I have a woocommerce website hosted using AWS Lightsail on the bitnami stack. It's using apache and php-fpm. I use Cloudflare for DNS and some basic caching.
The site has been up and running on Lightsail for 9 months but for the past 3 weeks has been timing out frequently. There doesn't seem to be a particular pattern to when it times out but will time out roughly every 5 requests, it does seem to time out most often on product pages.
I've tried disabling all plugins and have tried switching the theme to a basic one and the issue remains.
To my knowledge nothing has changed to trigger this but can't be sure and not sure how to check.
I've tried to upgrade the spec of the lightsail instance(2 x cpu, 4GB memory). Have increased the memory allocated to php to 1024 Mb, have tried to increase the max number from 5 to 10 of child processes that php-fpm allows(I'm not convinced this change has been picked up properly and can't find a way to definitely check what it is set to.
I've enabled debug logging in wordpress and whilst there are some php warnings they don't seem to relate to the requests that timeout.
In the php-fpm log I do see the following error a few times a day however it doesn't seem to relate directly to timeouts.
WARNING: [pool www] server reached max_children setting (5), consider raising it
I've enabled the php status page and have tried to monitor this whilst triggering a timeout request and it doesn't look like there is a long running process at that moment.
I'm running out of ideas to diagnose and fix the root cause of this issue and would really appreciate any suggestions on what I can try.
Is there a way to log the connection timeouts on the server?
