Questions
6 of 19
How can we modify documents?
Difficulty: 5/10
update operators, upserts & atomicity, schema migration
Update operations modify the fields and values of a document while keeping other fields and values unchanged. Replace operations substitute all fields and values in an existing document with specified fields and values while keeping the _id field value unchanged.
The Node.js driver provides the following methods to change documents:
- 1
updateOne()
- 2
updateMany()
- 3
replaceOne()
The top level of an update document contains one or more of the following update operators:
- 1
$set: replaces the value of a field with a specified one
- 2
$inc: increments or decrements field values
- 3
$rename: renames fields
- 4
$unset: removes fields
- 5
$mul: multiplies a field value by a specified number
javascript
Scenario Questions
0-2 years experience
- 1We have a collection of user profiles, each document has a 'loginCount' field. How would you increment that field each time the user logs in?
- 2A bug requires you to rename the field 'addr' to 'address' in all existing documents. Which MongoDB update operator would you use and how?
- 3If you need to add a new field 'isActive' with default true to a single document identified by its _id, how would you write that update?
2-5 years experience
- 1Our service needs to add a new tag to an array field 'tags' for all products that match a certain category, but we must avoid duplicate tags. How would you construct the update?
- 2During a rollout we discovered that some orders are missing a 'status' field. We want to set it to 'pending' for those documents, but only if the field doesn't exist. How would you implement this in MongoDB, and what considerations are there for concurrency?
- 3We attempted an update with upsert:true but observed that duplicate documents were created under high load. Walk me through why that might happen and how you'd fix it.
5-8 years experience
- 1Our analytics pipeline needs to migrate a large collection from storing timestamps as strings to ISODate objects. Describe how you would perform this migration with minimal impact on read/write latency.
- 2We have a sharded cluster and need to run a bulk update that modifies a field used as a shard key. What are the risks and how would you approach this operation?
- 3Explain how you would use MongoDB's change streams to audit document modifications in a high‑throughput microservice, and what performance trade‑offs you would consider.
8+ years experience
- 1Our organization is moving from a monolithic MongoDB deployment to a multi‑region, polyglot persistence architecture. How would you design a strategy for evolving document schemas across services while ensuring backward compatibility?
- 2We need to deprecate a field used by several legacy services without causing downtime. Outline a phased migration plan, including how you’d handle versioning and data validation.
- 3Discuss the pros and cons of using server‑side scripts (e.g., $where) versus application‑level logic for complex document modifications at scale.
Follow-up Questions
- What could go wrong if you forget to include the upsert option in that scenario?
- How would you verify that your bulk migration didn't introduce data inconsistencies?
- Can you describe a way to make the update safe under high concurrent writes?