Skip to main content

BYOLS self-hosting troubleshooting

This troubleshooting guide is primarily aimed at self-hosting customers who have installed Tines on their own Linux server via the zip file that we distribute.

1. Run the Tines troubleshooting helpers

The Tines zip file ships with some helper scripts to make it easy to quickly grab a range of troubleshooting information for sharing with Tines support:

cd /opt/tines
source commands.sh
tines-tenant-info
tines-troubleshoot

2. Check that the Docker daemon and the Tines containers are running

The output of sudo docker ps should look very similar to this:

# sudo docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
52eda5382421 tines/tines-nginx:latest "/entrypoint.sh ngin…" 3 days ago Up 3 days 0.0.0.0:80->80/tcp, 0.0.0.0:443->443/tcp nginx
e58272bbcc47 tines/tines-app:latest "start-tines-sidekiq" 3 days ago Up 3 days tines-sidekiq
03db7528d913 tines/tines-app:latest "start-tines-app" 3 days ago Up 3 days tines_tines-app_1
0d98aeca3433 redis:6.0.9-alpine "docker-entrypoint.s…" 3 months ago Up 3 days 0.0.0.0:6379->6379/tcp redis
a499a6f65147 postgres:11.9-alpine "docker-entrypoint.s…" 3 months ago Up 3 days 0.0.0.0:5432->5432/tcp postgres

Docker daemon not running

When the output of sudo docker ps looks like this:

Cannot connect to the Docker daemon at unix:///var/run/docker.sock. Is the docker daemon running?

... it means that the Docker daemon itself isn't running and will need to be started before the Tines containers can be started.

Depending on your system, the following might work for starting it:

sudo systemctl start docker

Tines containers not running

When the output of sudo docker ps looks like this:

# sudo docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES

... it means the Docker daemon is running but the containers are not. The containers can be started via:

cd /opt/tines
sudo docker-compose up -d

Tines containers continuously restarting

If successive runs of sudo docker ps show you output similar to the following:

# sudo docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
66f42f399746 tines/tines-nginx:latest "/entrypoint.sh ngin…" 24 seconds ago Up 20 seconds 0.0.0.0:80->80/tcp, 0.0.0.0:443->443/tcp nginx
4defb65ad456 tines/tines-app:latest "start-tines-sidekiq" 24 seconds ago Up Less than a second tines-sidekiq
d10fd54693bc tines/tines-app:latest "start-tines-app" 26 seconds ago Restarting (2) 6 seconds ago tines_tines-app_1
abe1fd42895c redis:6.0.9-alpine "docker-entrypoint.s…" 28 seconds ago Up 26 seconds 0.0.0.0:6379->6379/tcp redis
f7352cd0f05b postgres:11.9-alpine "docker-entrypoint.s…" 28 seconds ago Up 25 seconds 0.0.0.0:5432->5432/tcp postgres
# sudo docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
66f42f399746 tines/tines-nginx:latest "/entrypoint.sh ngin…" 28 seconds ago Up 25 seconds 0.0.0.0:80->80/tcp, 0.0.0.0:443->443/tcp nginx
4defb65ad456 tines/tines-app:latest "start-tines-sidekiq" 28 seconds ago Restarting (2) 4 seconds ago tines-sidekiq
d10fd54693bc tines/tines-app:latest "start-tines-app" 30 seconds ago Restarting (2) 3 seconds ago tines_tines-app_1
abe1fd42895c redis:6.0.9-alpine "docker-entrypoint.s…" 32 seconds ago Up 30 seconds 0.0.0.0:6379->6379/tcp redis
f7352cd0f05b postgres:11.9-alpine "docker-entrypoint.s…" 32 seconds ago Up 29 seconds 0.0.0.0:5432->5432/tcp postgres

...with repeated mentions of containers "restarting", it means that something is misconfigured and the Tines application is unable to run correctly. In this situation Docker will continue to attempt to restart the containers until they start to work.

When this happens, you'll need to capture the logs of the containers that are restarting and share them with Tines support. The command for doing this is sudo docker logs <container-name>. The "container name" is in the last column of the output above, so for that example you would run:

sudo docker logs tines-sidekiq
sudo docker logs tines_tines-app_1

The Tines zip file also ships with a helper script to make it easy to quickly grab all of the containers' logs for sharing with Tines support:

cd /opt/tines
source commands.sh
tines-logs

3. Check that the web UI is loading

Visit the domain you have configured to point to the server running tines in your browser e.g. https://tines.my-company.com

nginx container failing to talk to tines-app container

If you continue to see the following message for more than a minute:

nginx error screen

...it means that the nginx container is failing to talk to the tines-app container. You'll need to capture the logs of the nginx container to share with Tines support:

sudo docker logs nginx

or

cd /opt/tines
source commands.sh
tines-logs

On rare occasions, restarting all containers can resolve a transient issue and restore access:

cd /opt/tines
sudo docker-compose down
sudo docker-compose up -d

Tines application producing an error

If you repeatedly see the following message after taking the same sequence of steps in the application:

Tines error screen

...then capture and share those steps with Tines support. If you can, also capture the full set of logs immediately after it happens for sharing with Tines support:

cd /opt/tines
source commands.sh
tines-logs