I am encountering an issue while trying to deploy my Serverless framework application using the serverless-python-requirements plugin with the following configuration:
custom:
pythonRequirements:
dockerizePip: true
When I run the sls deploy command, I receive the following error:
Serverless: Running docker run --rm -v /home/myuser/.cache/serverless-python-requirements/679f0c7a123e5d80212530da94d545f92ad890ecf43e8288e2b6bc243b4e138b_slspyc:/var/task:z lambci/lambda:build-python3.8 /bin/sh -c 'python3.8 -m pip install -t /var/task/ -r /var/task/requirements.txt && chown -R 0:0 /var/task && find /var/task -name *.so -exec strip {} ;'
Serverless: Stdout:
Serverless: Stderr: ERROR: Could not open requirements file: [Errno 2] No such file or directory: '/var/task/requirements.txt'
It seems like the requirements.txt file is not being found in the /var/task/ directory within the Docker container. I have verified that the requirements.txt file is present in my project directory.
It seems like the requirements.txt file is not being found in the /var/task/ directory within the Docker container.
Has anyone faced this issue before, and how can I resolve it?
The error message indicates that the requirements.txt file is not being found in the Docker container at the expected location /var/task/. This is often due to permission issues with the default cache location.
Here are some steps to resolve this issue:
-
Default Cache Directory: The serverless-python-requirements plugin writes dependencies to
~/.cache/serverless-python-requirements/...before mounting them in the Docker container with -v. If the user running the sls deploy command does not have the necessary permissions to access this cache directory, the/var/taskdirectory in the container will be empty, causing therequirements.txtfile to be missing. Note that Docker will not trigger an error if the volume source does not exist or lacks permissions. -
Change Cache Directory Permissions: Ensure that the user running the sls deploy command has the correct permissions to access the cache directory. See AWS Lambda troubleshooting documentation: https://docs.aws.amazon.com/lambda/latest/dg/troubleshooting-deployment.html
-
Specify Custom Cache Location: If you cannot change the permissions, you can specify a different cache location where you have the necessary permissions using the cacheLocation option in your
serverless.yml:
custom:
pythonRequirements:
dockerizePip: true
cacheLocation: /path/to/mycacheLocation
- Disable Global Cache: Alternatively, you can disable the global cache using the useStaticCache option. This will make Docker bind directly to
path/to/serverless/project/.serverless/requirements:
custom:
pythonRequirements:
dockerizePip: true
useStaticCache: false
- CI/CD Pipeline Issues with Dind: If you are running this in a CI/CD pipeline, there might be issues related to Docker-in-Docker (DinD). For more information on resolving these issues, refer to this GitHub issue.
I hope it can help !