Amazon Linux EC2 Troubleshooting -- 2

My Amazon Linux EC2 instance is showing signs of stress—occasional service crashes and noticeable load spikes. The stack is Nginx in front of PHP-FPM, and everything sits on a vanilla AWS setup with the usual security groups and EBS volumes. I need a fresh set of eyes to jump in over SSH, review the configuration, and stabilise the box.

Scope of work
• Pinpoint the root cause of the crashes and high-load episodes, using CloudWatch metrics, log files, and any profiling tools you prefer.
• Audit Nginx, PHP-FPM, and systemd settings, making sensible tweaks to worker limits, memory, and time-outs.
• Check kernel limits, swap usage, and basic Amazon Linux optimisations to be sure nothing obvious is throttling performance.
• Verify that core AWS items—security groups, IAM roles, EBS throughput, and instance type—are aligned with best practice.
• Leave me with clear notes on what you changed (or didn’t), so the next deployment is painless.

Acceptance criteria
1. Nginx and related services run for 24 h without an unexpected restart.
2. Average CPU and memory stay within a normal operating window (we can define this on kick-off).
3. I receive a concise change log and any tuned config files.

If you are comfortable in AWS CLI, CloudWatch, and basic DevOps housekeeping, this should be a quick win. SSH access and CloudWatch dashboards will be provided as soon as we agree to proceed.