Profiler for slow queries in MongoDB - Step-by-Step Execution

Concept Flow - Profiler for slow queries

Enable Profiler

↓

Profiler Logs Queries

↓

Check Query Execution Time

↓

Log Query

↓

Analyze Logs

The profiler is turned on, it watches queries, logs those that are slow, and then you analyze these logs to improve performance.

Execution Sample

MongoDB

db.setProfilingLevel(1, { slowms: 100 })
db.system.profile.find({ millis: { $gt: 100 } })

Enable profiling for queries slower than 100ms and then find those slow queries in the profiler collection.

Execution Table

Step	Action	Profiler State	Query Logged	Output
1	Enable profiler with slowms=100	Profiling ON, slowms=100	No	Profiler ready
2	Run query taking 50ms	Profiling ON	No	Query runs, not logged (50 < 100)
3	Run query taking 150ms	Profiling ON	Yes	Query logged (150 > 100)
4	Query profiler collection for slow queries	Profiling ON	N/A	Returns logged slow query details
5	Disable profiler	Profiling OFF	No	Profiling stopped

💡 Profiler stops logging when disabled or slowms threshold not met

Variable Tracker

Variable	Start	After Step 1	After Step 2	After Step 3	After Step 4	Final
profilingLevel	0 (off)	1 (on)	1 (on)	1 (on)	1 (on)	0 (off)
slowms	100 (default)	100	100	100	100	100
loggedQueriesCount	0	0	0	1	1	1

Key Moments - 2 Insights

Why doesn't every query get logged when the profiler is on?

Only queries slower than the slowms threshold are logged, as shown in execution_table step 2 and 3 where 50ms query is ignored but 150ms query is logged.

What happens if you disable the profiler?

Profiler stops logging any queries immediately, as shown in execution_table step 5 where profilingLevel changes to 0 and no queries are logged.

Visual Quiz - 3 Questions

Look at the execution_table, at which step is the first slow query logged?

A Step 2

B Step 3

C Step 4

D Step 5

💡 Hint

Check the 'Query Logged' column in execution_table rows

According to variable_tracker, what is the profilingLevel after enabling the profiler?

A 0 (off)

B 2 (debug)

C 1 (on)

D 100

💡 Hint

Look at profilingLevel value after Step 1 in variable_tracker

If slowms was set to 200, what would happen to the query taking 150ms in step 3?

A It would not be logged

B Profiler would disable

C It would be logged

D Query would run faster

💡 Hint

Compare query time with slowms threshold in execution_table step 3

Concept Snapshot

Profiler for slow queries in MongoDB:
- Enable with db.setProfilingLevel(level, {slowms: ms})
- Logs queries slower than slowms
- Check logs in system.profile collection
- Disable with db.setProfilingLevel(0)
- Helps find slow queries to optimize

Full Transcript

This visual execution shows how MongoDB's profiler works for slow queries. First, you enable the profiler with a threshold (slowms) that sets the minimum query time to log. Queries faster than this threshold are ignored. When a query runs slower than slowms, it is logged in the system.profile collection. You can then query this collection to see details of slow queries. Finally, you can disable the profiler to stop logging. Variables like profilingLevel and slowms control this behavior, and the loggedQueriesCount tracks how many slow queries have been recorded.