How to Ignore directories and files for a deployment

Serverless Framework
1

Hi!

Problem: my zipped deployment folder that contains my python code is too big!
I looked inside and there’s stuff in there that should be ignored according to my .gitignore

Here is my .gitignore

image

Here is the contents of my .gitignore

image
image383×501 28.9 KB

Here is the contents of the (too-big) zipped folder that is deployed to S3
(I DL it from S3 and unzipped the contents)

image

I should note that my service project is nested one level below another service project, like so…
-root
—main.project.with.own.deployment
------my.project.with.big.zip
---------.gitignore
---------handler.py
--------etc…

This is the command I’m using to deploy

sudo sls deploy -f numpy

Thanks!

2

Serverless does not pay attention to .gitignore. You have a few options for you to take that I use myself.

  1. You can add options to the serverless.yml to indicate what to exclude and what to include. You can read about that at https://serverless.com/framework/docs/providers/aws/guide/packaging/

  2. What I do most recently is use gulp to build a deployment folder containing all my files/folders. Then within that folder is what I actually deploy. This allows me to have more processes in place for CI/CD.

3

Perfect. Thanks so much!

4

We a single serverless.yml that contains a bunch of resources and 4 python lambdas. Each lambda needed a different set of python libraries and pre-compiled binaries. I found the best way to hald this was to “package individually” and exclude everything:

package:
  individually: true
  exclude:
    - ".*/**"

Then in each lambda, include only the binaries/libs I needed. e.g.,

functions:
  DocPdfSplit:
    handler: doc_pdf_split.s3created
    package:
      include:
        - doc_pdf_split.py
        - lib.py
        - ghostscript-9.20/**
    events:
      - s3:
          bucket: ${self:custom.s3_name}
          event: s3:ObjectCreated:*

and

  PageTxtCombine:
    handler: page_txt_combine.s3created
    package:
      include:
        - page_txt_combine.py
        - lib.py
        - elasticsearch/**
        - urllib3/**
    events:
      - s3:
          bucket: ${self:custom.s3_name}
          event: s3:ObjectCreated:*
5

What @shentenfruede said above works well putting “exclude” under “package”. On Mac I needed to use:

package:
  individually: true
  exclude:
    - "*/**"
6

With the new release, include and exclude are depreciated and will be removed completely in future releases. Please adhere to

patterns
Here is an example:

# Welcome to Serverless!
#
# This file is the main config file for your service.
# It's very minimal at this point and uses default values.
# You can always add more config options for more control.
# We've included some commented out config examples here.
# Just uncomment any of them to get that config option.
#
# For full config options, check the docs:
#    docs.serverless.com
#
# Happy Coding!

service: python-sls-test

frameworkVersion: '3'
configValidationMode: error

custom:
  SERVICE_NAME: ${env:SERV_NAME, 'python-default'}
  REGION: ${env:AWS_REGION, 'us-east-1'}

provider:
  name: aws
  runtime: python3.9
  stage: ${opt:stage, 'dev'}
  region: us-east-1
  deploymentMethod: direct
  memorySize: 512
  deploymentPrefix: serverless
  deploymentBucket:
    name: sls-vj-deploy
    maxPreviousDeploymentArtifacts: 3



# you can add packaging information here
package:
 
 patterns:
   - '!exclude-me.py'
   - '!venv/**'
   - '!README.md'
   - '!build/**'
  # Package each function individually 
 individually: true


functions:
  testFunction:
    handler: lambdas/handler.test
    package: 
      patterns:
        - lambdas/**