Commit Graph
6 Commits
Author SHA1 Message Date
Evan JarrettandClaude Opus 5 a95c89aaef docs: record how a hold falls off the relay and stays off
Around 2026-08-20 the Bluesky relay marked every ATCR hold offline and stopped
dialing. Nobody noticed for two weeks, and it surfaced only indirectly as "pull
and push counts are up to 24 h stale". The stats were a symptom; the fleet was
simply disconnected.

It cannot recover on its own. Indigo's relay gives up on a host after 16
consecutive dial failures and returns from the redialer, and only a fresh
requestCrawl revives it. The hold sends requestCrawl exactly once, at boot
(server.go:400), with no ticker and no check that any relay is subscribed. So a
dropped hold is invisible until its process restarts, silently.

Documents the current mechanics, the failure mode, how to diagnose it with
getHostStatus and a frozen repo rev, and how to recover. The automatic
re-crawl is described as a deferred proposal and explicitly NOT implemented,
by decision: a jittered ticker guarded on subscriber liveness, plus surfacing
the subscriber count, since the deeper problem is that this was silent.

Two things found while writing it, both recorded. The proposal needs plumbing
that does not exist: EventBroadcaster has no exported subscriber count, and
Subscriber does not retain the userAgent, so "is a relay listening" cannot
currently be answered. And ResubscribeAllHosts selects only active hosts, so an
offline host is not recovered even by a relay restart.

Carries a replay warning. ca539b1 fixed a panic on subscriber disconnect during
firehose backfill, and the exposure condition is that a backfill goroutine
exists at all, which Subscribe skips when the cursor is current. So a caught-up
relay never triggered it and a hold whose relays are far behind is exposed on
every reconnect. Verify a deployed hold contains ca539b1 before provoking a
re-crawl; efabb677 does not.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PDqoCE1j3njokkZ9b1C5n9
2026-09-02 22:30:37 -05:00
Evan Jarrett 6758996300 add SBOM package diffing, verify hold-service captain records
- diff view gains a Packages tab with added/removed/changed/unchanged
  package tables and purl-derived type/license/upstream links
- captain records verified against the DID's atcr_hold service before
  caching (processor + batch backfill), preventing forged holds
- fix empty-handle updates clobbering cached handles and colliding on
  the UNIQUE constraint
- move fillPrevCIDs into repo.go; DirectRepoOperator is now canonical,
  repomgr kept as a test oracle
- surface read-only crew status in hold selector
- reconcile docs
2026-06-13 12:49:03 -05:00
Evan Jarrett e3843db9d8 Implement did:plc support for holds with the ability to import/export CARs.
did:plc Identity Support (pkg/hold/pds/did.go, pkg/hold/config.go, pkg/hold/server.go)

  The big feature — holds can now use did:plc identities instead of only did:web. This adds:
  - LoadOrCreateDID() — resolves hold DID by priority: config DID > did.txt on disk > create new
  - CreatePLCIdentity() — builds a genesis operation, signs with rotation key, submits to PLC directory
  - EnsurePLCCurrent() — on boot, compares local signing key + URL against PLC directory and auto-updates if they've drifted (requires rotation key)
  - New config fields: did_method (web/plc), did, plc_directory_url, rotation_key_path
  - GenerateDIDDocument() now uses the stored DID instead of always deriving did:web from URL
  - NewHoldServer wired up to call LoadOrCreateDID instead of GenerateDIDFromURL

  CAR Export/Import (pkg/hold/pds/export.go, pkg/hold/pds/import.go, cmd/hold/repo.go)

  New CLI subcommands for repo backup/restore:
  - atcr-hold repo export — streams the hold's repo as a CAR file to stdout
  - atcr-hold repo import <file>... — reads CAR files, upserts all records in a single atomic commit. Uses a bulkImportRecords method that opens a delta session, checks each record for
  create vs update, commits once, and fires repo events.
  - openHoldPDS() helper to spin up a HoldPDS from config for offline CLI operations

  Admin UI Fixes (pkg/hold/admin/)

  - Logout changed from GET to POST — nav template now uses a <form method=POST> instead of an <a> link (prevents CSRF on logout)
  - Removed return_to parameter from login flow — simplified redirect logic, auth middleware now redirects to /admin/auth/login without query params

  Config/Deploy

  - config-hold.example.yaml and deploy/upcloud/configs/hold.yaml.tmpl updated with the four new did:plc config fields
  - go.mod / go.sum — added github.com/did-method-plc/go-didplc dependency
2026-02-14 15:17:53 -06:00
Evan Jarrett 4ac2b97c33 remote at sign from tangled urls 2026-02-10 20:48:24 -06:00
Evan Jarrett 57593a8683 remove the filesystem and buffered upload ability on the holds. going forward the only supported storage is s3. adds extra mocks and tests around uploading 2026-01-19 16:59:03 -06:00
Evan Jarrett ec90f43d3e add atcr-hold readme 2025-10-24 09:54:10 -05:00