Harbor for Beginners, Part 1: Getting Started

A friendly first tour of the Harbor registry for people from a VMware background. What Harbor is, how to read the Projects page, and what each screen does, with nothing changed and nothing to break.

Harbor for Beginners · Part 1 of 12

If you have spent your time in the VMware world, a container registry can sound like something from a different planet. Take a breath. Harbor is friendlier than it first looks, and a lot of it lines up with things you already do every day. In this first part we are just going to walk around and get our bearings. We will not change a single setting, so nothing can break. By the end you will know what Harbor is for and what each screen does.

What is Harbor?

Harbor is a place to keep container images, and for AI work, model images. An image is just an app or a model that has been packaged up and is ready to run. When your platform goes to start a model or an app, it has to grab that package from somewhere. Harbor is that somewhere. Think of it as the storeroom that everything else pulls from.

If you want it to click quickly, line it up against vSphere ideas you already use. The words are new, the concepts are not.

Harbor calls itYou can think of it asWhat it holds
ProjectA Content LibraryA named folder of images that you can lock down with permissions.
RepositoryA template in that libraryOne named image, for example a single model.
Artifact or TagA version of that templateA specific build, like v1, or a unique fingerprint called a digest.
Robot accountA service accountA login for scripts and pipelines, so you do not hand out your own.

How to log in

Getting in is nothing fancy. You open Harbor in a browser, type a username and password, and click Log In. There is a Remember me box, and a small More info link in the corner that takes you to the docs. That is the whole thing. If you have ever signed into a web tool, you have already done this.

Harbor sign-in screen
The Harbor sign-in screen.

The Projects page

Once you are in, you land on the Projects page. This is home, so it is worth getting comfortable here.

Glance at the top right first. Those two little boxes are a quick headcount, how many projects exist and how many repositories they hold between them. In this lab that is fifteen projects holding well over a hundred repositories, so there is already plenty in here to explore.

The big table underneath is the list of projects. Every column is telling you something handy, so let us read them once and then they will feel obvious forever.

ColumnWhat it is telling you
Project NameThe name of the folder. Click it to step inside.
Access LevelPublic or Private. Public means anyone can pull from it without logging in.
RoleYour role in that project. A dash means you are just a visitor, not a member.
TypeProject for a normal one, or Proxy Cache for one that mirrors another registry.
Repositories CountHow many images live inside.
Creation TimeWhen it was made.
Harbor Projects dashboard with summary tiles and the projects table
The Projects page. Notice the dash in the Role column, that is the tell that you are looking, not owning.

See how your Role is a dash on every row? That is completely normal on a shared system. You can wander into any public project and read it, but you cannot push images to it or change its settings, because none of them are yours. That changes in Part 2, where you make your own project and become its boss.

Opening a project

Click any project name and you drop into it. You will see a few tabs along the top: Summary, Repositories, and Scanner. Repositories is the one you will care about most. It is the actual list of images in that project, and it shows how many versions each one has, how many times it has been pulled, and when it last changed.

A project open on the Repositories tab
Inside a project. Each row is one image, and the Artifacts and Pulls columns show how busy it is.

A registry built for AI fills up with names you will recognize. Open the model store and there they are, model after model sitting side by side, small chat models, coding models, vision models, ones that turn text into numbers for search, and ones that turn speech into text. This is the very shelf a model deployment reaches into when it needs an image, which is a nice way to see how the registry ties into everything else you have built.

The model store project showing many model repositories
The model store, full of the model images the platform deploys from.

Give the Scanner tab a quick look while you are here. Harbor ships with a built-in image scanner, and most setups plug in one called Trivy. The tab just shows its name, a little Healthy badge, and a version number. The thing to take away is simple: checking images for security problems is part of Harbor itself, not a separate tool you add on later. You will put it to work in Part 5.

The Scanner tab showing Trivy with a Healthy status
The built-in scanner. A green Healthy badge means it is ready to go.

The logs

The Logs link on the left is the history book. It keeps a record of who did what and when, the pushes, pulls, deletes, and setting changes. Right now it may be empty, which only means nothing has happened in the slice of time you are looking at. Once you start pushing your own images in later parts, this is where your footprints show up.

The Logs and Audit Logs view
The activity log. Quiet for now, but it fills up the moment you start working.

What you can do right now

Even as a visitor you can get plenty of practice without touching anything you should not:

  • Wander into any public project and read its list of images.
  • Open an image and look at its versions, its size, and how many times it has been pulled.
  • Check a project Scanner tab and confirm it is healthy.
  • Stare at the table until Public versus Private, and that Role dash, feel like second nature.

What you cannot do yet is push images or change settings. For that you need a project where you are a member, and making one is exactly what we do next.


Harbor for Beginners, Part 1 of 12. Product names belong to their owners. The names and IDs in the screenshots come from a lab and are stand-ins.

About The Author


Discover more from Journal of Intelligent Infrastructure – By Dr Pranay Jha

Subscribe to get the latest posts sent to your email.

Leave a Reply

Your email address will not be published. Required fields are marked *

Architect’s Toolkit

About the Author

Dr. Pranay Jha is a Cloud and AI Consultant with 18+ years of experience in hybrid cloud, virtualization, and enterprise infrastructure transformation. He specializes in VMware technologies, multi-cloud strategy, and Generative AI solutions. He holds a PhD in Computer Applications with research focused on Cloud and AI, has published multiple research papers, and has been a VMware vExpert since 2016 and a VMUG Community Leader.

Discover more from Journal of Intelligent Infrastructure - By Dr Pranay Jha

Subscribe now to keep reading and get access to the full archive.

Continue reading