1
1
mirror of https://github.com/namibia/awesome-cheatsheets.git synced 2024-11-30 16:53:59 +00:00
awesome-cheatsheets/tools/aws.sh
2020-09-21 17:06:46 +07:00

40 lines
2.5 KiB
Bash

##############################################################################
# AWS
##############################################################################
# General
aws help
aws --version # Show the current AWS CLI version
aws configure # Configure your AWS Key ID, AWS Secret, default region and default output format for the AWS CLI
aws configure --profile <profile_name> # Configure using the profile name. By default, the list of profile is stored in ~/.aws.credentials (Linux and MacOS)
# EC2
## We need to specify a region to use ec2 commands. We can configure a default region with "aws configure" or set the AWS_DEFAULT_REGION environment variable before the command line
## Example: AWS_DEFAULT_REGION=us-east-1 aws ec2 describe-instances
aws ec2 describe-instances # Desribe all instances in the current region
aws ec2 describe-instances --instance-ids <instance_id_1> <instance_id_2> # Describe specific instances by their IDs
aws ec2 describe-instances --filters Name=<instance_name> # Filter and describe instances by name
aws ec2 start-instances --instance-ids <instance_id_1> <instance_id_2> # Start previously stopped instances by their IDs
aws ec2 stop-instances --instance-ids <instance_id_1> <instance_id_2> # Stop running instances by their IDs
aws ec2 terminate-instances --instance-ids <instance_id_1> <instance_id_2> # Shutdown the specific instances by their IDs
# S3
## To specify the root directory of a S3 bucket, use this syntax: s3://<bucket_name>
aws s3 ls # List S3 objects and common prefixes under a prefix or all S3 buckets
aws s3 ls s3://<bucket_name> # List objects and common prefixes under a specified bucket and prefix
aws s3 mb s3://<bucket_name> # Create a specific S3 bucket
aws s3 rb s3://<bucket_name> # Remove an empty specific S3 bucket by name
aws s3 mv <local_file_path> s3://<bucket_name>/<destination_file_path> # Move a file in local_file_path to a specific bucket in destination_file_path
## Example: aws s3 mv text.txt s3://mybucket/text.txt
aws s3 mv s3://<bucket_name_1> s3://<bucket_name_2> --recursive # Move all objects from bucket_name_1 to bucket_name_2
aws s3 sync <source> <target> # Sync all contents from source to a target directory. This will copy and update all missing or outdated files or objects between source and target
## Examples: aws s3 sync . s3://mybucket
## aws s3 sync s3://bucket_1 s3://bucket_2
aws s3 sync <source> <target> --delete # Sync all contents from source to target, but this will remove all missing files and objects from the target that are not present in source