Troubleshooting

The most common problem new users run into is a misconfiguration that prevents the plugin from initializing. When this happens you will likely see a message like this when you try to install or enable the plugin:

Error response from daemon: dial unix /run/docker/plugins/<id>/netapp.sock: connect: no such file or directory

This simply means that the plugin failed to start. Luckily, the plugin has been built with a comprehensive logging capability that should help you diagnose most of the issues you are likely to come across.

The method you use to access or tune those logs varies based on how you are running the plugin.

Managed plugin method

If you are running Trident using the recommended managed plugin method (i.e., using docker plugin commands), the logs are passed through Docker itself, so they will be interleaved with Docker’s own logging.

To view them, simply run:

# docker plugin ls
ID                  NAME                DESCRIPTION                          ENABLED
4fb97d2b956b        netapp:latest       nDVP - NetApp Docker Volume Plugin   false

# journalctl -u docker | grep 4fb97d2b956b

The standard logging level should allow you to diagnose most issues. If you find that’s not enough, you can enable debug logging:

# install the plugin with debug logging enabled
docker plugin install netapp/trident-plugin:<version> --alias <alias> debug=true

# or, enable debug logging when the plugin is already installed
docker plugin disable <plugin>
docker plugin set <plugin> debug=true
docker plugin enable <plugin>

Binary method

If you are not running as a managed plugin, you are running the binary itself on the host. The logs are available in the host’s /var/log/netappdvp directory. If you need to enable debug logging, specify -debug when you run the plugin.