Subversion repository recovery

An SVN repository can lose one revision file or revprop while most earlier history and working copies remain usable.

Subversion repositories contain revision data, revision properties, transaction state, configuration, hooks, locks, and backend-specific metadata. FSFS stores history as files, while older repositories may use Berkeley DB, so the correct recovery path depends on the repository format and exact failed revision.

Understand the symptom

svnadmin verify identifies the first unreadable revision but does not repair missing history.

WHY

What is happening

The Subversion book explains that verify walks repository revisions and stops at corruption, while svnadmin recover applies specifically to Berkeley DB consistency. FSFS recovery often depends on backups, dumps, mirrors, working copies, and intact revision ranges.

What you may see

Symptoms that change the recovery path.

01

svnadmin verify stops at one revision

A revision or representation cannot be read or checksummed.

02

No such revision file

An FSFS rev or revprop file is missing from its shard.

03

Repository needs recovery

A Berkeley DB backend may be inconsistent or locked.

04

Checkout works only to an older revision

Later history may depend on the damaged range.

Protect the source

Actions that can make the situation worse.

01

Live file replacement can create false history

Revision numbers alone do not prove content or checksum identity.

02

Dumping may stop at the first bad revision

Do not assume later revisions are absent or useless.

03

Working copies contain uncommitted data

Cleanup or fresh checkout can discard local changes.

Do this now

Safe steps before evaluation.

01

Take the repository offline

Stop commits, hooks, replication, packing, upgrades, and maintenance.

02

Copy the full repository

Keep format, UUID, revs, revprops, locks, transactions, hooks, and permissions.

03

Run verification on a copy

Record the first failure, backend, shard layout, and command version.

04

Collect alternate history

Preserve hotcopies, dump files, mirrors, svnsync targets, and working copies.

NEPFOSYS process

How a data-first evaluation proceeds.

01

Repository map

NEPFOSYS identifies backend, format, UUID, revision ranges, shards, revprops, and readable history.

02

History reconstruction

Intact dumps, mirrors, backups, revision files, and working-copy content are compared on protected copies.

03

Repository validation

Revisions, paths, properties, branches, tags, checksums, and checkout behavior are verified.

A clear next step

Preserve the original device before another repair or recovery attempt.

Start a case with the exact symptoms, incident timeline, prior attempts, encryption details, and the files that matter most. NEPFOSYS will route the case to the appropriate DFW intake or nationwide mail-in path.

Start with a case review

Questions, answered

What customers ask before starting.

Will svnadmin recover fix FSFS corruption?

No. That command is for Berkeley DB consistency, not missing FSFS revision content.

Can working copies restore repository history?

They can preserve file states and local changes, but not all revision metadata or every historical version.

Should the repository be upgraded during recovery?

No. Preserve and recover the original format before migration.

DFW local · Nationwide mail-in

Protect the device. Start with the facts.

Start my recovery case