MongoDB C100DBA Exam Dumps

The oplog (operations log) in a MongoDB replica set is stored in the following collection:

local.oplog.rs

• local: This is a special database that exists on every MongoDB instance and is not replicated.
• oplog.rs: The specific collection that contains the oplog.
• The oplog is a capped collection, meaning it has a fixed size and automatically overwrites old entries as new ones are added.
• All replica set members maintain their own copy of the oplog to track changes for replication.

The correct answer is to enable MongoDB's audit log system. The audit log records authentication attempts, authorization failures, and optionally all CRUD operations and administrative actions. This is the only MongoDB-native feature designed specifically for compliance auditing and provides verifiable evidence of who accessed what data and when. Audit logs can be written to syslog, JSON, or BSON format for integration with SIEM systems.

The second option mistakes the oplog for an audit trail. The oplog records replication operations needed to keep secondaries in sync, not access patterns or authentication. The third option relies on application-level logging, which is error-prone, can be tampered with, and is not the authoritative source MongoDB uses. Query profiling (fourth option) focuses on performance optimization of slow queries, not comprehensive access auditing. The fifth option references a non-existent feature. The audit log is a critical component of MongoDB's application administration and security posture tested on the C100DBA exam.

To extract data from MongoDB into a CSV file, you would use mongoexport.

mongoexport is a command-line utility that exports data from MongoDB to various formats:

• CSV format - mongoexport --collection myCollection --out output.csv --csv
• JSON format - Default output format
• Flexible field selection - Can specify which fields to export using --fields option
• Query filtering - Can export only documents matching a specific query using --query

Example: mongoexport --db myDatabase --collection myCollection --csv --fields _id,name,email --out data.csv

This tool is useful for data analysis in spreadsheet applications, data migration, and creating backups in a human-readable format.

Optimizing a slow $lookup stage involves several complementary strategies: (1) creating an index on the join field in the foreign collection allows MongoDB's query engine to use the index rather than performing a collection scan; (2) positioning $match before $lookup reduces the working set passed to the join, decreasing the number of join iterations required; (3) using the sub-pipeline form of $lookup (available in MongoDB 3.6+) allows you to filter, project, and limit the foreign collection results before the join, further reducing data volume. These three optimizations work together and are all applicable in modern MongoDB versions.

Option A is correct because an index on the foreign collection's join field is a primary optimization that enables index-assisted execution. Option B is correct because moving $match earlier in the pipeline is a standard aggregation optimization principle. Option C is correct because the sub-pipeline option allows you to filter the foreign collection before joining, reducing data volume. Option D is incorrect because application-layer joining typically requires fetching more data to the client and lacks MongoDB's optimizations; server-side $lookup is generally more efficient. Option E is incorrect because the shard key status of the foreign collection does not directly affect $lookup efficiency; $lookup works across sharded and unsharded collections, though there may be other cluster topology considerations.

During an election in a MongoDB replication cluster, all nodes with a priority greater than 0 participate in the election process. Elections are triggered when the primary becomes unavailable or steps down. Nodes with priority 0 cannot become primary but can still vote in elections. During an election, nodes exchange heartbeat messages and votes to determine which eligible node should become the new primary. The node that receives votes from a majority of the replica set members becomes the new primary. Factors considered during election include: node priority, data freshness (oplog position), and connectivity. It's important to configure priorities appropriately based on your infrastructure and requirements (e.g., higher priority for nodes in the primary data center).