typed-secrets 4.4: persistence and resume hold references: rehydration stops reading the vault, every persisted copy uses the persisted form, and run events are masked

z1m ago

#3798

data_writer_property_test: a secret holding a newline and a literal backslash-n escapes the persisted-content property (seed 1648929202)

z1m ago

#3785

typed-secrets: the accessor adapter reads a legacy row stamped with the empty namespace as foreign, so its tagged fields stay reference text on the typed path

z2h ago

#3790

typed-secrets query view: route a row's JSON content by whether the adapter adapted the record, not by the service's namespace rule

z2h ago

#3789

flaky: for_each_name_property_test lets the text around a sentinel equal the secret's value, so the untyped value search replaces both and the property fails by seed

z2h ago

#3732

typed-secrets 4.3: the expression-facing query view registers Secret in DataQueryService

z3h ago

#3791

flaky: LitestreamReplicator 'a failed launch releases its lock probe, so the next acquire starts a sidecar' fails by timing in verify-build

s20m ago

#3714

CEL duration arithmetic is off by a second when a duration has negative nanos (cel-js 7.6.1)

s5h ago

#3731

typed-secrets 4.2: live and reconstructed step outputs and child outputs return references through the accessor adapter

z6h ago

#3773

A coded UserError a model method throws reaches a workflow step without its code, because InProcessExecutor keeps only the message

z8h ago

#3779

Dashboard Workflows page takes ~19s on a serve with many extension workflows: workflow.search resolves each workflow's owner with a full extension-workflow scan

z8h ago

#3670

codegen: BROKEN_REGEXES override for the @swamp/aws/invoicing InvoiceUnit name lacks the u flag, so it rejects ordinary names

s22h ago

#3742

dashboard: run page polling stacks history.get and history.logs requests while serve is slow

z1d ago

#3741

serve: add child spans and attributes to workflow.history.get (and run_summary/run_search) to diagnose slow dashboard loads

z1d ago

#3740

serve: webhook-triggered workflow runs can't be cancelled — cancel can't find them, so it reports serve's own pid as "the process running it is still alive"

z1d ago

#3726

A workflow step fails on a static validation error in a method it does not invoke, where swamp behavior run of the same method succeeds

z1d ago

#3739

serve: a run orphaned by a container restart can't be cancelled — the restarted serve reuses the dead owner's pid, so cancel reports the owner still alive

z9h ago

#3735

typed-secrets 4.6: turn the typed path on for every entry point, remove the internal option, and delete the regular-expression rewrite, the value search and the plaintext resolution of tagged fields

1d ago

#3734

typed-secrets 4.5: the inline resolutions refuse or mask: placement, task selectors, child references and deferred targets refuse secrets, assert messages mask them, and the deprecated resolvers go

1d ago

#3730

typed-secrets 4.1: workflow passes bind sentinels as Secrets: typed guards and asserts, forEach items and identities, child bindings, webhook inputs, and the synchronous authorization step

z9h ago

#3628

typed-secrets 3.5: local entry points, validation diagnostics, context.runModel and model.method delivery

z1d ago

#3708

model.method in a workflow guard fails with 'Do not know how to serialize a BigInt' when its inputs hold an int

z1d ago

#3706

data query: filter on data lifecycle so deleted resources can be excluded

z1d ago

#3713

CEL expression results that hold a timestamp, duration or uint become an empty object

z1d ago

#3707

Catalog rebuild restores pre-rename versions of a name renamed away and written again

z1d ago

#3529

Remote resolveSecret allowlist is always empty for workflow steps that use vault.get

z1d ago

#3686

swamp issue redacts git commit SHAs as secrets and fast-check replay paths as IP addresses

s1d ago

#3627

typed-secrets 3.4: destination refusals, identifier stand-ins, shell safety and masking

z2d ago

#3530

Remote readResource does not apply the legacy schema fallback for untagged sensitive records

z2d ago

#3685

Flaky: datastore_v3_outbox_model_property_test throws MayLandLaterError on a random seed (aged commit, failed discard-and-mark checkpoint, power cut)

z2d ago

#3645

v3 outbox model property test fails: a commit dropped after a failed append-discard-and-mark reports MayLandLaterError instead of the expected outcome

2d ago

#3606

Workflow step dataOutputOverrides vaultName is accepted but silently dropped

s2d ago

#3626

typed-secrets 3.3: sensitive-field accessors, data-origin sentinels and reference validation

z2d ago

#3641

Generated vault keys carry raw data-name and vary characters that strict vault providers reject

3d ago

#3625

typed-secrets 3.2: vault.get as a CEL function, bare-token rewriting, AST extraction and up-front existence checks

z3d ago

#3619

typed-secrets 3.1: the production Secret, its refusals, the fault context, the container rule and the boundary conversion, behind the internal option

z3d ago

#3620

Flaky test: extension_pull_dependency_conflict_test fails with 'failed to copy esbuild binary' during deno bundle

3d ago

#3609

Extension tarball analyzer ignores webhook entry points

3d ago

#3608

Registry drops contentMetadata.extensions (method grafts) at confirm-push

3d ago

#3041

digitalocean codegen leaves an orphaned security_secret.ts model file after its endpoint left the spec

s6d ago

#3552

A --repo-dir or SWAMP_REPO_DIR through a symlink fails with 'unable to open database file'

s6d ago

#3543

@swamp/1password: vault delete of a missing item probably fails instead of being a no-op

6d ago

#3544

@swamp/1password: item or vault names containing 'authorization' are misreported as auth failures

s6d ago

#3558

typed-secrets 2.2: VaultSecretBag entries hold references, one registry per run, 128-bit sentinels

z6d ago

#3214

Show wait keys and pending signals to remote clients and the dashboard

z6d ago

#3560

Flaky: LitestreamReplicator.acquire 'concurrent acquires in one process share one sidecar and one hold' fails 'no hold leaked' under full-suite load

z6d ago

#3563

VaultService.requireExistenceProvider labels the vault instance name as its type

z6d ago

#3561

VaultService capability errors label the vault instance name as its type

z6d ago

#3557

typed-secrets 2.1: VaultExistenceProvider, its conformance suite, and has on the built-in providers

z6d ago

#3559

MockVaultProvider.get and delete throw synchronously instead of rejecting

z6d ago

← Back to list

01Issue

BugClosedSwamp CLIPublic

Assigneeszack

Relationships

#3707 Catalog rebuild restores pre-rename versions of a name renamed away and written again

Opened by zack · 10/21/2026

Summary

A catalog rebuilt from disk (full backfill) and the write-through catalog disagree about the history of a data name that was renamed away and then written again. The rebuild brings back the versions from before the rename; write-through had removed them.

Steps to reproduce

With FileSystemUnifiedDataRepository and a populated write-through CatalogStore:

save data name a (v1).
rename a to b — write-through removes a's rows, writes b@1, records the forward a -> b, and a@2 on disk is the rename marker.
save a again (v3) — write-through upserts a@3 and ends the forward.
Rebuild a fresh catalog from the same files (DataQueryService.query on an unpopulated catalog).

Actual

write-through rows: a@3 (latest), b@1 (latest)
rebuilt rows: a@1, a@3 (latest), b@1 (latest)

The backfill indexes every version of a from listVersions, skipping only the rename marker version, so the pre-rename a@1 returns as a non-latest history row.

Expected

A rebuilt catalog matches the write-through catalog. Either the backfill drops versions below the most recent rename marker of a name, or write-through keeps them; pick one and make both paths agree.

Impact

Latest rows match, so data.latest(), default queries and the swamp doctor datastores completeness check are unaffected. History queries (isLatest == false, version == 1) return a pre-rename version only on rebuilt catalogs.

Found while writing the rebuild-equals-write-through property test for swamp-club#3698; that test's generator excludes re-saving a renamed-away name because of this.

02Bog Flow

Closed

10/21/2026, 2:42:16 PM

No activity in this phase yet.

03Sludge Pulse

zack assigned zack10/21/2026, 2:27:50 PM

zack commented 10/21/2026, 2:42:16 PM

Closing. Triage found that this affects only v2 datastores.

Reproduced on a v2 filesystem datastore with the CLI. Write data name a, run swamp data rename to move it to b, write a again, then delete .swamp/data/_catalog.db and query. The rebuilt catalog has an extra non-latest a@1 row that the write-through catalog never had. Latest rows match on both sides.

The extra row comes from the v2 full backfill (DataQueryService.backfillAsync and backfillSync), which indexes every version on disk except the marker versions. v3 datastores never run that backfill: repository_factory.ts creates the query service with backfill turned off when the v3 read model supplies the catalog. A v3 catalog is rebuilt by replaying commits or loading the compacted catalog state. Those paths drop the old name's rows on a rename and add back only the version written afterwards, which is what write-through does. The v3 conclusion comes from reading the code; it was not reproduced on a v3 datastore.