Significant drop in reported throughput after upgrading from 7.15 through to 8.1.2

[Alex_Leonov]

Kibana version:
8.1.2

Elasticsearch version:
8.1.2

APM Server version:
Fleet/Agent/APM Integration 8.1.2

APM Agent language and version:
elastic-apm-node, 3.31.0

Browser version:
N/A

Fresh install or upgraded from other version?
Upgraded 7.15 to 7.17, then to 8.0 and then 8.1.2

Is there anything special in your setup?
Running on Elastic Stack in a self-managed K8s cluster, Fleet Agent exposed through an Nginx ingress

Description of the problem including expected versus actual behavior. Please include screenshots (if relevant):
After upgrading, throughput (for the server agent only!) has dropped by a hundred times. Nothing has changed in the application's code, and it is actually receiving the same amount of requests (checked that in the Google Cloud dashboard for the service). It is as if only 1% of transactions get reported or displayed.

Steps to reproduce:

1. Everything is working in 7.15, high throughput
2. Upgrade
3. Everything is working in 8.1.2, but the amount of transactions has dropped to the floor

[simitt (Silvia Mitter)]

Hi @Alex_Leonov ,
thanks for reporting and apologies for the late response. We are not aware of a known issue reducing throughput in 8.1.2 .
Could you take a look at the APM agent + APM Server logs and enable Stack Monitoring to see if anything suspicious is showing up please (any dropped events, server timeouts, etc.)?