Skip to content
learn.dm

Common Mistakes

skip() and Over-fetching

Deep offsets and SELECT * for documents. What it costs, and what to do instead.

IntermediateAbout 5 minutes

Problem. Page 500 of orders, or a list endpoint that ships every field including nested blobs the UI never shows.

db.orders.find({}).skip(4990).limit(10)
db.users.find({ city: "Bangalore" })  // then delete keys in Node
Bad — skip walks every prior key; find() pulls the whole document
db.orders.find({
  userId,
  $or: [
    { createdAt: { $lt: lastDate } },
    { createdAt: lastDate, _id: { $lt: lastId } }
  ]
}).sort({ createdAt: -1, _id: -1 }).limit(10)

db.users.find({ city: "Bangalore" }, { name: 1, city: 1, _id: 0 })
Better — cursor + projection on the server

skip(n) is O(n). Offset is fine for a 20-row admin table. toArray() on unbounded find is the Node version of the same class of bug. countDocuments on every page for '4 of 80,221' is a second heavy query — often drop the total.

Interview question

Why is skip() plus fetching whole documents a production bug?

Think about it first.

Practice

Related lessons