# CloudRepo Edge storage

> Keep your Edge's copy of your files on the VM's own disk, or in an S3 bucket of your own with your own certificate authority.

> **Early access:** The CloudRepo Edge is in early access. We turn it on one organization at a time.

Your Edge keeps a copy of every file your builds pull, so the next pull comes from your own network. It keeps that copy in one place: the VM’s own disk, or an S3 bucket of your own. Pick one. The Edge will not start with both, or with neither.

Both live in `/etc/cloudrepo-edge/edge.env`, the settings file from step 3 of the install guide.

## On the VM’s own disk

This is what the install guide sets up:

`/etc/cloudrepo-edge/edge.env`

```ini
EDGE_HOST=your-org.mycloudrepo.io
EDGE_INGRESS=your-org.mycloudrepo.io:443
EDGE_API_URL=https://api.cloudrepo.io
EDGE_STORE_DIR=/var/lib/cloudrepo-edge/store
EDGE_SPOOL_DIR=/var/lib/cloudrepo-edge/spool
```

`EDGE_STORE_DIR` is where the copy lives. There is no size to set: the store uses the disk it is on. When a new file would leave less than 20% of that disk free, the Edge first removes the files your builds pulled longest ago, until 30% is free. It never writes past the last 5%. Past that, a pull still works: the Edge serves it from CloudRepo and does not keep it. To keep more, give the store a disk of its own.

`EDGE_SPOOL_DIR` holds each file while it arrives, before it joins the store. Keep it with either store.

## In an S3 bucket of your own

What the bucket needs:

- A bucket for this Edge alone, and a key for that bucket alone that can read, write and list its objects.
- Support for writes with `If-None-Match: *`, which the Edge uses to claim the bucket.
- Path-style addressing: your provider’s endpoint, then the bucket’s name as the whole path.

In `/etc/cloudrepo-edge/edge.env`, take out the `EDGE_STORE_DIR` line and put the bucket’s lines in its place:

`/etc/cloudrepo-edge/edge.env`

```ini
EDGE_HOST=your-org.mycloudrepo.io
EDGE_INGRESS=your-org.mycloudrepo.io:443
EDGE_API_URL=https://api.cloudrepo.io
EDGE_BUCKET_URL=https://s3.example.com/cloudrepo-edge
EDGE_BUCKET_REGION=us-east-1
EDGE_SPOOL_DIR=/var/lib/cloudrepo-edge/spool
```

`EDGE_BUCKET_URL` is your endpoint and the bucket’s name. It starts with `https://`, or `http://` for a bucket inside your own network. `EDGE_BUCKET_REGION` is the bucket’s region, `us-east-1` if you leave it out.

### The bucket’s key

Copy the key from your own secret store. First make its two files, open to the Edge’s user alone:

**Terminal**

```bash
sudo install -o 65532 -g 65532 -m 0600 /dev/null /etc/cloudrepo-edge/bucket/access-key-id
sudo install -o 65532 -g 65532 -m 0600 /dev/null /etc/cloudrepo-edge/bucket/secret-access-key
```

Put your access key id in the first file. Run this command on its own, paste the id, press Enter, then Ctrl-D:

**Terminal**

```bash
sudo tee /etc/cloudrepo-edge/bucket/access-key-id >/dev/null
```

Put your secret access key in the second file the same way:

**Terminal**

```bash
sudo tee /etc/cloudrepo-edge/bucket/secret-access-key >/dev/null
```

### Your own certificate authority

Skip this if a public CA signed your bucket’s certificate. If your own CA signed it, as it often does for storage inside your network, put that CA’s certificate (PEM, here `bucket-ca.pem`) beside the key:

**Terminal**

```bash
sudo install -o 65532 -g 65532 -m 0600 bucket-ca.pem /etc/cloudrepo-edge/bucket/ca.pem
```

Then add this line to `/etc/cloudrepo-edge/edge.env`:

`/etc/cloudrepo-edge/edge.env`

```ini
EDGE_BUCKET_CA_FILE=/var/run/cloudrepo-edge/bucket/ca.pem
```

That is where the Edge sees the key’s directory, so the file needs no mount of its own. The Edge then checks your bucket’s certificate against this CA alone, and uses it for nothing else. The file is mode 0600 and at most 64 KiB of PEM certificates, and the bucket’s URL must start with `https://`. If anything is wrong with it, the Edge stops before it sends anything, and its log names `EDGE_BUCKET_CA_FILE` and what is wrong.

### Register with the bucket

With a bucket, step 4 of the install guide registers the Edge with a different command. Use the one in the install guide’s “Keep the store in an S3 bucket” section, in place of step 4’s last command. It mounts the key’s directory, `/etc/cloudrepo-edge/bucket`, where the Edge reads it, and leaves out the store’s directory, which a bucket does not use. If a key file is missing or the bucket cannot be read, `register` stops and says what is wrong, and your join token is not spent.

### Check the bucket

The Edge’s health checks pass even when the bucket is out, because the Edge then serves every pull from CloudRepo and keeps nothing. So read the Edge’s log too: a line with `"level":"WARN"` that names the bucket says what is wrong.

---

The page: https://www.cloudrepo.io/docs/manage/edge-storage.html
