Fix typos and make minor language edits (#17809)

Fix typos and make minor language edits

I was not sure whether "sports" was a typo for "supports" or not. However, I changed it to "has" which seems to work well.
This commit is contained in:
Peter Hauke 2023-07-24 11:06:31 +02:00 committed by GitHub
parent 4f59d8cafb
commit 47348174e4
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 14 additions and 14 deletions

View File

@ -33,15 +33,15 @@ badges.
![Docker Hub Official and Verified Publisher images](images/hub-official-images.png){:width="700px"}
When building your own image from a Dockerfile, ensure you choose a minimal base
image that matches your requirements. A smaller base image not just offers
image that matches your requirements. A smaller base image not only offers
portability and fast downloads, but also shrinks the size of your image and
minimizes the number of vulnerabilities introduced through the dependencies.
You should also consider using two types of base images: the first image for
development and unit testing and the second for testing during the latest
You should also consider using two types of base image: the first image for
development and unit testing and the second image for testing during the later
stages of development and production. In the later stages of development, your
image may not require some build tools such as compilers, build systems, or
any debugging tools. A small image with minimal dependencies can considerably
image may not require build tools such as compilers, build systems, and
debugging tools. A small image with minimal dependencies can considerably
lower the attack surface.
### Use multi-stage builds
@ -56,11 +56,11 @@ from one stage to another, leaving behind things you dont need in the final
image. This can result in a concise final image.
This method of creating a tiny image doesn't only significantly reduce
complexity, but also the change of implementing vulnerable artifacts in your
complexity, but also reduces the chance of implementing vulnerable artifacts in your
image. Therefore, instead of images that are built on images, that again are
built on other images, multi-stage builds allow you to 'cherry pick' your
artifacts without inheriting the vulnerabilities from the base images on which
they rely on.
they rely.
For detailed information on how to configure multi-stage builds, see
[multi-stage builds](../build/building/multi-stage.md).
@ -73,7 +73,7 @@ instructions which allows you to automate the steps you would normally
libraries and install custom software. These appear as instructions in the
Dockerfile.
Building your image is a snapshot of that image, at that moment in time. When
Building your image is a snapshot of that image at that moment. When
you depend on a base image without a tag, youll get a different base image
every time you rebuild. Also, when you install packages using a package
installer, rebuilding can change the image drastically. For example, a
@ -100,7 +100,7 @@ Consider the following best practices when rebuilding an image:
- Each container should have only one responsibility.
- Containers should be immutable, lightweight, and fast.
- Dont store data in your container. Use a shared data store instead.
- Dont store data in your containers. Use a shared data store instead.
- Containers should be easy to destroy and rebuild.
- Use a small base image (such as Linux Alpine). Smaller images are easier to
distribute.
@ -109,7 +109,7 @@ Consider the following best practices when rebuilding an image:
- Auto-scan your image before deploying to avoid pushing vulnerable containers
to production.
- Analyze your images daily both during development and production for
vulnerabilities. Based on that, automate the rebuild of images if necessary.
vulnerabilities. Based on that, automate the rebuilding of images if necessary.
For detailed best practices and methods for building efficient images, see
[Dockerfile best practices](develop-images/dockerfile_best-practices.md).
@ -120,12 +120,12 @@ In addition to following the best practices outlined on this page when
developing images, it's also important to continuously analyze and evaluate the
security posture of your images using vulnerability detection tools.
Docker tools come with features helps you stay up to date about vulnerabilities
Docker tools come with features that help you to stay up to date about vulnerabilities
that affect images that you build or use.
- Docker Hub supports an automatic
[vulnerability scanning](../docker-hub/vulnerability-scanning.md) feature,
which when enabled, automatically scans images when you push them to a Docker Hub
which when enabled automatically scans images when you push them to a Docker Hub
repository. Requires a [Docker subscription](../subscription/index.md).
- Docker Hub also supports an early-access
[advanced image analysis](../scout/advanced-image-analysis.md) feature, which extends
@ -134,11 +134,11 @@ that affect images that you build or use.
- For the CLI, there's the
[`docker scout` CLI plugin](../engine/reference/commandline/scout.md)
which lets you explore vulnerabilities for images using the terminal.
- Docker Desktop sports a detailed image view for images in your local image
- Docker Desktop has a detailed image view for images in your local image
store, that visualizes all of the known vulnerabilities affecting an image.
All of these security features are powered by the same technology:
[Docker Scout](../scout/index.md). These features help you achieve a holistic
[Docker Scout](../scout/index.md). These features help you to achieve a holistic
view of your supply chain security, and to provide actionable suggestions for
what you can do to remediate those vulnerabilities.