Troubleshooting High Queue Size and Reducing it
Question
Your application uses Amazon SQS and Auto Scaling to process background jobs.
The Auto Scaling policy is based on the number of messages in the queue, with a maximum instance count of 100
Since the application was launched, the group has never scaled above 50
The Auto scaling group has now scaled to 100, the queue size is increasing and very few jobs are being completed.
The number of messages being sent to the queue is at normal levels.
What should you do to identity why the queue size is unusually high and to reduce it?
Answers
Explanations
Click on the arrows to vote for the correct answer
A. B. C. D.Answer : B.
Here the best option is to look at the application logs and resolve the failure.
You could be having a functionality issue in the application that is causing the messages to queue up and increase the fleet of instances in the Autoscaling group.
For more information on centralized logging system implementation in AWS, please visit this link :
https://aws.amazon.com/answers/logging/centralized-logging/The issue described in the question is that the Auto Scaling group has scaled to its maximum instance count of 100, but the queue size is increasing and very few jobs are being completed. The number of messages being sent to the queue is at normal levels. The goal is to identify why the queue size is unusually high and reduce it.
Answer B is the most appropriate solution to identify the root cause of the problem. Analyzing application logs can help identify the possible reasons for message processing failure. This will help identify issues such as stuck messages or messages with incomplete information, which could cause the queue size to increase. Once the cause of the failure has been identified, steps can be taken to resolve it.
Answer A is not a good solution because it would lead to over-provisioning of resources, which can be costly. Additionally, the scaling group has never scaled above 50, so it is unlikely that scaling to 200 would help.
Answer C is not a good solution either because creating additional scaling groups would not necessarily help reduce the queue size. Instead, it would increase the number of instances processing messages in parallel, which could increase costs without addressing the underlying issue causing the queue size to increase.
Answer D is not a good solution because CloudTrail logs for Amazon SQS would not help identify the root cause of the problem, which is likely related to message processing failure or incomplete information.
In conclusion, the best solution is to analyze the application logs to identify possible reasons for message processing failure and resolve the cause for failures.
