic stack integration, just disable property OPENVIDU_PRO_ELASTICSEARCH_HOST. See OpenVidu Docs

image
image921×440 75.9 KB

If you want the Elasticsearch service in your local deployment, then you must have it up and running. The official docker-compose deployment of OpenVidu Pro already includes it: openvidu/docker-compose.yml at master · OpenVidu/openvidu · GitHub.

3

thanks for ur response a made both :
OPENVIDU_PRO_KIBANA_HOST=
OPENVIDU_PRO_ELASTICSEARCH_HOST=

and i still have :

openvidu-server_1 | [WARN] 2021-01-26 15:56:34,102 [main] io.openvidu.server.pro.config.KibanaConfig - IOException when reaching Kibana REST API with method GET at path /api/status: null
openvidu-server_1 | [ERROR] 2021-01-26 15:56:34,103 [main] io.openvidu.server.pro.config.KibanaConfig - If property ‘OPENVIDU_PRO_KIBANA_HOST’ is defined, then it is mandatory that OpenVidu Server Pro is able to connect to it
openvidu-server_1 | [ERROR] 2021-01-26 15:56:34,103 [main] io.openvidu.server.pro.config.KibanaConfig - Terminating OpenVidu Server Pro

4

I don’t think OPENVIDU_PRO_KIBANA_HOST is disabled. Please, reivew it again and make sure it is empty.

5

this the .env file
image

6

How did you deployed? Do you actually have a kibana and elasticsearch running?

Please can you show me the output of docker ps and:

sudo su
cd /opt/openvidu
./openvidu report

This will generate a report with all the information of the deployment.

I think you did not follow instructions properly or something is missing.

Regards,
Carlos

8

I made a new vm and installed openvidu
Have I to install kibana and elasticsearch separately ?

docker ps

image
image1854×319 55.8 KB

report :

https://sharetxt.xyz/id?55

10

Just to confirm, are you trying to deploy OpenVidu Pro in your local machine? I see you’re using DOMAIN_OR_PUBLIC_IP=127.0.0.1. We didn’t test such scenario but I don’t think this would work as it is.

  1. The main reason is that a Media Node need to be installed externally, and install OpenVidu and Media Nodes in the same machine makes some problems which needs to be solved manually and is not an easy task.
  2. As it is installed locally this could cause some problems because of internal ports being used.

I can see two possible solutions here:

Option 1 Deploy a dev environment with two instances in your cloud provider or infrastructure, one with OpenVidu and another one with Media Node (to have at least one media node configured). I recommend to make it accesible to the internet with a valid Domain name pointing to it’s public IP.

Option 2 Deploy two VM, install OpenVidu Pro Master Node in one, and a Media Node in the other. I think that if you use the “DOMAIN_OR_PUBLIC_IP” with the private IP of your VM it should work.

Also you need to add in OpenVidu Pro the Media Node with its own private Ip as stated here: OpenVidu Docs

11

thank you so much for your help

12

The problem here is that I didn’t install the Media Node, I just tested the Openvidu server and it doesn’t boot and shows the problems mentioned above