Table of Contents
Question
A systems administrator is troubleshooting an application that is configured to auto-scale with a minimum of two nodes and a maximum of four. The application will scale out if the CPU utilization of one of the nodes exceeds 80% for more than five minutes and will scale in if the CPU utilization of one of the nodes drops under 20% for more than ten minutes. There is a reverse proxy in front of the application. The systems administrator notices two of the nodes are often running over 80% for a long period of time, which is triggering the creation of the other two nodes; however, they are being created and terminated while the load in the first two remains over 50% all the time. Which of the following should the administrator configure to fix this issue?
A. Disable DNS caching in the reverse proxy.
B. Reduce the minimum node count to one.
C. Disable TLS tickets.
D. Reduce the scale-out rule to 50%.
E. Increase the scale-in rule to 50%.
Answer
A. Disable DNS caching in the reverse proxy.
Explanation
The correct answer is A. Disable DNS caching in the reverse proxy.
A reverse proxy is a server that sits between the client and the application servers and acts as an intermediary for requests and responses. A reverse proxy can provide benefits such as load balancing, caching, security, and compression.
However, a reverse proxy can also cause issues with auto-scaling applications if it is not configured properly. One of the possible issues is DNS caching, which is the process of storing DNS records locally for faster resolution. DNS caching can improve performance, but it can also prevent the reverse proxy from discovering new nodes that are created by the auto-scaling mechanism. This can result in uneven distribution of traffic among the nodes, causing some nodes to be overloaded and others to be underutilized.
To fix this issue, the systems administrator should disable DNS caching in the reverse proxy, or reduce the TTL (time to live) value of the cached records to a very low number. This will allow the reverse proxy to resolve the DNS names of the nodes more frequently and balance the traffic more evenly among them.
The other options are incorrect because:
- B. Reduce the minimum node count to one. This is false because this will not solve the issue of uneven traffic distribution among the nodes. Reducing the minimum node count to one will only reduce the number of nodes that are always running, but it will not prevent the reverse proxy from sending most of the traffic to one or two nodes due to DNS caching.
- C. Disable TLS tickets. This is false because this has nothing to do with the issue of uneven traffic distribution among the nodes. TLS tickets are a mechanism that allows clients and servers to resume a previous TLS session without performing a full handshake. Disabling TLS tickets may affect the performance and security of the TLS connections, but it will not affect how the reverse proxy resolves the DNS names of the nodes.
- D. Reduce the scale-out rule to 50%. This is false because this will not solve the issue of uneven traffic distribution among the nodes. Reducing the scale-out rule to 50% will only make the auto-scaling mechanism more sensitive and create more nodes when one of them reaches 50% CPU utilization. However, this will not prevent the reverse proxy from sending most of the traffic to one or two nodes due to DNS caching.
- E. Increase the scale-in rule to 50%. This is false because this will not solve the issue of uneven traffic distribution among the nodes. Increasing the scale-in rule to 50% will only make the auto-scaling mechanism less sensitive and terminate fewer nodes when one of them drops below 50% CPU utilization. However, this will not prevent the reverse proxy from sending most of the traffic to one or two nodes due to DNS caching.
CompTIA Cloud+ CV0-003 certification exam practice question and answer (Q&A) dump with detail explanation and reference available free, helpful to pass the CompTIA Cloud+ CV0-003 exam and earn CompTIA Cloud+ CV0-003 certification.