Skip to content

Latest commit

 

History

History
207 lines (148 loc) · 4.69 KB

aws_cli.md

File metadata and controls

207 lines (148 loc) · 4.69 KB
Raw

AWS Command Line

  • AWS CLI is a case sensitive framework

Configure

- Set up initial configuration

$ aws configure
$ AWS Access Key ID [****************WYFA]: XXXXXXXXXXXXXXXXXXXXXXXXX
$ AWS Secret Access Key [****************W5F2]: XXXXXXXXXXXXXXXXXXXXXXXXXXXXX
$ Default region name [us-east-1]:
$ Default output format [None]:
  • aws credential are stored in ~/.aws/
  • Output formats are: JSON (default), text, table

- Change the default output format

$aws ec2 describe-instances --output table | text | json

- Getting Help

$ aws --version
$ aws aim help

- Set up Autocomplete

http://docs.aws.amazon.com/cli/latest/userguide/cli-command-completion.html

  • get the shell and aws_completer location
$ echo $SHELL
/bin/bash
$ which aws_completer
/staff/ppeiris/anaconda2/envs/tensorflow/bin/aws_completer
  • set up with the complete command
$ complete -C '/staff/ppeiris/anaconda2/envs/tensorflow/bin/aws_completer' aws

Now you can double tap TAB key to autocomplete aws commands

- Dry Run

Dry run allow to run the command line options without really making any real effect. This wont give us any output, it will give a message saying the command would have success or fail.

$ aws ec2 describe-regions --dry-run

Output

An error occurred (DryRunOperation) when calling the DescribeRegions operation: Request would have succeeded, but DryRun flag is set.

Process Data

- Filters Option

You can use filters to limit the output of a command. This filtering happen on the server side.

$ aws ec2 describe-instances --filter Name=instance-type,Values=t2.micro

- Query Option

You can also use the query option to limi the results. This happen on the client side (i.e: this is very similar to grep option)

$ aws ec2 describe-regions --query 'Regions[?RegionName==`us-west-1`]'

Output

[
    {
        "RegionName": "us-west-1",
        "Endpoint": "ec2.us-west-1.amazonaws.com"
    }
]

Process JSON data using Jamespath query language

JAMESpath is a query language for JSON data (http://jmespath.org/).

Install using conda

conda install -c conda-forge jmespath

Install JAMESpath terminal

https://github.com/jmespath/jmespath.terminal

pip install jmespath-terminal

- How to use jpterm

Following command get the list of reagions in JSON format and open up the jamespath terminal to interact with the data.

aws ec2 describe-reagions | jpterm

jamespath

image credit to https://github.com/jmespath/jmespath.terminal

  • Exit using Ctr+C
  • You can build comples queries in jamespath terminal and then use then with Query option directly to use in programatic way.

CLI Commands for AWS S3

- Create a bucket (make bucket - mb)

  • create a bucket call 'unique-name-of-the-bucket'.
  • More help using - aws s3 mb help
$ aws s3 mb s3://unique-name-of-the-bucket

- Copy files to a bucket (copy - cp)

  • Copy the file call FILE_NAME to the bucket and a directly within in call dir-name
  • More help using - aws s3 cp help
$ aws s3 cp FILE_NAME s3://unique-name-of-the-bucket/dir-name/

- List all the buskets

  • This command will display all the buskets that are in the account.
$ aws s3 ls

- List all the files within a bucket

aws s3 ls s3://unique-name-of-the-bucket

- Move all the files from S3 to local dir (mv --recursive)

  • This move all the files from S3 bucket to local folder (remove from the S3)
$ aws s3 mv s3://unique-name-of-the-bucket . --recursive

- Move files from local dir to S3

  • exclude and include flags can be use to pick which files need to be move
$ aws s3 mv . s3://unique-name-of-the-bucket --recursive --exclude '*' --include '*.txt'

- Sync local filed with S3 (|S3 to local | S3 to S3)

  • This will sync local files to S3
    • This update the files that nees updates
    • New files get created in S3
$ aws s3 sync . s3://unique-name-of-the-bucket

- Sync files with delete option

  • This flag delete the files that are deleted from the local folder
$ aws s3 sync . s3://unique-name-of-the-bucket --delete

- Upload a file with public read permission (acl)

$ aws s3 cp FILE_NAME s3://unique-name-of-the-bucket --acl public-read

- Delete files from S3 (rb)

  • Delete files from S3
$ aws s3 rb FILE_NAME s3://unique-name-of-the-bucket
  • Delete all the files from S3 & delete the bucket
$ aws s3 rb s3://unique-name-of-the-bucket --force

Update aws cli to version 2 in mac

$ curl "https://awscli.amazonaws.com/AWSCLIV2.pkg" -o "AWSCLIV2.pkg"
$ sudo installer -pkg AWSCLIV2.pkg -target /