Using Meshery CLI

Meshery’s command line interface is mesheryctl. Use mesheryctl to both manage the lifecyle of Meshery itself and to access and invoke any of Meshery’s application and cloud native management functions. mesheryctl commands can be categorized as follows:

  • mesheryctl - Global overrides and flags
  • mesheryctl app - Cloud Native Application Management
  • mesheryctl filter - Cloud Native Filter Management
  • mesheryctl mesh - Cloud Native Lifecycle & Configuration Management
  • mesheryctl perf - Cloud Native Performance Management
  • mesheryctl pattern - Cloud Native Pattern Configuration & Management
  • mesheryctl system - Meshery Lifecycle and Troubleshooting

Configuring Meshery Deployments with meshconfig

Meshconfig refers to a configuration file found at ~/.meshery/config.yaml. Your meshconfig file must contain one or more contexts in order for any mesheryctl system command to work. Each context represents a Meshery deployment.

Each of the system commands are used to control Meshery’s lifecycle like system start, stop, status, reset and so on.

Meshery CLI FAQ

Question: What is the meshconfig?

Like kubeconfig for kubectl, meshconfig is the name of your mesheryctl config file. You can find your meshconfig file in its default location of ~/.meshery/config.yaml. By default, mesheryctl will look to this location for your meshconfig. You can override the default location at anytime with the use of the global parameter, --config <my-other-meshconfig>.

Question: What is a context?

A meshconfig context represents a single Meshery deployment. Using contexts, you can configure different Meshery deployments with environment-specific settings and you can easily switching between your individual Meshery deployments by updating your current-context.

Question: Why are contexts necessary?

Many Meshery users have more than one Meshery deployment. Contexts allow you to deploy different versions of Meshery, update your release channel subscription settings, selectively install one or more Meshery Adapters, and so on. Contexts allow you to configure your invididual Meshery deployments.

Question: What is current-context?

current-context identifies the Meshery deployment that when any mesheryctl command is invoked will use the environment described in the current-context. You can switch between contexts. Only one context can be the current-context.

Question: What’s the difference between contexts and environments?

Contexts configure Meshery deployments (server, adapters, operator and so on), while environments define a collection of Kubernetes clusters and cloud native infrastructure under management in Meshery.

Question: What does the default meshconfig look like?

The following template is used to create a config file from scratch. Not all of the following variables are required to be included. Some of the variables may have a null value or may be excluded (e.g. “adapters”).

contexts:
    <context1-name>:
      endpoint: <url to meshery server rest api>
      token: <name of token variable in this config file>
      platform: <type of platform: ”docker” or “kubernetes”>
      # Future: specify type of kubernetes (e.g. eks)
      adapters: <collection of names of Meshery adapters:
          “istio”,“linkerd”,”consul”,”nginx-sm”,”octarine”,”tanzu-sm”,”citrix-sm”,”kuma”,”osm”,”nsm”>

   <context2-name>:
    endpoint: <url to meshery server rest api>
    token: <name of token variable in this config file>
    platform: <type of platform: ”docker” or “kubernetes”>
    current-context: <context name>

tokens:
- name: <token1-name>
  location: <token-location>
- name: <token2-name>
  value: <token-value>
  # Future: allow embedding of token certificate

Try it out and see for yourself. Run mesheryctl system context create test and mesheryctl system context view test.

Question: How do endpoints work in meshconfig?

Endpoints specify the access URL for the Meshery UI, for a deployment. Endpoints are developed based on platform:

  • Docker: Docker users can specify the endpoint in the meshconfig. The port specified in this will be used to generate the endpoint. The endpoint is of the form http://localhost:port, where port is taken from the meshconfig.
  • Kubernetes: Deployments with kubernetes as the platform have an endpoint generated by service discovery using the Kubernetes API. This endpoint overwrites the endpoint specified in the meshconfig.

Question: Can I get an API token using mesheryctl?

Yes, if you need to establish a session with your Meshery Server, you can authenticate using mesheryctl, using mesheryctl system login.

Advanced Installation

Users can control the specific container image and tag (version) of Meshery that they would like to run by editing their local ~/.meshery/meshery.yaml (a docker compose file). Aligned with the Meshery container image, instead of leaving the implicit :stable-latest tag behind image: layer5/meshery, users will instead identify a specific image tag like so:


bash
version: '3'
services:
  meshery:
    image: layer5/meshery:v0.5.0
    labels:
      - "com.centurylinklabs.watchtower.enable=true"

Suggested Reading

For an exhaustive list of mesheryctl commands and syntax:

Guides to using Meshery’s various features and components.

Recent Discussions with "mesheryctl" Tag

Discussion Forum

Don't find an answer to your question here? Ask on the Discussion Forum.

Suggested Reading