Skip to content

docs: state that live restore is incompatible with Swarm mode - #26505

Open
Duo-ee wants to merge 2 commits into
docker:mainfrom
Duo-ee:26424-live-restore-swarm
Open

Duo-ee wants to merge 2 commits into
docker:mainfrom
Duo-ee:26424-live-restore-swarm

Conversation

@Duo-ee

@Duo-ee Duo-ee commented Oct 31, 2026 •

Copy link
Copy Markdown

Description

The "Live restore and Swarm mode" section only described which containers the
option applies to. That reads as if live restore and Swarm mode can be enabled
together on the same daemon, with live restore handling standalone containers
and Swarm handling the services. They can't: the daemon refuses to start when
live-restore is enabled on a node that is part of a Swarm.

This adds a short paragraph to that section which:

  • states that live restore and Swarm mode are mutually exclusive,
  • shows the error the daemon logs when both are enabled, and
  • explains how to get the daemon running again.

Change is limited to content/manuals/engine/daemon/live-restore.md. I didn't
touch the existing wording above the new paragraph.

A follow-up commit corrects the recovery step. Removing the node from the Swarm
doesn't clear the node's local Swarm state, and the daemon reads that state
before it connects to the cluster, so the check keeps failing. docker swarm leave also needs a running daemon. Disabling live restore is what gets the
daemon back; leaving the Swarm comes after that.

Verification

  • npx --no-install rumdl fmt and rumdl check: no issues.
  • vale on the file: the added lines report nothing. The file also reports one
    existing error and one suggestion at lines 29 and 30 (Docker Desktop and
    click); those are unchanged by this PR and were already reported on
    upstream/main.
  • I couldn't run docker buildx bake validate|test locally (no Docker on this
    machine), so the Hugo build layers are unverified until CI runs.

Related issues or tickets

Closes #26424

Reviews

  • Technical review
  • Editorial review
  • Product review

Generated by DeepSeek Harness, a coding agent. It drafted the change and this
description.

The live restore section only described which containers the option applies to, which reads as if live restore and Swarm mode can be enabled together. The daemon refuses to start when both are enabled.
@Duo-ee
Duo-ee requested a review from dvdksn as a code owner October 31, 2026 09:36
@netlify

netlify Bot commented Oct 31, 2026 •

Copy link
Copy Markdown

✅ Deploy Preview for docsdocker ready!

Name Link
🔨 Latest commit 052e153
🔍 Latest deploy log https://app.netlify.com/projects/docsdocker/deploys/6ae5fdf5b25d45bfbaf08a6b
😎 Deploy Preview https://deploy-preview-26505--docsdocker.netlify.app
📱 Preview on mobile
Toggle QR Code...

QR Code

Use your smartphone camera to open QR code link.

To edit notification comments on pull requests, go to your Netlify project configuration.

@github-actions github-actions Bot added the area/engine Issue affects Docker engine/daemon label Oct 31, 2026
Removing a node from the Swarm doesn't clear the node's local Swarm state, so
the daemon still fails the compatibility check when it restarts. Pointing
readers at that option sent them into a loop: docker swarm leave needs a running
daemon, and the daemon doesn't start while live restore is enabled.

Disabling live restore is the step that unblocks the daemon. Mention leaving the
Swarm only after the daemon is back.

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

area/engine Issue affects Docker engine/daemon

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Bad wording/missing information for live restore and swarm mode

1 participant