# How to use AWS CLI v2 and s3cmd with e3 Object Storage

## e3 Object Storage: Linux CLI Guide

Welcome! This guide shows you how to use **AWS CLI v2** and **s3cmd** with **eazyBackup e3 Object Storage**.

You can copy and paste the example commands as-is, replacing the placeholders where noted.

### What You Need

* Your e3 `ACCESS_KEY`
* Your e3 `SECRET_KEY`
* Your bucket name
* Linux shell access

Our S3-compatible endpoint used in examples:

```bash
s3.ca-central-1.eazybackup.com
```

#### Environment variables:

{% hint style="info" %}
Shell variables (like ACCESS\_KEY, SECRET\_KEY, BUCKET, and ENDPOINT) are just named placeholders in your terminal session. We use them in KB examples so you can set your e3 details once, then copy/paste the rest of the commands without editing every line.
{% endhint %}

Set the recommended environment variables in your terminal session:

```bash
export ACCESS_KEY="<YOUR_E3_ACCESS_KEY>"
export SECRET_KEY="<YOUR_E3_SECRET_KEY>"
export ENDPOINT="s3.ca-central-1.eazybackup.com"
export ENDPOINT_URL="https://$ENDPOINT"
export REGION="ca-central-1"
export BUCKET="<YOUR_BUCKET_NAME>"
```

### Option A: AWS CLI v2

AWS CLI is Amazon's official CLI, but it works great with e3 when you provide a custom endpoint.

#### A more permanent setup: use an `e3` profile

Most users prefer a named AWS profile so they do not have to export credentials every time.

1. Create an `e3` profile:

```bash
aws configure --profile e3
```

When prompted:

* AWS Access Key ID: your e3 access key
* AWS Secret Access Key: your e3 secret key
* Default region name: `ca-central-1`
* Default output format: (optional) `json`

2. Set e3-friendly S3 settings for that profile:

```bash
aws configure set region "ca-central-1" --profile e3
aws configure set s3.addressing_style virtual --profile e3
```

#### Where AWS CLI stores credentials (`~/.aws/credentials`)

AWS CLI profiles are stored in two files in your home directory:

* `~/.aws/credentials`: secrets (access keys)
* `~/.aws/config`: profile settings (region, S3 addressing style, etc.)

If you use `aws configure --profile e3`, AWS CLI will create/update these files for you.

You can also create them manually.

Create the directory:

```bash
mkdir -p ~/.aws
chmod 700 ~/.aws
```

Create or edit `~/.aws/credentials`:

```ini
[e3]
aws_access_key_id = YOUR_E3_ACCESS_KEY
aws_secret_access_key = YOUR_E3_SECRET_KEY
```

Lock down permissions:

```bash
chmod 600 ~/.aws/credentials
```

Create or edit `~/.aws/config`:

```ini
[profile e3]
region = ca-central-1
s3 =
  addressing_style = virtual
```

Then use the profile in commands:

```bash
aws --profile e3 --endpoint-url "https://s3.ca-central-1.eazybackup.com" s3 ls
```

Notes:

* You will still supply the e3 endpoint on commands using `--endpoint-url "$ENDPOINT_URL"` (We have found that AWS CLI does not reliably store an endpoint override in the profile for all commands).
* Keeping e3 in its own profile helps avoid accidentally talking to AWS S3 with the same commands.

#### Install AWS CLI v2 on Linux

```bash
curl -L --show-error \
  "https://awscli.amazonaws.com/awscli-exe-linux-x86_64.zip" -o /tmp/awscliv2.zip
unzip -oq /tmp/awscliv2.zip -d /tmp
sudo /tmp/aws/install --bin-dir /usr/local/bin --install-dir /usr/local/aws-cli --update
aws --version
```

#### Configure for e3

```bash
aws configure set default.region "$REGION"
aws configure set default.s3.addressing_style virtual
```

Why `virtual`:

* It uses bucket subdomain style (`bucket.endpoint`), which we recommend for e3.

#### Convenience: an `aws-e3` wrapper (optional)

If you want to avoid typing `--endpoint-url` and `--profile` every time, add this to your shell profile (for example `~/.bashrc`):

```bash
aws-e3() { aws --profile e3 --endpoint-url "https://s3.ca-central-1.eazybackup.com" "$@"; }
```

Then you can run:

```bash
aws-e3 s3 ls
aws-e3 s3 ls "s3://$BUCKET/"
```

#### AWS CLI Examples

List buckets:

```bash
aws --profile e3 --endpoint-url "$ENDPOINT_URL" s3 ls
```

List objects in a bucket:

```bash
aws --profile e3 --endpoint-url "$ENDPOINT_URL" s3 ls "s3://$BUCKET/"
```

Upload a file (PUT):

```bash
echo "hello from e3 via aws cli" > /tmp/e3-aws-test.txt
aws --profile e3 --endpoint-url "$ENDPOINT_URL" s3 cp /tmp/e3-aws-test.txt "s3://$BUCKET/e3-aws-test.txt"
```

Download a file (GET):

```bash
aws --profile e3 --endpoint-url "$ENDPOINT_URL" s3 cp "s3://$BUCKET/e3-aws-test.txt" /tmp/e3-aws-test.downloaded.txt
head -n 1 /tmp/e3-aws-test.downloaded.txt
```

Delete object:

```bash
aws --profile e3 --endpoint-url "$ENDPOINT_URL" s3 rm "s3://$BUCKET/e3-aws-test.txt"
```

Get object metadata:

```bash
aws --profile e3 --endpoint-url "$ENDPOINT_URL" s3api head-object \
  --bucket "$BUCKET" --key "e3-aws-test.txt"
```

### Option B: s3cmd

`s3cmd` is a popular third-party S3 client and works with e3 when configured correctly.

#### Install s3cmd on Linux

Ubuntu/Debian:

```bash
sudo apt-get update
sudo apt-get install -y s3cmd
s3cmd --version
```

#### Configure s3cmd for e3

Create a dedicated config:

```bash
cat > /tmp/e3-s3cmd.cfg <<EOF
[default]
access_key = $ACCESS_KEY
secret_key = $SECRET_KEY
host_base = $ENDPOINT
host_bucket = %(bucket)s.$ENDPOINT
use_https = True
signature_v2 = False
bucket_location =
verbosity = WARNING
EOF
chmod 600 /tmp/e3-s3cmd.cfg
```

{% hint style="info" %}
Important:&#x20;

* Set both `host_base` and `host_bucket`.&#x20;
* If `host_bucket` is missing, `s3cmd` will fall back to the AWS hostname like `bucket.s3.amazonaws.com`, which can cause failures even when keys are correct.
  {% endhint %}

#### s3cmd Examples

List buckets:

```bash
s3cmd -c /tmp/e3-s3cmd.cfg ls
```

List objects in a bucket:

```bash
s3cmd -c /tmp/e3-s3cmd.cfg ls "s3://$BUCKET/"
```

Upload a file (PUT):

```bash
echo "hello from e3 via s3cmd $(date -Is)" > /tmp/e3-s3cmd-test.txt
s3cmd -c /tmp/e3-s3cmd.cfg put /tmp/e3-s3cmd-test.txt "s3://$BUCKET/e3-s3cmd-test.txt"
```

Download a file (GET):

```bash
s3cmd -c /tmp/e3-s3cmd.cfg get "s3://$BUCKET/e3-s3cmd-test.txt" /tmp/e3-s3cmd-test.downloaded.txt
head -n 1 /tmp/e3-s3cmd-test.downloaded.txt
```

Delete object:

```bash
s3cmd -c /tmp/e3-s3cmd.cfg del "s3://$BUCKET/e3-s3cmd-test.txt"
```

Get object info:

```bash
s3cmd -c /tmp/e3-s3cmd.cfg info "s3://$BUCKET/e3-s3cmd-test.txt"
```

### Troubleshooting

#### `InvalidAccessKeyId` with s3cmd

Most common cause:

* `host_bucket` is not set correctly.

Fix:

* Use `host_bucket = %(bucket)s.s3.ca-central-1.eazybackup.com`.

### FAQ

#### Is e3 the same as AWS S3?

e3 is **S3-compatible**, not AWS itself.\
That means most standard S3 tools work, but you will need to provide the e3 endpoint and correct addressing configuration.

#### Which client should I use?

* We prefer **AWS CLI v2** for most workflows.
* Use **s3cmd** if you prefer its command style, but be sure to configure `host_bucket` explicitly.

***

{% hint style="info" %}
Need help? Share your exact command and error output with support, and include whether you are using AWS CLI or s3cmd.
{% endhint %}