week-1.txt

PERFORMANCE AND MONITORING:

Quiz:

1. How many MMS monitoring agents should you run for a production system? 
  - More then 1 for redundancy. 

2. What is typically the cause of a low ulimit warning in MMS monitoring? 
  - The resources available to mongod may be too limited.

3. Below are several groupings for stats you can view in MMS monitoring. Select those groups containing stats that are usually correlated with one another. 
  - opcounters, lock% | lock%, journal stats, flush average | non-mapped virtual memory, connections | cursors, connections

4. Extrapolating from this lesson, which of the following might cause different members of a replica set to have different opcounter values? Check all that apply. 
  - You are reading only from the primary. | You have read preference set to secondaryPreferred. | The primary will receive reads from each secondary as they read the oplog.

5. Which of the following will generate a host down alert? Check all that apply. 
  - A mongod process monitored by the agent is killed. | A secondary in a replica set goes down. | An arbiter in a replica set goes down.

6. How can you add users to a group? Check all that apply. 
  - You can send email invitations to users who are not in MMS | You can add existing MMS users to groups yourself | You can add anyone with MongoDB Jira credentials yourself. 

7. Consider the following output from mongostat (see mongostat_output.txt) - Which of the following statements is true? 
  - The data set is staying the same size.

8. If a disk was highly utilized and struggling to keep up, which CPU-level statistic would you expect to see increased? 
  - %iowait


================================================================================================================




H-w 1.2:
  
  > use m202
  > db.answers.insert({'homework':'homework_1.2', 'answer':11});

=========================================================================================================

H-w 1.4 answers:

  - The server is under heavy write load
  - Queries accessed data not in memory starting shortly before 12:30