Improve Error Logging for Unhealthy PHP-FPM Containers #60
Description
Hello team,
First off, thank you for this useful script — I’ve been using it in a production ECS cluster, and it’s been quite helpful!
This is not about a bug but rather a feature enhancement request.
Recently, I encountered a failure in the php-fpm process due to max_children limits being reached. However, when the container became unhealthy and I manually ran the script, I noticed that:
No logs were captured regarding the failure.
The script exited with a status code of 0, even though there was a process-level issue.
I was only able to identify the problem by inspecting the php-fpm log files directly.
Enhancement Suggestion:
It would be great if the script could capture and persist error logs when something goes wrong (e.g., logging to a file or stderr). This would allow these logs to be piped into centralized monitoring tools like Datadog or CloudWatch, enabling better visibility and alerting.
Thanks again!