So the CantusDB server is down...

This page lists common troubleshooting steps when the Cantus Database server is down.

How to find out the server is down

Obviously, one way you'll discover that the server is down is that someone will alert you that they tried to access Cantus DB but couldn't because of some error. Hopefully, though, you'll receive a notification that the server is down before it gets to this point. How? Through a service called Uptime Robot, we ping www.cantusdatabase.org every 5 minutes and get an email alert if the site is unreachable. If you want to get this notification, talk to whoever manages the lab gmail account about setting up forwarding of these notifications to you.

Check the logs

First, check the logs on the server to see if you can see any obvious errors:

1. ssh into the server and head to the Docker Compose application home directory. On current servers, this is at /home/cantusdb/code/CantusDB/.
2. Run docker logs [name of container] to see the logs for a particular container. The -f flag on this command will follow logs. The --since and --until parameters can be useful to segment the logs (see the Docker documentation for details).

Restart the Compose application

A good start to getting the application working again is to restart the Docker Compose application. Simply run docker compose down; docker compose up -d in the terminal to do so.

Restart the server

If restarting the application doesn't work or if you cannot ssh into the server, you can also restart the VM on which the application is running:

1. Log in to the Arbutus platform management app.
2. Head to the Compute > Instances panel.
3. Find the server you are troubleshooting (either ct-prd-cantus-database or ct-stage-cantus-database).
4. In the "Actions" column, choose "Soft Reboot Instance". Depending on continued issues, you can try "Hard Reboot Instance".

Remember that the application will need to be restarted if you have restarted the VM.