Docker-compose apache guacamole gets stuck

Question

Im using docker-compose version 1.25.0 with nginx container and guacamole + guacd container version 1.2.0,

Enviroment works fine until once in 2-3 days guacamole + guacd containers gets stuck with this logs :

guacamole :

guacamole_1       | 14:48:48.651 [http-nio-8080-exec-68] INFO  o.a.g.tunnel.TunnelRequestService - User "guacadmin" disconnected from connection "21875". Duration: 10349153 milliseconds
guacamole_1       | 14:48:59.902 [http-nio-8080-exec-77] INFO  o.a.g.environment.LocalEnvironment - GUACAMOLE_HOME is "/root/.guacamole".
guacamole_1       | 14:49:11.606 [http-nio-8080-exec-81] INFO  o.a.g.environment.LocalEnvironment - GUACAMOLE_HOME is "/root/.guacamole".
guacamole_1       | 14:49:23.103 [http-nio-8080-exec-87] INFO  o.a.g.environment.LocalEnvironment - GUACAMOLE_HOME is "/root/.guacamole".
guacamole_1       | 14:49:40.457 [http-nio-8080-exec-94] INFO  o.a.g.environment.LocalEnvironment - GUACAMOLE_HOME is "/root/.guacamole".
guacamole_1       | 14:49:48.032 [http-nio-8080-exec-102] INFO  o.a.g.environment.LocalEnvironment - GUACAMOLE_HOME is "/root/.guacamole".
guacamole_1       | 14:49:56.853 [http-nio-8080-exec-107] INFO  o.a.g.environment.LocalEnvironment - GUACAMOLE_HOME is "/root/.guacamole".

guacd:

guacd_1           | guacd[15306]: INFO: Accepted format: 16-bit PCM with 2 channels at 44100 Hz
guacd_1           | guacd[15306]: ERROR:        User is not responding.
guacd_1           | guacd[15306]: INFO: User "@48e59b0d-c88b-405c-8a88-c6d5dc75cf4b" disconnected (0 users remain)
guacd_1           | guacd[15306]: INFO: Last user of connection "$730dba5f-ceca-468a-85da-ac7d55fc9a41" disconnected
guacd_1           | guacd[15306]: INFO: Internal RDP client disconnected
guacd_1           | guacd[6]: INFO:     Connection "$730dba5f-ceca-468a-85da-ac7d55fc9a41" removed.
guacd_1           | guacd[15254]: ERROR:        User is not responding.
guacd_1           | guacd[15254]: INFO: User "@1b440518-4540-4935-90d7-b58887fa08d5" disconnected (0 users remain)
guacd_1           | guacd[15254]: INFO: Last user of connection "$7f2b67df-d6b6-486a-9718-0b6dd10b7efd" disconnected
guacd_1           | guacd[15228]: ERROR:        User is not responding.
guacd_1           | guacd[15228]: INFO: User "@8b94a59f-9486-4616-920d-eab33e2bf9be" disconnected (0 users remain)
guacd_1           | guacd[15228]: INFO: Last user of connection "$47eab730-314c-4429-9b81-d16ed2ac458a" disconnected
guacd_1           | guacd[15254]: INFO: Internal RDP client disconnected
guacd_1           | guacd[15228]: INFO: Internal RDP client disconnected
guacd_1           | guacd[6]: INFO:     Connection "$47eab730-314c-4429-9b81-d16ed2ac458a" removed.
guacd_1           | guacd[6]: INFO:     Connection "$7f2b67df-d6b6-486a-9718-0b6dd10b7efd" removed.
guacd_1           | guacd[20341]: ERROR:        Error connecting to RDP server

And my api code gets (<Response [502]>,) to GET request for token from guacamole, containers are stuck and no other movment in the logs.

it all start working again after i docker-compose down everything . but i dont understand what is causing the issue, it all works fine until it dont and only a full restart makes it working again.

How can i know what is causing it ? is there any way to make docker restart when container is stuck ? i am considering updating guacd and guacamole from 1.2.0 to 1.3.0 i just dont get how all of a sudden (after 6 month of everything working fine) it started having this bug.

score 0 · Answer 1 · answered Sep 09 '21 at 22:51

0

Docker sent out a major product and terms of service update sometime around Aug 30. That may be the source of your problem. Rather than fighting with the RDP error stack, try running the following docker-compose.yml file based on the modified guacamole image from oznu. It includes the guacamole web client, the guacd server, and a postgres database. The containers have been running on my machine for about a week now with no problems.

version: "2"
services:
  guacamole:
    image: oznu/guacamole
    container_name: guacamole
    volumes:
      - postgres:/config
    ports:
      - 8080:8080
volumes:
  postgres:
    driver: local

answered Sep 09 '21 at 22:51

jrbe228

356
3
11

Hey, thank you for your answer. Problems starts 2 month ago and i cannot replicate it on another machine because on this machine there is a lot of traffic, i did update docker to 20 docker-compose to 1.29 and guacamole + guacd from 1.2 to 1.3 - so far so good. – Batchen Regev Sep 14 '21 at 13:36
Good to hear! Let me know if anything changes – jrbe228 Sep 14 '21 at 15:39
Well nothing changed, once or twice a day the guacd container gets stuck, i can see in `docker-compose ps` that container is healthy, but the logs stop on container like there is no new logs the weird thing i found is that if i restart container and everything starts working again i can see old logs from when it got stuck - meaning if i see no logs after 5:00 and i restart at 7:00 i see all the logs from 5 to 7, im thinking of converting my environment from docker to Kubernetes, ill update again – Batchen Regev Sep 30 '21 at 12:06

Docker-compose apache guacamole gets stuck

1 Answers1