The problem
A hospitality website hosted on AWS Lightsail had become painfully slow and was intermittently going offline. Visitors were hitting timeout errors and even server access was becoming unreliable.
What made it difficult
The hosting dashboard suggested the server was barely using its CPU, so the obvious explanation didn’t fit the symptoms.
What I found
After working through the server, WordPress, database and traffic layers, I found that the instance had exhausted its available burst processing capacity. The server appeared lightly loaded in AWS, but internally it was spending most of its time waiting for CPU.
What I did
I cloned the site onto a larger Lightsail instance, verified WordPress, Apache, PHP and the database independently, tested the site before cutover, then moved the existing static IP across.
The result
The site went from intermittent timeouts and very high server load to normal response times, near-zero load and stable operation.
This is the kind of work I often end up doing for clients. The visible problem might be “the website is slow” or “the site keeps going down,” but the actual cause can sit anywhere across hosting, WordPress, DNS, security, caching or server configuration. My job is to work through the layers and fix the thing that is actually broken.
Read the full technical write-up: How I fixed intermittent Cloudflare 524 errors on an AWS Lightsail WordPress site, or browse more case notes.