Remove the Waypoint Server

The Waypoint Server creates several resources in Docker and Kubernetes that should be removed to either reinstall Waypoint or to completely remove it from a system.

Run waypoint destroy on your workspaces prior to removing the server to prevent hanging resources.

Uninstall

Starting in 0.2.1, the server uninstall command will remove the server and supporting containers and services that are created via install. By default, it also creates a snapshot and deletes the server context.

For Docker, uninstall removes the Waypoint server-specific volumes, but refrains from doing a global volume prune. If you created deployments with Waypoint, there will likely be volumes left over from those deployments. You can remove these volumes individually, or globally via:

$ docker volume prune -f

If the uninstall fails, see below for manual removal.

Manual removal

Waypoint Server in Docker

Ensure that you have the latest hashicorp/waypoint Docker image.

$ docker pull hashicorp/waypoint:latest

waypoint install for Docker creates a container and a volume. These resources should be removed when Waypoint Server is no longer needed. These are some example docker commands that should clean up after a Waypoint Server installation.

$ docker stop waypoint-server
$ docker rm waypoint-server
$ docker volume prune -f

Waypoint Server in Kubernetes

waypoint install for Kubernetes installs the Waypoint Helm chart to your Kubernetes cluster. To delete the resources created by the Helm chart installation, the commands below will help with the cleanup of the server and static runner:

$ helm uninstall waypoint
$ helm uninstall waypoint-static
$ kubectl delete pvc data-waypoint-server-0
$ kubectl delete pvc data-default-waypoint-static-runner-0

Pack Builder No Such Image

If the pack builder exits during build with an error similar to:

» Building...
Creating new buildpack-based image using builder: heroku/buildpacks:18
...
! failed to create 'detector' container: Error response from daemon: No such
  image: pack.local/builder/6c647962686577687270:latest

You may need to remove the Docker volumes that are responsible for storing cached images for pack.

$ docker volume list
...

Locate the volumes named starting with pack-cache- and remove them with docker volume rm.

Remote Operations Failure

If your Waypoint operation fails, task introspection can help troubleshoot the issue. Logs from the on-demand runner may not be readily available from the platform in which the runners are launched, and task introspection can get the information needed to troubleshoot.

In the example below, after a build failure, waypoint task inspect was used with the task ID. The Run Job results show that there was an error with the Docker plugin which required arguments which were missing:

waypoint task inspect 01G669RE32WM0JBBGZW3JETYSJ

» On-Demand Runner Task Configuration
      Task ID: 01G669RE32WM0JBBGZW3JETYSJ
   Task State: STOPPED
    Workspace: default
      Project: example-go
  Application: example-go

» Run Job Configuration
  Job Configuration:
         Job ID: 01G669RE30ND9SKP50RYWP1B5H
      Operation: Build
  Target Runner: 01G669RE300HV2D1QPSYMSKHP2
      Workspace: default
        Project: example-go
    Application: example-go

  Job Results:
           State: Error
   Complete Time: 33 minutes ago
  Error Messsage: There was an error while executing a Waypoint plugin for this operation!

One or more required arguments for the plugin was not satisfied. This is usually due to a missing or incompatible set of plugins. For example, only certain build plugins are only compatible with certain registries, and so on. Please inspect the missing argument, the set
of plugins you are using, and the documentation to determine if your plugin combination is valid.

Plugin function: github.com/hashicorp/waypoint/builtin/pack.(*Builder).BuildODR-fm

==> Missing arguments:

    - docker.AccessInfo


» Watch Job Configuration
  Job Configuration:
         Job ID: 01G669RE32M1MGQBYAVND110SS
      Operation: WatchTask
  Target Runner: *
      Workspace: default
        Project: example-go
    Application: example-go

  Job Results:
          State: Success
  Complete Time: 33 minutes ago

» Start Job Configuration
  Job Configuration:
         Job ID: 01G669RE32M558VJZKRGYQ9W3S
      Operation: StartTask
  Target Runner: *
      Workspace: default
        Project: example-go
    Application: example-go

  Job Results:
          State: Success
  Complete Time: 33 minutes ago

» Stop Job Configuration
  Job Configuration:
         Job ID: 01G669RE32HSPGP4AXDSA0NZ7S
      Operation: StopTask
  Target Runner: *
      Workspace: default
        Project: example-go
    Application: example-go

  Job Results:
          State: Success
  Complete Time: 33 minutes ago

Investigating Deployed Applications

Waypoint includes several commands to support debugging and monitoring while developing your application.

Exec Into the Application Container

After deploying your application, you can use waypoint exec to run commands in the context of the most recent deployment. Typically, waypoint exec will be used for running database migrations and debugging. However, you can use it for any purpose.

Use the exec command to open a shell prompt.

$ waypoint exec /bin/bash

Since you are in the deployment directory, you will observe that Waypoint automatically executes against the currently deployed application.

From within the Docker container, validate that this is the actual application by listing out the directory hosting the application's compiled files. For example, if you used the pack builder this directory will be /workspace.

$ cd / && ls

You should observe an output that contains the file structure for the current deployment.

List the processes that are running in the container.

$ ps aux

Type exit to leave the interactive Docker session.

$ exit

View Waypoint Application and Entrypoint Logs

In the application's directory, run the logs command to observe the running logs for your deployment. This will include logs from the entrypoint if it is in use.

$ waypoint logs

You will observe output similar to the following. These logs are from the existing deployment.

2020-09-24T06:20:18.162Z 2MGFF4:
2020-09-24T06:20:18.163Z 2MGFF4: > node-js-getting-started@0.3.0 start /workspace
2020-09-24T06:20:18.163Z 2MGFF4: > node index.js
2020-09-24T06:20:18.163Z 2MGFF4:
2020-09-24T06:20:18.383Z 2MGFF4: Listening on 3000

Press Ctrl-C to exit the logs command.

You may increase the verbosity of entrypoint logs by setting the WAYPOINT_LOG_LEVEL environment variable. See the documentation on entrypoint logs for more information.

Access the Waypoint web UI

The Waypoint server includes a web-based user interface that you can use to view builds, deployments, and releases for projects and applications.

$ waypoint ui -authenticate

$ waypoint user token

Review the application metadata and associated operations and logs in the browser.