docs: update README for unmanaged data volume, SFTP, cached AMI

Storage architecture changed 2026-06-05 (volume imported by ID, not
CFN-managed) — the old Expanding Storage procedure no longer worked.
This commit is contained in:
Adam Moussa 2026-06-05 15:50:05 -04:00
parent 2c45931741
commit cf9b233aee

View file

@ -4,13 +4,16 @@ Personal file share server on AWS — Samba for macOS Finder integration and Fil
## Architecture
- **EC2** — `t4g.small` (ARM64, Amazon Linux 2023) in the private subnet
- **EC2** — `t4g.small` (ARM64, Amazon Linux 2023) in the private subnet. The AMI is cached in the committed `cdk.context.json` (`cachedInContext: true`), so deploys never pick up a new AL2023 release implicitly — an AMI change forces instance replacement and must be deliberate (`cdk context --reset <ami key> && cdk synth`).
- **Samba** — SMB file share at `/data/share`, optimized for macOS (`vfs_fruit`)
- **FileBrowser** — Web UI on port 8080, backed by the same `/data/share` directory
- **EBS** — 500 GiB gp3 data volume (separate from root), encrypted
- **DLM** — Daily EBS snapshots, 30-day retention
- **SFTP** — password auth for user `adam` (`ForceCommand internal-sftp`, same password as SMB); used by Hazel for automated uploads
- **EBS data volume** — `vol-04d951cccacc435b5`, 500 GiB gp3 encrypted, mounted at `/data`. **Unmanaged import**: the stack references it by ID (`Volume.fromVolumeAttributes` + `CfnVolumeAttachment`), so CloudFormation can attach it but can never create, replace, or delete it — the data survives instance replacement and even stack deletion. UserData waits for the attachment, then mounts the existing filesystem; a `blkid` guard ensures a disk that already has a filesystem is never formatted.
- **DLM** — Daily EBS snapshots at 06:00 UTC, 30-day retention (targets the `file-share-backup=true` tag, set directly on the volume)
- **SSM** — Session Manager for instance access (no SSH key)
> **History:** the data volume was originally an inline `blockDevice`, which destroyed data on instance replacement (2026-05-27 incident), then a stack-managed standalone volume, which was orphaned when an uncommitted deploy got reverted by CD (2026-06-05 incident). The unmanaged-import design ends that failure class.
## Access
Requires VPN connection to the office network (10.10.0.0/16).
@ -52,12 +55,11 @@ The stack outputs the instance's private IP for SMB and FileBrowser access.
## Expanding Storage
The 500 GiB data volume can be expanded without downtime:
The data volume is **not managed by CloudFormation** (imported by ID), so changing a size in `lib/file-share-stack.ts` has no effect. Expand it directly — no downtime:
1. Modify the volume size in `lib/file-share-stack.ts`
2. Deploy: `npx cdk deploy`
3. SSH into the instance via SSM and resize the filesystem:
1. `aws ec2 modify-volume --volume-id vol-04d951cccacc435b5 --size <new-GiB>`
2. Wait for the modification to leave `modifying`: `aws ec2 describe-volumes-modifications --volume-ids vol-04d951cccacc435b5`
3. Resize the filesystem via an SSM session:
```bash
sudo growpart /dev/xvdf 1 # if partitioned
sudo resize2fs /dev/xvdf
sudo resize2fs /dev/nvme1n1 # xvdf surfaces as nvme1n1 on Nitro instances
```