fix(gateway): local infra defaults + qdrant collection retry-on-failure

- AI_LLM_BASE_URL default -> http://127.0.0.1:4014/v1 (was imrnes :20128/api/v1)
- QDRANT_URL fallback -> http://127.0.0.1:6333 (was imrnes :6333)
- ensureQdrantCollection: reset memoised promise on failure so a mid-way
  recreate abort (DELETE done, PUT failed) does not leave the collection
  permanently missing until process restart
- tests: qdrantEnsure.test.ts (3 cases: retry-on-failure, recreate, idempotent)
This commit is contained in:
asepharyana committed 2026-09-02 12:57:54 +07:00
1 parent e5304fde29
commit 4c38d53972
3 files changed
+89 -8

No files matched your search

@@ -151,10 +151,7 @@ export const configSchema = z
.transform((v) => v === "true")
.default(false),
AI_LLM_API_KEY: z.string().optional(),
AI_LLM_BASE_URL: z
.string()
.url()
.default("http://100.121.180.82:20128/api/v1"),
AI_LLM_BASE_URL: z.string().url().default("http://127.0.0.1:4014/v1"),
AI_LLM_MODEL: z.string().default("text"),
// Vision uses the SAME router/base URL as text moderation
// (AI_LLM_BASE_URL) but a different model alias. The dedicated NVIDIA