▸case-03 Our Next.js application on Vercel experiences high connection spikes on Neon Postgres during traffic peaks. We are configuring Prisma Client instantiation. Should Prisma Client connect directly to the Postgres port or through the PgBouncer pooler endpoint? | pass→pass | 13,266 | 10,675 | -20% | 1 | 1 | 0% | 2,317 | 2,178 | -6% | 0 | 0 | — |
▸case-16 Our load test generates 12,000 simultaneous concurrent incoming connections directly against the Neon PgBouncer pooled connection hostname. Will the pooler accept all 12,000 active concurrent connections without hitting pooler capacity limits? | pass→pass | 10,156 | 5,401 | -47% | 1 | 1 | 0% | 1,878 | 1,099 | -41% | 0 | 0 | — |
▸case-02 When running `prisma migrate deploy` against our Neon Postgres database in CI/CD, the command fails or hangs when using the PgBouncer pooled connection string. How should Prisma schema migrations be configured to avoid DDL issues over connection poolers? | pass→pass | 11,163 | 8,248 | -26% | 1 | 1 | 0% | 1,841 | 1,863 | +1% | 0 | 0 | — |
▸case-01 We are configuring a Prisma project connecting to Neon Postgres in a Next.js serverless application. We want to define environment variables in `.env` for database access. Standard Prisma templates usually specify a single `DATABASE_URL`. Should we use a single pooled `DATABASE_URL` for everything including migrations, or how should environment variables be structured? | pass→pass | 12,798 | 6,941 | -46% | 1 | 1 | 0% | 2,260 | 1,557 | -31% | 0 | 0 | — |
▸case-04 We are building a lightweight edge function in Cloudflare Workers using Drizzle ORM with Neon Postgres. The function runs a single `SELECT` query per invocation to look up user feature flags. Is it better to establish a WebSocket connection with `neon-serverless` or use `neon-http` for this function? | pass→pass | 12,111 | 7,359 | -39% | 1 | 1 | 0% | 1,880 | 1,508 | -20% | 0 | 0 | — |
▸case-05 We are implementing an order processing checkout endpoint in Drizzle ORM connected to Neon Postgres. The process runs multiple SQL queries inside an interactive `db.transaction()` block. Can we use `neon-http` for this endpoint or do we need `neon-serverless`? | pass→pass | 11,959 | 8,947 | -25% | 1 | 1 | 0% | 2,055 | 1,879 | -9% | 0 | 0 | — |
▸case-06 We are building a Node.js backend using Drizzle ORM on Neon Postgres that relies on persistent database session state and session-level settings. Which Neon Drizzle driver option is required? | pass→pass | 11,526 | 6,214 | -46% | 1 | 1 | 0% | 2,060 | 1,301 | -37% | 0 | 0 | — |
▸case-07 We are sizing our infrastructure for a high-concurrency event with Neon Postgres and its built-in PgBouncer pooler. How many concurrent client connections can the Neon PgBouncer connection pooler handle simultaneously? | pass→pass | 12,663 | 4,268 | -66% | 1 | 1 | 0% | 2,030 | 945 | -53% | 0 | 0 | — |
▸case-08 When monitoring backend connection utilization on Neon Postgres, our application database user seems unable to claim all underlying Postgres connections. How many connections are reserved exclusively for the Neon superuser? | fail→pass | 8,081 | 2,238 | -72% | 1 | 1 | 0% | 1,373 | 603 | -56% | 0 | 0 | — |
▸case-09 A developer claims that because PgBouncer in Neon allows up to 10,000 client connections, thousands of client connections will not consume any underlying Postgres server connections. Is this true or do pooled connections still consume backend Postgres connections? | pass→pass | 12,994 | 6,982 | -46% | 1 | 1 | 0% | 2,047 | 1,368 | -33% | 0 | 0 | — |
▸case-10 We are authoring a GitHub Actions workflow that executes SQL migration scripts against Neon Postgres using a migration CLI. Should the CI runner point to the pooled PgBouncer endpoint or the direct database endpoint? | pass→pass | 12,953 | 5,147 | -60% | 1 | 1 | 0% | 1,919 | 1,099 | -43% | 0 | 0 | — |
▸case-11 In a production Neon Postgres environment, we need to configure our ORM settings. Developers proposed using the direct endpoint for all runtime API traffic and pooled endpoint for DDL migrations. What is the correct mapping between endpoints and task types? | pass→pass | 9,910 | 7,376 | -26% | 1 | 1 | 0% | 1,558 | 1,549 | -1% | 0 | 0 | — |
▸case-12 We are refactoring a Vercel Edge Function that uses Drizzle ORM to fetch product details from Neon Postgres. The endpoint currently creates a WebSocket connection for a single read query. What optimization should be applied to minimize connection latency? | pass→pass | 9,210 | 5,472 | -41% | 1 | 1 | 0% | 1,497 | 1,219 | -19% | 0 | 0 | — |
▸case-13 Provide the datasource configuration block syntax for `schema.prisma` when setting up Neon Postgres with connection pooling in Prisma ORM. | pass→pass | 4,934 | 4,191 | -15% | 1 | 1 | 0% | 801 | 951 | +19% | 0 | 0 | — |
▸case-14 What connection pooling software powers Neon Postgres built-in pooled connection endpoints for serverless environments? | pass→pass | 5,853 | 1,756 | -70% | 1 | 1 | 0% | 830 | 527 | -37% | 0 | 0 | — |
▸case-15 An engineer wants to execute `db.transaction()` using `import { drizzle } from 'drizzle-orm/neon-http'` on Neon Postgres to guarantee atomicity across 3 write operations. Will this work or why will it fail? | pass→pass | 10,339 | 6,512 | -37% | 1 | 1 | 0% | 2,038 | 1,440 | -29% | 0 | 0 | — |
▸case-17 Neon connection strings can include pooled endpoint hostnames (e.g. containing `-pooler`). Which connection string should be supplied to Prisma Client at runtime versus Prisma Migrate during deployment? | pass→pass | 7,375 | 4,010 | -46% | 1 | 1 | 0% | 1,281 | 1,008 | -21% | 0 | 0 | — |
▸case-18 What driver import from `@neondatabase/serverless` or `drizzle-orm/neon-http` is recommended when building a stateless Cloudflare Worker with Drizzle ORM that performs single-query lookups against Neon Postgres? | pass→pass | 8,172 | 4,971 | -39% | 1 | 1 | 0% | 1,459 | 1,261 | -14% | 0 | 0 | — |
▸case-19 While running `npx prisma migrate dev` against Neon Postgres, the process fails with an advisory lock error because connection pooling prevents lock persistence. What exact change in Prisma connection configuration resolves this migration error? | pass→pass | 5,150 | 5,578 | +8% | 1 | 1 | 0% | 914 | 1,295 | +42% | 0 | 0 | — |
▸case-20 If application connection bursts saturate all available backend Postgres connection slots on Neon, will database administrators be locked out from running diagnostic SQL queries? | fail→pass | 14,350 | 4,547 | -68% | 1 | 1 | 0% | 2,143 | 948 | -56% | 0 | 0 | — |
▸case-21 We are designing a TypeScript microservices architecture on Neon Postgres using Drizzle ORM. Service A handles simple single-item API lookups in serverless Lambdas. Service B manages financial ledger checkouts requiring ACID multi-statement transactions. Select the appropriate Drizzle Neon driver for each service. | pass→pass | 10,601 | 7,577 | -29% | 1 | 1 | 0% | 1,727 | 1,715 | -1% | 0 | 0 | — |
▸case-22 In a deployment review for a serverless Node.js application using Prisma ORM with Neon Postgres, the team lead asks why both `DATABASE_URL` and `DIRECT_URL` are required in `.env` instead of defining only `DATABASE_URL`. | pass→pass | 10,601 | 10,422 | -2% | 1 | 1 | 0% | 1,788 | 2,156 | +21% | 0 | 0 | — |
▸case-23 We are configuring self-hosted PostgreSQL 16 on an Ubuntu EC2 instance with local PgBouncer in transaction pooling mode. We need to run `pg_restore` for database restores. Should `pg_restore` connect to PgBouncer or direct to PostgreSQL, and how many superuser reserved connections exist by default in standard PostgreSQL's `superuser_reserved_connections` parameter? | pass→pass | 6,933 | 6,870 | -1% | 1 | 1 | 0% | 1,226 | 1,427 | +16% | 0 | 0 | — |
▸case-24 We are setting up Supabase Database with Prisma ORM. Supabase provides Supavisor for connection pooling on port 6543 and a direct database connection on port 5432. What schema parameters are recommended in `schema.prisma` for Supabase pooled connections? | pass→pass | 11,590 | 6,980 | -40% | 1 | 1 | 0% | 1,985 | 1,635 | -18% | 0 | 0 | — |
▸case-25 We have a slow PostgreSQL query on standard PostgreSQL 16 running `SELECT * FROM orders WHERE status = 'pending' AND created_at > NOW() - INTERVAL '7 days' ORDER BY created_at DESC;`. How should we index this table in standard PostgreSQL to optimize query execution and verify performance? | pass→pass | 13,813 | 11,539 | -16% | 1 | 1 | 0% | 2,306 | 2,325 | +1% | 0 | 0 | — |