AWS S3 - CLI: Copy and Sync data, Static Website Hosting, mount S3 Bucket as file system with s3fs

AWS CLI

#

Setup and Config

#

Troubleshooting

#

If AWS CLI sync does not work it could be cause of a wrong server time.

AWS CLI with S3

#

Copy and Sync data

#

Create & Delete S3 Buckets

#

# Create S3 Bucket
aws s3 mb s3://bucketname

# Remove S3 Bucket
aws s3 rb --force s3://bucketname

Sync to S3 Bucket

#

# Uploads only new or changed files
aws s3 sync /source/path s3://bucketname/destination

# Uploads only new or changed files: Delete files from the destination that no longer exist in the source directory
aws s3 sync /source/path s3://bucketname/destination --delete

# Only *.pdf files
aws s3 sync --exclude="*" --include="*.pdf" /source/path s3://bucketname/destination

# Exclude *.pdf files
aws s3 sync --exclude="*.pdf" /source/path s3://bucketname/destination

# Use specific account
aws s3 sync /source/path s3://bucketname/destination --profile account1

Sync from S3 Bucket

#

# Download only new or changed files
aws s3 sync s3://bucket-name/source /destination/path

Copy to & from S3 Bucket

#

# Upload files from bucket (include sub dirs)
aws s3 cp --recursive /source s3://bucketname/destination

# Download files from bucket (include sub dirs)
aws s3 cp --recursive s3://bucketname /destination

Note: “aws s3 cp” command does not delete files from the destination bucket when they no longer exist in the source directory.

Static Website Hosting

#

Create S3 Bucket
aws s3 mb s3://BucketName

Copy index.html file to Bucket:
aws s3 cp /path/index.html s3://BucketName/index.html

Policy: Allow read-only access to every object in the bucket

bucketpolicy.json

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Sid": "PublicReadGetObject",
            "Effect": "Allow",
            "Principal": "*",
            "Action": [
                "s3:GetObject"
            ],
            "Resource": [
                "arn:aws:s3:::BucketName/*"
            ]
        }
    ]
}

Add Policy to S3 Bucket:

$ aws s3api put-bucket-policy --bucket $BucketName \
--policy file://$pathToPolicy/bucketpolicy.json

Define name of index document:
aws s3 website s3://$BucketName --index-document index.html

Open Website:
http://BucketName.s3-website-us-east-1.amazonaws.com

AWS s3fs

#

Create IAM Permission

#

• Create a S3 Bucket with standard settings, “Block public access” should be enabled by default
• Create an IAM Permission to access the S3 Bucket
• Attach the Permession to an IAM User
• Create an “Access Key” for the User

IAM Permission: Replace name_of_your_s3-bucket with the name of your S3 Bucket

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": [
                        "s3:GetBucketLocation",
                        "s3:ListAllMyBuckets"
                      ],
            "Resource": "arn:aws:s3:::*"
        },
        {
            "Effect": "Allow",
            "Action": "s3:*",
            "Resource": [
                "arn:aws:s3:::name_of_your_s3-bucket",
                "arn:aws:s3:::name_of_your_s3-bucket/*"
            ]
        }
    ]
}

Part 2: Automatic mount S3 Bucket on system startup

#

Uncomment the user_allow_other string in /etc/fuse.conf file to allow S3 for other (non root) users:
vi /etc/fuse.conf

Replace name_of_your_s3-bucket with the actual name of the S3 Bucket.

Part 3: Manually mount S3 Bucket

#

Alternative the S3 Bucket can also be manually mounted.

Use mount | grep sftp_data or df -h | grep sftp_data or just ls /mount_pathto check whether the bucket is mounted. It is not possible to mount the S3 Bucket when the user is in the mountpoint directory while trying to mount the S3 Bucket.