[diegosouzapw]

Refs #11500 (Issue 1 only — the plan-file's Issues 2/3 need a separate scoped design decision and are explicitly out of this PR's scope; Issue 4 was already fixed, no code change needed).

Root cause

The #9927 fix deduped raw [Encryption] Decryption failed... logging only inside decryptConnectionFields(). The lazy-decrypt rollout (createLazyRowProxy/createLazyConnectionView in src/lib/db/providers/lazyConnectionView.ts, used by getProviderConnections() on every CredentialHealth/model-sync cycle) calls decrypt() directly with no {quiet:true} and no dedup tracking — a fresh Proxy wraps a fresh row object every cycle, so the raw failure line re-fires on every single cycle for the same corrupt/stale-key credential.

Fix

New decryptQuiet() wrapper in src/lib/db/encryption.ts shares the existing loggedDecryptFailures Set with decryptConnectionFields(), keyed by (provider, connectionId, field, ciphertext) so a credential already flagged via either path logs at most once. Wired into both createLazyRowProxy and createLazyConnectionView.

Regression test

tests/unit/lazy-connection-view-decrypt-dedupe.test.ts — simulates 3 separate sync cycles over the same corrupt credential via both lazy-decrypt entry points; asserts the raw failure line logs exactly once. Passing 2/2.

Gates run

• typecheck:core — clean, exit 0
• eslint --suppressions-location (changed files) — exit 0
• check-file-size.mjs — OK
• check-complexity.mjs — OK (2684 vs baseline 2774)
• check-cognitive-complexity.mjs — OK (1208 vs baseline 1223)
• check-changelog-integrity.mjs — OK

⚠️ base-red inherited: #11449 — ESLint errors (5), docs-sync migrations count (unrelated to this diff)

[diegosouzapw]

Validated in local merge-train (train tip a3a96af7308584900f0fab3ab4afd8499391230a, boarded #11663 #11664 #11665 #11667 #11668 #11669 #11670 #11678 on .113 /srv/omniroute-train) — typecheck:core, file-size, complexity, cognitive-complexity, changelog-integrity all GREEN. npm run test:unit showed 21 failures across 15 test files, none touching this PR's diff; reproduced identically (same test names, same asserts) on origin/release/v3.8.51 alone with zero PRs boarded (targeted re-run, 21/94 fail, same failure set) — confirmed base-red/environment-load drift per merge-gates.md §3, not introduced by any of the 8 boarded PRs.