Skip to content

HDDS-16147. Document OM metadata backup including bucket snapshots - #526

Draft
jojochuang wants to merge 3 commits into
apache:masterfrom
jojochuang:docs-om-metadata-backup
Draft

HDDS-16147. Document OM metadata backup including bucket snapshots#526
jojochuang wants to merge 3 commits into
apache:masterfrom
jojochuang:docs-om-metadata-backup

Conversation

@jojochuang

@jojochuang jojochuang commented Aug 11, 2026

Copy link
Copy Markdown
Contributor

Summary

HDDS-16147: https://issues.apache.org/jira/browse/HDDS-16147

  • Add Backup and recovery under Administrator Guide → Operations with an OM metadata backup runbook (/v2/dbCheckpoint with includeSnapshotData=true, Ozone 2.2+).
  • Include om-metadata-backup.sh for multi-batch checkpoint downloads (same logic as OM HA bootstrap).
  • Document prescriptive cold restore steps from an offline tarball when no live OM leader is available, and clarify that legacy v1 /dbCheckpoint omits bucket snapshot data.
  • Cross-link from Replacing Ozone Manager disks standalone recovery path.

What is the link to the Apache Jira?

https://issues.apache.org/jira/browse/HDDS-16147

How was this patch tested?

  • pnpm run build — passes (no broken links)
  • Markdown lint / cspell on changed docs

Test plan

  • Review backup procedure for Ozone 2.2+ v2 endpoint accuracy
  • Review cold-restore hardlink reconstruction steps
  • Confirm sidebar placement under Operations → Backup and recovery

Add Administrator Guide backup-and-recovery section with v2 checkpoint backup steps, multi-batch helper script, and prescriptive cold restore from offline tarballs; cross-link from OM disk replacement.

Co-authored-by: Cursor <cursoragent@cursor.com>
@jojochuang
jojochuang requested a review from sadanand48 August 11, 2026 16:06
@jojochuang jojochuang changed the title Document OM metadata backup including bucket snapshots (draft — Jira pending) HDDS-16147. Document OM metadata backup including bucket snapshots Aug 11, 2026

usage() {
cat <<'EOF'
Usage: om-metadata-backup.sh [options] --base-url URL

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I feel we could add a java based tool itself instead of adding bash code. I don't see the hardlink construction logic here and we could re-use the existing code that follower uses to construct the db from the tar

jojochuang and others added 2 commits August 13, 2026 15:52
Replace manual live-OM restore steps with the HDDS-16171 repair CLI and add download subcommand reference in the Ozone Repair guide.

Co-authored-by: Cursor <cursoragent@cursor.com>
…load.

Lead with the repair CLI for live-OM backup and restore, keep HTTP/curl as
an alternative, and align command examples with the HDDS-16171 tool behavior.

Co-authored-by: Cursor <cursoragent@cursor.com>

## Alternative: HTTP checkpoint download

Use `curl` or `om-metadata-backup.sh` only when the backup host has no Ozone CLI. This runs the same `/v2/dbCheckpoint` transfer, but the saved tarball stays in inode-based form and needs hard link reconstruction before install.

@sadanand48 sadanand48 Aug 25, 2026

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

problem with direct curl for V2 is that it doesn't construct the paths from hardlink file. We can document this limitation if we are going with the alternative , while the backup is available it needs the unpack logic from follower

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants