Move Docker containers — and their data — from one host to another, from a web UI, in a few clicks. It handles the whole container, not just the image: named volumes, anonymous volumes, bind mounts, user-defined networks, published ports, environment, capabilities, restart policy, healthchecks and resource limits. You pick what travels, per container and per mount. Two ways to move things: | Mode | What happens | When to use it | | --- | --- | --- | | **Host to host over SSH** | The source connects to the target over SSH and streams everything straight into the target's `docker cp` / `docker load`. Nothing touches disk in between. | The two hosts can reach each other. | | **Migration package** | Builds a self-contained folder or `.tar` holding the data, the images, and a plain-bash `install.sh`. Carry it on a disk, run the script on the target. | Air-gapped targets, or when you want the move reviewed and replayed later. | The target needs **nothing installed**: no agent, no Python, no Go. Just `sshd`, a working `docker` CLI, `bash` and `gzip`. --- ## Quick start ### Run it in a container (recommended) On the **source** host: ```bash git clone https://git.azuze.fr/kawa/DockMV.git dockmv && cd dockmv docker compose up -d --build docker compose logs dockmv # prints the URL, including the access token ``` Then open the printed URL. It binds to `127.0.0.1` only; reach it from your laptop with a tunnel: ```bash ssh -L 8080:127.0.0.1:8080 you@source-host ``` ### Run it bare-metal The binary is fully static and embeds the web UI, so there is nothing to install alongside it. ```bash make build # needs Go 1.25+ and Node 20+ ... or just `go build .` if you skip the UI rebuild ./dockmv serve ``` Prebuilt for several platforms: ```bash make release # dist/dockmv-linux-amd64, -linux-arm64, -darwin-arm64, -windows-amd64 ``` `dockmv` needs access to the Docker socket on the source host, so run it as a user in the `docker` group (or as root). --- ## Using it 1. **Containers tab** — everything on the source host, grouped by compose project. Tick the ones to move. Use *select all*, *select running*, or the compose-project checkbox for batch selection. 2. **Expand a row** (`▸`) to choose per-container details: the name on the target, whether the image is pulled or transferred, whether networks and ports come along, and — per mount — whether to **copy the data**, **create it empty**, or **not mount it at all**. Bind mounts can be relocated to a different path on the target; named volumes can be renamed. 3. **Apply to selected** in the toolbar does the same thing to every selected container at once (*copy all data*, *skip binds*, *image: pull on target*, …). 4. **Right panel** — add the target host, hit *connect*, review the options, then either **migrate over SSH** or **build a package**. 5. **Preview the commands** shows the exact `docker` invocations that will run on the target. Nothing is hidden. 6. **Jobs tab** — live progress per container and per mount, with the full command log. Start with **dry run** ticked. It performs every check and prints every command without changing anything on the target. --- ## How the data is actually moved The interesting part is that there is exactly **one** mechanism for every kind of data location: ``` source daemon ──CopyFromContainer(/mount/path)──▶ tar stream ──gzip──▶ ssh ──▶ docker cp -a - ctr:/parent ``` The source container's own mount path is read through the Docker archive API — the same thing `docker cp` uses. That means: - named volumes, anonymous volumes and bind mounts are all handled identically; - no helper image is pulled, and the container's image does not need `tar` inside it; - it works whether the container is running or stopped; - file ownership, permissions, symlinks and hardlinks are preserved (`docker cp -a`). On the target the container is **created first, started last**. Creating it is what makes Docker materialise the named volumes and bind directories; the data is then copied into the stopped container, and only then is it started. A mount the container declares **read-only** cannot be written through the container itself. For those, a throwaway container is created (never started) with the same volume attached writable, the data is copied into it, and it is removed straight after. ### What is faithfully reproduced Image (by pull or by layer transfer), command, entrypoint, environment, labels, working directory, user, hostname, published and exposed ports, all mount types, user-defined networks with their subnets and the container's aliases, DNS settings, extra hosts, capabilities, devices, sysctls, ulimits, security options, restart policy, stop signal and timeout, healthcheck, log driver and options, memory/CPU/pids limits, privileged, read-only rootfs, init, and the PID/IPC/UTS/userns modes. Settings that come from the **image** are deliberately not re-emitted — the recreated container carries only genuine run-time overrides, so it stays readable and keeps working when the image is later updated. ### What it will not do for you - **`--rm` is never reapplied.** A migrated container that deletes itself cannot be inspected. - **`--volumes-from` is not reproduced.** You are warned; migrate the other container and mount explicitly. - **`--network container:other`** requires the other container to be migrated too. You are warned. - **Swarm services** are out of scope. This tool moves plain containers. - **Live databases**: copying a running database's files gives you a crash-consistent snapshot at best. The default is to stop the source container while copying — leave it on. For anything you really care about, take a dump instead and migrate that. - **Cross-architecture moves**: an `amd64` image will not run on an `arm64` target. The preflight shows the target's architecture; check it. --- ## Safety The tool can stop containers and read every volume on the host, so it is treated as a privileged admin tool: - It binds to **`127.0.0.1` by default**. Binding anywhere else automatically generates an access token and prints it. - **SSH host keys are verified** exactly like OpenSSH. An unknown key is refused until you approve the fingerprint in the UI; a *changed* key is refused outright until you explicitly replace it. Trusted keys go to `/known_hosts`. - **Credentials are not persisted unless you ask.** By default the password or key lives in memory for the session. Ticking *remember* writes it to `/connections.json`, mode `0600`. - **Nothing on the target is overwritten by default.** If a container name already exists the item fails; you choose *skip*, *rename* or *replace* explicitly. An existing **volume** is reused and merged into, never silently deleted, unless you pick *replace*. - Every command that runs on the target is echoed into the job log. - Bind mounts of `/var/run/docker.sock`, `/proc`, `/sys`, `/dev` and `/` are flagged, and a mount at `/` is refused outright. --- ## The migration package `build package` produces: ``` shop-migration/ install.sh self-contained bash; read it, it is the whole contract manifest.json machine-readable description of everything inside README.txt instructions for whoever runs it images/ docker image archives (.tar.gz) data/ one archive per mount, per container ``` On the target: ```bash ./install.sh --dry-run # print every command, change nothing ./install.sh # restore ``` The installer never parses the manifest — every command is written out literally, so it can be read and audited before running. It checksums each payload before feeding it to Docker, and supports: ``` --dry-run print every command without changing anything --yes do not ask for confirmation --no-start create the containers but leave them stopped --conflict MODE fail (default) | skip | replace | rename --rename-suffix S suffix used by --conflict rename --skip-verify do not checksum the payloads --only NAME[,NAME...] restore only these containers --docker CMD docker command to use --sudo prefix docker with sudo -n ``` --- ## Target host requirements | Requirement | Why | | --- | --- | | `sshd`, reachable from the source | transport | | `docker` CLI + a working daemon | everything | | the login user can use docker | either in the `docker` group, or tick *run docker through `sudo -n`* | | `gzip` | compressed transfers; without it the tool falls back to uncompressed | | `bash` | only for the migration package installer | The **connect** button runs a preflight and tells you which of these are missing, plus the target's free disk space and architecture. --- ## Command line ``` dockmv [serve] [flags] start the web interface (default) dockmv inspect [flags] print the source inventory as JSON dockmv version ``` `serve` flags: ``` --addr string address to listen on (default "127.0.0.1:8080") --token string require this token on every request; "auto" generates one --data-dir string connections and trusted host keys (default: OS config dir) --package-dir string where migration packages are written (default /packages) --docker-host string source docker daemon (default: the DOCKER_HOST environment) -v verbose logging ``` `inspect` is handy for scripting and for reporting bugs: ```bash dockmv inspect --sizes | jq '.containers[] | {name, image, mounts}' ``` --- ## HTTP API Everything the UI does is available over HTTP. Pass the token as `X-Auth-Token` when one is set. ``` GET /api/health GET /api/source inventory + default selections GET /api/source/sizes volume sizes (slow) GET /api/connections POST /api/connections DELETE /api/connections/{id} POST /api/connections/{id}/probe read the SSH host key fingerprint POST /api/connections/{id}/trust approve that fingerprint POST /api/connections/{id}/test preflight the target GET /api/connections/{id}/inventory what is already on the target POST /api/plan/preview render the commands, run nothing POST /api/migrate/ssh start a host-to-host migration POST /api/migrate/package start a package build GET /api/jobs, /api/jobs/{id} GET /api/jobs/{id}/events server-sent events, live progress POST /api/jobs/{id}/cancel GET /api/packages, /api/packages/{name}/download ``` --- ## Development ``` web/ React + TypeScript UI (vite) internal/spec/ the transport model: a container, and how to render it back into docker flags internal/dkr/ source Docker daemon: inventory, archive streams, image save internal/sshx/ SSH transport, host key trust, driving the target's docker CLI internal/migrate/ the two engines: SSH streaming, and package + installer generation internal/job/ progress tracking for long-running work internal/api/ HTTP handlers and SSE internal/webui/ the built UI, embedded into the binary ``` ```bash make test # go test ./... make vet make ui # rebuild the embedded UI cd web && npm run dev # UI dev server on :5173, proxying /api to :8080 ``` ### Verifying it works On a Linux host with Docker (a VM is fine): ```bash go test ./... # unit tests: command rendering, plan resolution, # and the generated installer, checked with bash go test -tags e2e ./test/... -v # end to end, against the real daemon: # creates a container with a named volume, a # read-only bind mount and an anonymous volume, # writes files into all three, builds a package, # runs the generated install.sh, then reads the # files back out of the restored container ``` The e2e tests restore onto the same daemon under a suffixed name and clean up after themselves, so a single machine is enough: - `TestPackageRoundTrip` — builds a package and runs the generated `install.sh` for real. - `TestSSHMigration` — drives the host-to-host engine over a genuine SSH connection to `127.0.0.1`, so the whole transport (ssh, gzip streaming, the target's docker CLI, the staging container for read-only mounts, the verify step) is exercised. It needs `DM_SSH_HOST`, `DM_SSH_USER` and `DM_SSH_KEY`, and skips without them: ```bash ssh-keygen -t ed25519 -N '' -f ~/.ssh/dm_loop cat ~/.ssh/dm_loop.pub >> ~/.ssh/authorized_keys DM_SSH_HOST=127.0.0.1 DM_SSH_USER=root DM_SSH_KEY=~/.ssh/dm_loop \ go test -tags e2e ./test/... -run TestSSHMigration -v ``` Both were run against Debian 13 with Docker 29.7.2, alongside a browser pass over the web UI covering the trust prompt, a batch migration and a package build. `internal/webui/dist` is committed so that a plain `go build .` produces a working binary without a Node toolchain. Rerun `make ui` after changing anything under `web/`.