handleH2 error in apm log

I have an elasticsearch cluster in openshift 4.11 via the elastic operator 2.5.0.
apm server started as kind: ApmServer

```
apiVersion: apm.k8s.elastic.co/v1
kind: ApmServer
metadata:
name: test-apm
spec:
config:
apm-server.rum.enabled: true
max_procs: 16
apm-server.auth.anonymous.allow_agent:
- rum-js
- js-base
- java
- dotnet
- php
- opentelemetry/cpp
- python
apm-server.ssl.supported_protocols:
- TLSv1.0
- TLSv1.1
- TLSv1.2
- TLSv1.3
apm-server.rum.allow_headers:
- x-requested-with
apm-server.aggregation.transactions.enabled: true
apm-server.capture_personal_data: true
logging.level: warning
apm-server.auth.anonymous.enabled: true
apm-server.rum.allow_origins:
- '*'
apm-server.logging.level: warning
count: 4
elasticsearchRef:
name: elasticsearch
http:
service:
metadata:
annotations:
service.beta.openshift.io/inject-cabundle: 'true'
service.beta.openshift.io/serving-cert-secret-name: apmserver-openshift-tls
spec: {}
tls:
certificate:
secretName: apmserver-openshift-tls
selfSignedCertificate:
subjectAltNames:
- dns: apm-server.somedomain.com
kibanaRef:
name: kibana
podTemplate:
metadata:
creationTimestamp: null
spec:
containers:
- name: apm-server
resources:
limits:
cpu: '1'
memory: 2Gi
requests:
cpu: 100m
memory: 100Mi
version: 8.5.2
```

and I get a lot of errors in the logs:

{"log.level":"error","@timestamp":"2022-11-27T22:44:11.270Z","log.logger":"beater.http","log.origin":{"file.name":"gmux@v0.2.0/mux.go","file.line":81},"message":"handleH2 (172.22.12.6:51292) returned an error: write tcp 172.21.79.53:8200->172.22.12.6:51292: write: connection reset by peer","service.name":"apm-server","ecs.version":"1.6.0"}

pc
eck 2.5.0
elastic/kibana/apm - 8.5.2
openshift 4.11