I have this command in my docker file:
ADD static/ /www/static/
I have noticed that re-running docker build reuses the cache, even though the contents of the static/ directory have changed. Is this normal?
How does docker decide when a layer needs to be rebuilt? Just by looking at the command that needs to be executed, or by checking the actual operation performed? I assume is the former, since the latter would require re-running the operation, defeating the purpose of caching.
The workaround that I am using now is --no-cache but this makes building slower, since no layer is reused. Is there a better way?
I think the best option would be to mark one operation as non-cacheable. Is this possible?
According to Dockers website, the cache for a specific layer should be invalidated if the instruction has changed.
However, for ADD and COPY, the checksums of the files are compared and if these have changed, the cache is invalidated.
Therefore it seems that the contents of the files in static/ have not changed. So to be sure that you might see strange behaviour, please execute a checksum over the files in static/ before the first time you build, and before the second time - when you rebuild with the updated files.
Related
I am wondering if it is possible to swap out a layer of a container image for another. Here is my scenario:
I have a docker file that does the following:
It pulls the .net core 3.1 debian slim (buster) image
Builds the source code of my application and adds it as a layer on top of the .net core image
When a new version of the .net core 3.1 runtime image comes out, I would like to make a new image that has that new version, but has the same application layer on top of it.
I do NOT want to have to find the exact version of the code I used to build the application and re-build.
The idea is to replicate upgrading the machine and runtime, but not have any alteration to the application (less to test with the upgrade).
Is there a docker command that I can use to swap out a layer of an image?
This answer is broken into two distinct sections:
Part I: Build From Sources Instead: Explains why this idea should not be done.
Part II: A Proof-of-concept Implemented CLI Tool: A CLI tool I implemented specifically for this question, with a demonstration.
The first section can be skipped if potentially adverse side effects are not a concern.
Part I: Build From Sources Instead
No, there is probably not a docker command for this - and likely never will, due to a whole slough of technical issues. Even by design, this is not meant to be trivially possible; images and layers are meant to be immutable after being built. It is strongly recommended that the application image is instead rebuilt from the original sources, but with a new base image by modifying the FROM command.
There are numerous consistency issues that make this idea ill-advised, of which a few are listed below:
Certain Dockerfile commands do not actually create a layer, they update the final layer's internal manifest and the image's manifest. Commands that state Removing intermediate container XXXXXXXXXXXX are actually updating these aforementioned manifests, and not creating new layers. This requires correctly updating the only relevant changes when swapping from the old base image to the new base image; e.g. reconciling changes from ENV/LABEL/MAINTAINER/EXPOSE/CMD/ENTRYPOINT commands.
ENV commands that alter the application image's configuration from variables inherited in the previous base image may not be updated correctly. For example, in the application image, there might be the following command ENV command:
ENV PATH="/path:${PATH}"
If the application image's old base image layers are swapped out, and the new base image contains a different ${PATH} variable, it is ambiguous how to reconcile the differences without a manual decision by a developer.
If apt/apt-get/apk/yum is used to install Linux packages, these packages are installed in the application image as subsequent layers. The installed packages are not guaranteed to be compatible with the new base image if the underlying OS or executables change in the new base image layers.
Part II: A Proof-of-concept Implemented CLI Tool
"Upgrading" an image's base is technically possible by doing direct manipulation on the image archive of the already-built Docker application image. However, I cannot reiterate this enough - you should not even be attempting to edit the bottom layers of an existing image. Seriously - stop it, get some help. I am 90% sure this is a war crime.
For the sake of humoring this question thoroughly though, I have developed a proof-of-concept CLI tool written in Java over on my GitHub project concision/docker-base-image-swapper that is designed to swap base images (but not arbitrary layers). The design choices I made to resolve various consistency issues are a "best guess" on what action should be taken.
Included is a demonstration for swapping the base image for an already-built Java application image from JDK8 to JDK11, implemented in the demo/demo.sh script. All core code is ran in isolated Docker contains, so only Bash and Docker are necessary dependencies on the host to run this demonstration. The demo application image is only built once on JDK 8, but is run twice - once on the original JDK 8 image, and another time on a swapped-base JDK 11 image.
If you experience some technical difficulty with the tool, I may potentially be able to fix the issue. This project was quickly hacked together and has pretty poor code quality; furthermore, it likely suffers from various unaccounted edge cases. I might thoroughly rewrite this in Rust within the next month or two with a focus on maintainability and handling all edge cases.
Warning: I still do not advise trying to edit images; the use of the tool is at your own risk.
The Concept
There are three images relevant in this process, all of which are already-built (i.e. no original sources are needed):
application image: the application image that is currently based off of the old base image, but needs to be swapped to a newer base image.
old base image: the older base image that the application image is based off using a FROM command in the original Dockerfile.
new base image: the newer base image that should replace the application image's layers inherited solely from the FROM old-base-image layers.
By knowing which layers and configurations of the application image are inherited from the old base image, they can be replaced with the layers and configurations from the new base image. All image layers and manifests can be obtained as a tar archive by using the docker save command. With an archive(s) of all relevant three images, a tool can analyze the differences between
Warnings on Alternatives
Beware of doing simply a COPY --from=... from the old application image, as the original application's image configuration (through commands such as CMD, ENTRYPOINT, ENV, EXPOSE, LABEL, USER, VOLUME, WORKDIR) will not be properly replicated.
Two tools that provide this type of functionality that I have found are listed below.
Crane rebase https://github.com/google/go-containerregistry/blob/main/cmd/crane/rebase.md#using-crane-rebase
Buildpack (rebase) https://buildpacks.io/docs/concepts/operations/rebase/
NOTE: You do have to watch that you don't have items that need updates in higher layers. And that higher layers do NOT have updates in them that would conflict with the new base. Example using image below the base is a TOMCAT container. If layer 4 of the original container did a update tomcat package that would overshadow the update of same files on the NEW BASE. Which could result in a non functional system. So as always your mileage may vary.
Is it possible? Yes. But it's rarely done because of how error prone it is. For example, if the previous build would have created a library, but the library already exists in the original base image, that library won't be included in the first build. If the base image removes that library, then the resulting merged image will be missing that library since it's not in the new base layers and isn't in the old application layer.
There are a few ways I can think of to do this.
Option 1: If you know the specific files in one image, then use the COPY --from syntax to copy those files between images. This is the least error prone method, but requires that you know every file you want to include. The resulting Dockerfile looks like:
FROM new_base
# if every file is in /usr/local/bin:
COPY --from=old_image /usr/local/bin/ /usr/local/bin/
Option 2: you can export the images and create your own new image by combining the layers between the two. For this, there's docker save and docker load. That would look like:
docker save old_image >old_image.tar
docker save new_base >new_base.tar
mkdir old_image new_base new_image
tar -xvf old_image.tar -C old_image
tar -xvf new_base.tar -C new_base
cp old_image/*.json new_image/
# manually: copy each layer directory you want to save from old_image, you can view the nested tar files to find the proper ones
# manually: copy each layer directory from new_base into new_image
# manually: modify the new_image/manifest.json to have the new tag for your image, and adjust the layer references
tar -cvf new_image.tar -C new_image .
docker load <new_image.tar
Option 3: this could be done directly to the registry with API calls. You would pull the old manifest, adjust it with the new layers, and then push any new layers and then the new manifest. This would require a fair bit of coding (see regclient/regclient for how I've implemented some of these API's in Go).
Option 4: I know I've seen a tool that does this in a specific scenario but the name of it escapes me. I believe it required that you use their base images that were curated to reduce the risk of incompatibilities between versions, and limited what base images could be swapped, so I don't think it was a general purpose tool.
Note that option 2 and 3 both require either manual steps or for you to write some code if you want to automate it. Because of how error prone this is (as described above) I don't think you'll find anyone maintaining and supporting a tool to implement it. The vast majority rebuild from a Dockerfile using a CI tool.
I have developed a small utility script using Python which lets you append a tarball to an existing container image in a container registry (without having to pull existing image data), available at https://github.com/malthe/appendlayer.
The following illustrates how it works:
[ base layer ] => image1
[ base layer ] [ appended layer ] => image2
The script supports any registry that implements the OCI Distribution Spec.
The title of this question might suggest that it has already been answered, but trust me, I searched intensively here on SO :-)
As I understand it when building a docker-image the current folder will be packaged up and sent to the docker-deamon as the build-context. From this build-context the docker-image is build by "ADD"ing or "COPY"ing files and "RUN"ning the commands in the Dockerfile.
And furthermore: In case I have sensitive configuration-files in the folder of the DockerFile, these files will be sent to the docker-deamon as part of the build-context.
Now my question:
Lets say I did not use any COPY or ADD in my Dockerfile... will these configuration files be included somewhere in the docker-image? I ran a bash inside the image and could not find the configuration-files, but maybe they are still somewhere in the deeper layers of the image?
Basically my question is: Will the context of the build be stored in the image?
Only things you explicitly COPY or ADD to the image will be present there. It's common to have lines like COPY . . which will copy the entire context into the image, so it's up to you to check that you're not copying in things you don't want to have persisted and published.
It still is probably a good idea to keep these files from being sent to the Docker daemon at all. If you know which files have this information, you can add them to a .dockerignore file (syntax similar to .gitignore and similar files). There are other ways to more tightly control what's in the build context (by making a shadow install tree that has only the context content) but that's a relatively unusual setup.
As you said only COPY, ADD and RUN operations create layers, and therefore, only those operations add something to the image.
The build context is only the directory with the resources those operations (specifically COPY and ADD) will have access to while building the image. But it's not anything like a "base layer".
In fact, you said you ran bash and double checked that nothing sensitive was there. Another way to make sure about this is by checking the layers of the image. To do so, docker history --no-trunc <image>
docker build . will rebuild the docker image given the Dockerfile in the current directory, and ignore any paths matched from the .dockerignore file.
Any COPY statements in that Dockerfile will cause the build cache to be invalidated if the files on-disk are different from last time it built.
I've noticed that if you don't ignore the .git dir, simple things like git fetch which have no side-effect will cause the build cache to become invalidated (presumably because some tracking information within the .git dir has changed.
It would be very helpful if I knew how to see precisely which files caused the cache to become invalidated... But I've been unable to find a way.
I don't think there is a way to see which file invalidated the cache with the current Docker image design.
Layers and images since v1.10 are 'content addressable'. Their ID's are based on a SHA256 checksum which reflects their content.
The caching code just looks up the ID of the image/layer which will only exist in Docker Engine if the contents of the entire layer match (or possibly a collision).
So when you run docker build, a new build context is created for each command in the Dockerfile. A checksum is calculated for the entire layer that command would produce. Then docker checks to see if an existing layer is available with that checksum and run config.
The only way I can see to get individual file detail back would be to recompute the destination file checksums, which would probably negate most of the caching speed up. If you did want to do this anyway, the other problem is deciding which layer to check that against. You would have to lookup a previous image build tree (maybe by tag?) to find what the contents of the previous comparable layer were.
I'm building a Yocto image for a project but it's a long process. On my powerful dev machine it takes around 3 hours and can consume up to 100 GB of space.
The thing is that the final image is not "necessarily" the end goal; it's my application that runs on top of it that is important. As such, the yocto recipes don't change much, but my application does.
I would like to run continuous integration (CI) for my app and even continuous delivery (CD). But both are quite hard for now because of the size of the yocto build.
Since the build does not change much, I though of "caching" it in some way and use it for my application's CI/CD and I though of Docker. That would be quite interesting as I could maintain that image and share it with colleagues who need to work on the project and use it in CI/CD.
Could a custom Docker image be built for that kind of use?
Would it be possible to build such an image completely offline? I don't want to have to upload the 100GB and have to re-download it on build machines...
Thanks!
1. Yes.
I've used docker to build Yocto images for many different reasons, always with positive results.
2. Yes, with some work.
You want to take advantage of the fact that Yocto caches all the stuff you need to do your build in what it calls "Shared State Cache". This is normally located in your build directory under ${BUILDDIR}/sstate-cache, and it contains exactly what you are looking for in this case. There are a couple of options for how to get these files to your build machines.
Option 1 is using sstate mirrors:
This isn't completely offline, but lets you download a much smaller cache and build from that cache, rather than from source.
Here's what's in my local.conf file:
SSTATE_MIRRORS ?= "\
file://.* http://my.shared-computer.com/some-folder/PATH"
Don't forget the PATH at the end. That is required. The build system substitutes the correct path within the directory structure.
Option 2 lets you keep a local copy of your sstate-cache and build from that locally.
In your dockerfile, create the sstate-cache directory (location isn't important here, I like /opt for my purposes):
RUN mkdir -p /opt/yocto/sstate-cache
Then be sure to bindmount these directories when you run your build in order to preserve the contents, like this:
docker run ... -v /place/to/save/cache:/opt/yocto/sstate-cache
Edit the local.conf in your build directory so that it points at these folders:
SSTATE_DIR ?= "/opt/yocto/sstate-cache"
In this way, you can get your cache onto your build machines in whatever way is best for you (scp, nfs, sneakernet).
Hope this helps!
I'm learning to use Docker and I've come across a minor annoyance. Whenever I make a change to the Dockerfile,I run docker build -t tag . which goes through the entire Dockerfile as it should. This takes a good 5-6 minutes due to the dependencies in my project. Sometimes a command that I run will cause an error, or there will be a mistake in the Dockerfile. While the fix may take a couple seconds, I have to rebuild the entire thing which decreases my productivity. Is there a way to "continue from where the build last failed" after editing the Dockerfile? Thanks.
This is called the "build cache" and it is already a feature of Docker. Docker's builder will only use the cache up until the point where your Dockerfile has changed. There are some edge cases when using COPY or ADD directives that will cause the build cache to be invalidated (since it hashes files to determine if any have changed, and invalidates the cache if so). This means that if you are using COPY foo /foo and you have changed that file, the build cache will be invalidated. Also, if you do COPY . /opt/bar/ (meaning, you copy the entire directory to somewhere), even some small change like a Vim swap file or Dockerfile change will invalidate the cache!
The behavior of not using the build cache at all is invoked using --no-cache in your docker build command.
So basically, it's there, and you're using it, just that you're probably changing the Dockerfile at a very early point or hitting that lesser known edge case with a COPY/ADD directive, and the builder is invalidating everything after that point. And just to answer the question before you ask it, it would be very hard or impossible to continue using the cache after a change has invalidated the cache. Meaning, if you change your first Dockerfile line and invalidate the build cache, it is basically impossible to use the build cache past that point.
Is there a way to "continue from where the build last failed" after editing the Dockerfile?
No (as L0j1k's answer explains well)
That is why the best practice is to organize your Dockerfile from the stablest commands (the one which will never have to be changed/modified) to the most specific commands (the ones you might have to change quite a bit).
That way, your modifications will trigger only a build on the last few lines of your Dockerfile, instead of going through everything again, because you changed one of the first lines.