Open source alternative to Semrush and Ahrefs
TypeScript backlink-analysisgoogle-search-console-mcpkeyword-researchmcpseoseo-agentseo-auditseo-automationseo-skillsseo-toolssite-audit
Pin production to Postgres via committed Hyperdrive binding (#329)
Ben Senescu97f026562ec1
diff --git a/wrangler.jsonc b/wrangler.jsoncindex f6973f14..85f2862f 100644--- a/wrangler.jsonc+++ b/wrangler.jsonc@@ -64,14 +64,20 @@"migrations_dir": "drizzle",},],- // Postgres scale path (opt-in). D1 remains the default; leave this commented- // for free self-hosting. To run on Postgres set DATABASE_PROVIDER=postgres and- // uncomment a Hyperdrive binding pointing at your Postgres pooler. Without- // Hyperdrive, POSTGRES_DATABASE_URL is used as a direct-connection fallback- // (no edge pooling/caching — not recommended for the scaled hosted path).- // "hyperdrive": [- // { "binding": "HYPERDRIVE", "id": "<your-hyperdrive-config-id>" },- // ],+ // Postgres scale path. Production runs on Postgres: DATABASE_PROVIDER is set as+ // a worker *secret* (`wrangler secret put DATABASE_PROVIDER` -> "postgres") so it+ // survives `wrangler deploy` — which resets plain vars/bindings and is what+ // reverted prod to the D1 default before. This Hyperdrive binding points the+ // pooled connection at that Postgres. Without Hyperdrive, POSTGRES_DATABASE_URL+ // is the direct-connection fallback (no edge pooling/caching).+ //+ // SELF-HOSTERS on the free D1 default: DELETE this hyperdrive block before+ // deploying. The id below lives in OpenSEO's Cloudflare account, so `wrangler+ // deploy` will fail without access to it. D1 stays the default when+ // DATABASE_PROVIDER is unset.+ "hyperdrive": [+ { "binding": "HYPERDRIVE", "id": "9d64ccfb559f44449ce52a143912f898" },+ ],"r2_buckets": [{"bucket_name": "open-seo",