Monitoring and Troubleshooting Flashcards
Your application is using SQS to send and receive messages, your application needs to receive the messages as soon as they arrive and you need to ensure the architecture is as cost efficient as possible. Which of the following approaches will optimise the cost and performance of the application?
- Lower the message Visibility Timeout
- Enable Long Polling
- Reduce the total number of message queues
- Enable Short Polling
• Enable Long Polling
Your company is reaching the end of the financial year and the Finance team are running a lot of large database queries and scans against your DynamoDB tables. The database queries and scans are taking much longer to complete than expected, how can you make them more efficient?
- Reduce the page size to return fewer items per results page
- Set your queries to be eventually consistent
- Filter your results based on the Primary Key and Sort Key
- Run parallel scans
- Run parallel scans
* Reduce the page size to return fewer items per results page
Which service can you use to analyze and debug distributed applications, identify issues and locate performance bottlenecks?
CloudWatch
Systems Manager
CloudTrail
X-Ray
X-Ray
You are working on an application for an online training company which stores product data in DynamoDB. This week, the company is running a big promotion on a few courses and this is bringing lots of new traffic to your website, causing an increased number of queries to the database. Database queries are now running much slower than usual and the Operations Team are concerned that the DynamoDB table is being throttled. Which of the following approaches would you recommend to improve read performance?
- Redesign your table to use a more distinct partition key to enable the I/O load to be more evenly distributed across partitions
- Configure the application to use scans rather than queries and run multiple scans in parallel
- Configure a DAX cluster and point the DynamoDB API calls at the DAX cluster
- Add a Read Replica and point the DynamoDB API calls at the Read Replica
Configure a DAX cluster and point the DynamoDB API calls at the DAX cluster
What is the maximum execution duration for a Lambda request?
60 seconds 900 seconds 3 minutes 300 seconds 500 seconds
900 seconds
You deployed a new Lambda function a few days ago and your code seems to be executing successfully, however when you check CloudWatch there isn’t any log data for your function. What could be the reason for this?
- The CloudWatch agent has stopped
- There is an issue with S3 in your region
- The execution role for the Lambda function did not grant permissions to write log data to CloudWatch Logs
- Your code is taking too long to execute, it could be that your function does not have enough compute resources to generate the log files
The execution role for the Lambda function did not grant permissions to write log data to CloudWatch Logs
A recent increase in the amount of users of an application hosted on an EC2 instance that you manage has caused the instances OS to run out of CPU resources and crash. The crash caused several users’ unsaved data to be lost and your supervisor wants to know how this problem can be avoided in the future. Which of the following would you NOT recommend?
- Rewrite the application so that users’ unsaved data is frequently written to disk.
- Take a snapshot of the EBS volume and re-deploy as a larger instance type.
- Take frequent snapshots of the EBS volume during business hours to ensure users’ data is backed up.
- Create an auto-scaling group to add more servers when demand is high.
Take frequent snapshots of the EBS volume during business hours to ensure users’ data is backed up.
You are hosting your website in an S3 bucket located in us-east-1, however many of your users are located in India, Africa and Europe and they are experiencing long delays. How can you improve response times for these users?
- Implement Connect Direct
- Configure a CloudFront CDN
- Use Elasticache to cache the content in each Region
- Create 3 additional S3 buckets in regions local to your users and replicate the data across to the new buckets
Configure a CloudFront CDN
As you retrieve information from DynamoDB, you receive a ProvisionedThroughputExceededException error. Further investigation shows that you’re not exceeding your table’s read capacity throughput. What is causing this error?
• AWS metrics do not run in real time. This error will disappear.
• You have too many sort keys on your table.
• The table is warming up and this process consumes throughput.
• You are exceeding an individual partition’s throughput capacity, even if you’re not exceeding the overall table throughput capacity.
You are exceeding an individual partition’s throughput capacity, even if you’re not exceeding the overall table throughput capacity.
Your application is using SQS to send and receive messages. On average, it takes your application between 20 and 40 seconds to process a message and you have noticed that quite frequently, multiple application servers are attempting to process the same message which is causing issues within the application. What can you do to help prevent this from happening?,
- Lower the message visibility timeout
- Enable Long Polling
- Enable Short Polling
- Increase the message visibility timeout
Increase the message visibility timeout
Your application servers are behind an Application Load Balancer with sticky sessions configured. However during busy times you are occasionally finding that one of your application servers is becoming overloaded. Which of the following options could help avoid this from happening?
- Store session state locally on the EC2 instance
- Store session state in RDS
- Store session state in memory
- Store session state in an ElastiCache cluster
Store session state in an ElastiCache cluster
You can use X-Ray with applications running on which platforms? • EC2 • ECS • Lambda • Elastic Beanstalk • S3
- EC2
- ECS
- Lambda
- Elastic Beanstalk
