F5 Loadbalancer log Server sends too much data

Our loadbalancer administrator complains that his logfiles are flooded with nextcloud related errors :
http_process_state_prepend - Invalid action:0x109010 Server sends too much data. serverside (xx.xx.xx.xx:443 -> xx.xx.xx.xx:61889) clientside ((null connflow)) (Server side : vip=/xxxx profile=http server_ip=xx.xx.xx.xx)

So it looks like nextcloud sends a header saying I will transmit x-amount of bytes which is not honored and then leads to an error. I do not find any problem in nextcloud logs/apache logs.

Nextcloud version 13.0.4
Operating system Centos 7.4.1708
Apache 2.4.6
PHP 7.1.18

Any idea how to get rid of the errors in the loadbalancer?
We do not have users complaining that nextcloud is not functioning well and on the nextcloud server I do not find any errors/problems but for some reason the loadbalancers logs a lot of those http_process_state_prepend
Server sends too much data errors.

-> Found out that it is related to fixed bug:

-> upgraded our nextcloud server to 15.0.7 but our loadbalancer still throws the error in his logs.
According to F5 support problems caused by nextcloud server which responds with body in 304 http response.
fix
BaseResponse.php
if ($status === Http::STATUS_NO_CONTENT ||
$status === Http::STATUS_NOT_MODIFIED ||
($status >= 100 && $status <= 199)) {
// Those status codes are not supposed to have a body:
// https://stackoverflow.com/q/8628725
return ‘’;

replace by
if ($status === Http::STATUS_NO_CONTENT ||
$status = 204 ||
$status = 304 ||
$status === Http::STATUS_NOT_MODIFIED ||

                    ($status >= 100 && $status <= 199)) {
                    // Those status codes are not supposed to have a body:
                    // https://stackoverflow.com/q/8628725
                    return '';