We have installed and released pega deployment ver 8.8.3 on EKS.

While we operate/use application, 502 bad gateway often happens on the application.

What we have investigated:

We have checked the logs of AWS ALB and AWS Support said about below. Can you tell us how to resolve on pega platform?

• Upon reviewing the internal logs of the Application Load Balancer (ALB), we found numerous records indicating receipt of TCP FIN from the targets while there were outstanding requests on the ALB. Additionally, we verified that these records matched trace_ids from access logs that had been responded with HTTP 502. In such cases, the ALB responds with HTTP 502 to the client.
• For future investigation, it's necessary to examine the detailed situation on the target side. Unfortunately, we regret to inform you that we are unable to conduct this investigation on our end. We kindly ask you to check from your side using OS logs, application logs, or similar resources.
• If you are using Apache HTTP Server on the target web server and have set the Multi Processing Module (MPM) to the event mode, this configuration may be the cause of the HTTP 502 responses from the ALB.