README

Author: EnigmaCurry

contrib/k8s-lbry/README.md

@@ -40,6 +40,7 @@ cluster.
     + [Startup chainquery with a database snapshot](#startup-chainquery-with-a-database-snapshot)
   * [lbrynet](#lbrynet)
     + [IMPORTANT - Backup your cluster wallet](#important---backup-your-cluster-wallet)
+- [Uninstall](#uninstall)
 
 <!-- tocstop -->
 
@@ -95,7 +96,7 @@ You may also find [an older version of this same tutorial from before run.sh
 existed](https://github.com/EnigmaCurry/lbry-docker/blob/7bf2555e230aee6997c5192a98d716f8504a495a/contrib/k8s-lbry/README.md),
 it has a focus on using helm and kubectl directly. But after reading that, you
 may appreciate that the commands here are shorter, and easier to repeat. If you
-pay close attention to the output of `run.sh`, you will see that it prints out
+pay close attention to the output of [`run.sh`](run.sh), you will see that it prints out
 the full helm or kubectl commands as it runs them.
 
 ### Create a project directory
@@ -119,16 +120,19 @@ chmod a+x run.sh
 
 ### Setup alias and tab completion
 
-`run.sh` can be run directly without any setup. However, setting a bash alias
-for `run.sh` is the quickest way of setting up to run from any directory, as
-well as activating support for bash tab completion.
+`run.sh` can be run directly without any setup. However, without adding it to
+your `PATH`, you need to specify the full path to the script each time. Setting
+a bash alias for `run.sh` is the quickest way of setting up to run from
+anywhere, as well as activating support for bash tab completion.
+
+One time setup to install alias to `$HOME/.bashrc`:
 
 ```
 ./run.sh setup-alias
 ```
 
-It should prompt you if it is OK to edit your `$HOME/.bashrc`. Once you confirm,
-close your terminal session, then reopen it.
+It should prompt you if it is OK for the script to edit `$HOME/.bashrc`. Once
+you confirm, close your terminal session, then reopen it.
 
 Verify the new `k8s-lbry` alias to `run.sh` is working:
 
@@ -140,8 +144,8 @@ Notice that tab completion should work throughout typing the above command.
 
 ### k8s-lbry setup
 
-Setup will check for dependencies, update helm repositories, and creates an
-initial config file.
+Setup will check for dependencies, update helm repositories, and create an
+initial config file (`values-dev.yaml`).
 
 ```
 k8s-lbry setup
@@ -153,7 +157,7 @@ An Ingress Controller
 ([nginx-ingress](https://github.com/helm/charts/tree/master/stable/nginx-ingress))
 will help you to route outside internet traffic into your cluster. nginx-ingress
 will also help terminate TLS connections (SSL) so that your containers don't
-need to worry about encryption.
+need to worry about encryption of traffic.
 
 Install nginx-ingress into the `k8s-lbry` namespace:
 
@@ -182,14 +186,15 @@ depending on what you enable in `values-dev.yaml`.
 Wait for the nginx-ingress Load Balancer to show an External IP address:
 
 ```
-k8s-lbry kubectl get svc
+k8s-lbry kubectl get svc -w
 ```
 
 Press Ctrl-C to quit once you see the External IP listed (and not `<pending>`).
 
-You must create your own `values-dev.yaml` (or have told `run.sh setup` to
-create one for you), or to use a path other than `values-dev.yaml`, pass the
-`VALUES` environment variable to subsequent `run.sh` commands.
+You must edit your own `values-dev.yaml`. (The setup procedure created an
+initial configuration in the same directory as `run.sh`.) To use a different
+config file, set the `VALUES` environment variable before subsequent commands,
+specifying the full path to your values file.
 
 Edit `values-dev.yaml`. Change the following things:
 
@@ -201,7 +206,7 @@ Edit `values-dev.yaml`. Change the following things:
  * Change `echo-http-server.hostname` to any domain name you choose. (It must be
    a real internet domain that you control, if you choose to enable TLS.)
  * Each section has it's own `enabled` flag, to selectively turn on each
-   service. It is recommended to start with only lbrycrd enabled, then enable
+   service. It is recommended to start with only `lbrycrd` enabled, then enable
    other services as you need them by running `upgrade`.
 
 Save `values-dev.yaml`. 
@@ -214,10 +219,10 @@ k8s-lbry install
 
 ### k8s-lbry upgrade
 
-For helm, upgrade does not necessarily mean you are upgrading to a new version
-of any particular software, it just means to apply your configuration file to
-the cluster. If you edit `values-dev.yaml`, you apply your changes with
-`k8s-lbry upgrade`.
+For helm, `upgrade` does not necessarily mean you are upgrading to a new version
+of any particular software, `upgrade` just means to apply your configuration
+file to the cluster. If you edit `values-dev.yaml`, you then need to apply your
+changes with `k8s-lbry upgrade`.
 
 You can make changes to `values-dev.yaml` at any time. You can apply your
 configuration to your cluster by upgrading the release:
@@ -333,7 +338,7 @@ Run the mysql shell for the speech database.
 
 #### k8s-lbry logs [app_name]
 
-Stream the logs for a pod given the helm app name into your terminal.
+Stream the logs for a pod into your terminal, given the helm app name.
 
 Example:
 
@@ -385,8 +390,8 @@ k8s-lbry logs lbrycrd
 
 Press Ctrl-C to stop viewing the log.
 
-It is advisable to wait for lbrycrd to syncrhonize with the full blockchain
-before starting other services, so watch the logs until syncrhonization
+It is advisable to wait for lbrycrd to synchronize with the full blockchain
+before starting other services, so watch the logs until synchronization
 completes.
 
 You can utilize `lbrycrd-cli` as well:
@@ -529,3 +534,42 @@ deposit LBC:
 k8s-lbry lbrynet address unused
 ```
 
+## Uninstall
+
+If you wish to uninstall k8s-lbry from your cluster, here are the steps:
+
+ * Delete the helm releases:
+
+    ```
+     k8s-lbry helm delete k8s-lbry
+
+     k8s-lbry helm delete cert-manager
+
+     k8s-lbry helm delete nginx-ingress
+    ```
+
+   * By deleting the `nginx-ingress` release, the Load Balancer resource should
+     be automatically cleaned up. You can verify this yourself in your cloud
+     provider's console that no Load Balancer is still running.
+
+ * Delete the Persistent Volume Claims:
+
+    * In `values-dev.yaml` all of the persistence claims are labeled as
+      `"helm.sh/resource-policy": keep`. This means that helm will not
+      automatically delete the volume when it deletes the release.
+
+    * List all of your Persistent Volume Claims:
+
+    ```
+    k8s-lbry kubectl get pvc
+    ```
+
+    * Then delete each one you no longer want:
+
+    ```
+    k8s-lbry kubectl delete pvc [name-of-pvc]
+    ```
+
+    * Deleting the claim, should delete the volume. You can verify this yourself
+      in your cloud provider's console that no Volumes exist.
+