title: Move a volume description: Move a volume and the service that mounts it to another server.

A volume lives on one server. With Ployz Cloud you can move it, and the service that mounts it, to another server. Your service stops only for the final copy.

Every server with storage needs a newer Ployz before any volume moves. Until then the volume's Copies names the servers to upgrade, for example "Upgrade db-1 to a newer Ployz".

Move a volume to another server

A volume moves onto its mirror, so first add a mirror on the server you want it on.

1. Open the volume and click its mirror under Elsewhere in Copies.
2. Click Make primary. The mirror shows Primary in pink until you deploy.
3. Click Deploy.

A mirror on a server running 0.3.0-beta.6 offers no Make primary and reads "Make primary needs a newer Ployz on web-2.": that release can take the volume but can't put the moved service back in web traffic. Upgrade the server first.

Details lists the steps: "Stop · Final sync · Start", with "Catch up" first when the mirror last refreshed over an hour ago. Your service keeps running while the deploy catches up. Writes pause from Stop until Start, so the final sync carries every last change. If the mirror can't catch up, the deploy fails and the old server keeps serving.

The deploy moves the volume before it starts your services. While it runs, the volume's tray, the bottom bar and the deployment page say what it's doing, for example "Moving to web-2". A service that takes a while to stop or start doesn't fail the move.

Cancel lets the step under way finish, and the rest waits for your next deploy. If the deploy fails, the change that failed is dropped, and the changes it never reached wait for your next deploy. A move that already handed the volume to its new server isn't dropped, since only finishing it brings the volume back. See A move failed partway.

The old copy a move leaves

The old server keeps a read-only copy, like data-web-1, under Elsewhere. It serves nothing and never refreshes. Moving back sends only what changed since. If an earlier move onto that server failed, Ployz may send the whole volume instead.

• Its × deletes it right away, after you click Delete. This can't be undone.
• Its Restore stages a restore from its last point. See Restore a volume. While a staged restore reads the copy, × says "A restore reads this copy."

A move or a new mirror deletes old copies on other servers first, and Details says so.

Drain a server

Drain on a server's page moves each volume there off at the next deploy, to its mirror or else to the first other server that runs services. The page says how many, like "Next deploy moves 2 volumes". Each environment's next deploy moves its own volumes. Drain's moves wait for every server with storage to have a newer Ployz too. Drain skips a server running 0.3.0-beta.6. When a volume then has nowhere to go but that server, Drain names it: "Upgrade web-2 to move data there". A volume nowhere would take reads "data has nowhere to move".

Move from the CLI

The CLI moves a volume at once, without waiting for a deploy:

--wait prints each step as the move reaches it, and so does ployz volume runs data RUN:

The last line says how long your service was down. ployz volume runs data keeps it as the move's message. While the old replica is stopped, ployz status names it as not serving, with the code it exited with. The row goes once the replica on web-2 runs.

The old server keeps an old copy, data-web-1. If the volume already has a mirror on web-2, the move starts from it. To move to a third server, first remove the old copy with the ployz volume mirror rm line the move prints.

ployz volume move, release, restore, mirror, mirror rm and sync --full each start a run in Ployz Cloud and print its id. Add --wait to wait for it to finish. A volume takes one run at a time, and none while a deploy runs in its environment. ployz volume runs data lists what ran on the volume and how it ended. ployz volume release data stops a move or a mirror that's still running.

A deploy never waits for a move. Until web-2 is told to start your service, a deploy takes the move over. Your service deploys on the old server, and Ployz asks for the move again once the deploy ends. After that, a deploy follows the move and starts your service on web-2.

In a Protected environment, a move, restore, release, mirror removal or resize from a member, or from a token or an agent without environments:protect, waits for an owner or admin to approve it. ployz waits for the approval. Adding a mirror and syncing don't ask.

If a move stops partway or a command is refused, see Troubleshooting volumes.

Good to know

• A server must answer for its volume to move. If the volume's server is down, neither a move nor a release can run, and the run says which server did not answer. Copies offers no move while it's silent, since it may still be writing. See When its server goes down.