npme-uninstall.md

Uninstalling npm Enterprise

While installing npme handles the installation of dependencies, uninstalling npme does not uninstall its dependencies. Here's a quick guide to uninstall and remove the different components that make up the Enterprise product.

1. Uninstall the npme package

   sudo npm uninstall -g npme
   

2. Note storage paths in admin console settings

   If you wish to remove all packages and data from your registry, please make a note of all the Storage paths you have configured for your instance at https://<your-server>:8800/settings.

   The default root is /usr/local/lib/npme. You can remove these in step 6 below.

3. Uninstall the orchestration layer

   To uninstall and remove Replicated, please follow the instructions found here.

   Ubuntu/Debian:

   sudo service replicated stop
   sudo service replicated-ui stop 
   sudo service replicated-operator stop
   sudo docker rm -f replicated replicated-ui replicated-operator
   sudo docker images | grep "quay\.io/replicated" | awk '{print $3}' | xargs sudo docker rmi -f
   sudo apt-get remove -y replicated replicated-ui replicated-operator
   sudo apt-get purge -y replicated replicated-ui replicated-operator
   sudo rm -rf /var/lib/replicated* /etc/replicated* /etc/init/replicated* /etc/init.d/replicated* /etc/default/replicated* /var/log/upstart/replicated* /etc/systemd/system/replicated*
   

   CentOS/RHEL:

   sudo systemctl stop replicated replicated-ui replicated-operator
   sudo service replicated stop
   sudo service replicated-ui stop 
   sudo service replicated-operator stop
   sudo docker rm -f replicated replicated-ui replicated-operator
   sudo docker images | grep "quay\.io/replicated" | awk '{print $3}' | xargs sudo docker rmi -f
   sudo yum remove -y replicated replicated-ui replicated-operator
   sudo rm -rf /var/lib/replicated* /etc/replicated* /etc/init/replicated* /etc/default/replicated* /etc/systemd/system/replicated*
   

4. Stop and remove all Docker containers and images

   Assuming you are not using Docker for anything else on your server, you may remove all running Docker containers and images with the following commands:

   # stop all containers
   sudo docker stop $(sudo docker ps -aq)
   # remove all containers
   sudo docker rm $(sudo docker ps -aq)
   # remove all images
   sudo docker rmi -f $(sudo docker images -q)
   

5. Optionally uninstall Docker

   If you wish, you may also remove Docker itself. Use the instructions for your Linux distribution:

   Ubuntu/Debian instructions, sample:

   sudo apt-get autoremove --purge docker-engine
   sudo rm -rf /var/lib/docker
   

   CentOS/RHEL instructions, sample:

   sudo yum -y remove docker-engine
   sudo rm -rf /var/lib/docker
   

6. Optionally remove packages and Enterprise data

   To completely wipe the Enterprise data, you should also remove the directories noted from step 2. Note that this will remove all packages you have published thus far!

   # caution! make sure the directory below is correct
   # caution! this will remove all published packages
   sudo rm -r /usr/local/lib/npme
   

7. Optionally uninstall Node.js and npm

   Assuming you do not need Node.js or npm for anything else on your server, you may uninstall them via the system package manager you used to install Node.js.

   Ubuntu/Debian:

   sudo apt-get autoremove --purge nodejs
   

   CentOS/RHEL:

   sudo yum -y remove nodejs
   

This should remove all traces of npm Enterprise from your server.

Reinstalling npm Enterprise

Follow the same steps as outlined in the quickstart guide.

1. Reinstall Node.js and npm

   Ubuntu/Debian:

   sudo apt-get install -y nodejs
   sudo npm i -g npm@latest
   

   CentOS/RHEL:

   sudo yum -y install nodejs
   sudo npm i -g npm@latest
   

2. Install npme

   sudo npm install npme -g --unsafe
   

   Answer any prompts and wait for the installer to complete.

   Once it's done, verify docker and replicated daemons are running:

   sudo ps aux | grep -e docker -e replicated
   sudo docker ps -a
   replicated nodes
   

3. Complete installation via admin web console

   Use the admin web console at https://<your-server>:8800 to complete the installation and configure your instance.

   See standard install instructions.

4. Wait for the services to start

   

5. Check for error on startup

   On reinstallation, it's possible that the node used by Replicated may not be provisioned correctly, in which case you will receive an error on initial start of the appliance.

   

   To fix this, check the "Connected" and "Initialized" status of the single node under the Cluster page at https://<your-server>:8800/cluster.

   

   If they do not show as green/checked, then you will need to manually reset the node via the following CLI commands:

   # login with same credentials of admin web console
   sudo replicated login
   # get the app node name
   sudo replicated nodes
   # remove the host
   sudo replicated node rm <node-id>
   

   Next, visit the Cluster page again in the UI and click "Add Node".

   

   Select "Install Manually".

   

   Copy the displayed command:

   

   Run it in your SSH terminal. Note that the private_ip variable argument can be omitted.

   curl -sSL https://get.replicated.com/operator | sudo sh -s daemon-endpoint=<ip_from_ui>:9879 daemon-token=<token_from_ui> private-address=<ip_from_ui> public-address=<ip_from_ui>
   

   Then check the state of the node in the UI. It should now be connected and initialized.

   

   Start the registry services by going to the Dashboard and hitting the "Start Now" button.

   

   Once the services are up, verify a healthy node on the Cluster page.

   

   You are good to go! You can now configure your CLI client and start using your Enterprise registry!