@rsevilla could you help try to verified this bug on 4.7 version?

Hi,

I've done some tests with the patch and I can verify I haven't seen a significant performance drop when launching a perf test against a cluster with lot of NodePort services:

$ oc get clusterversion
NAME      VERSION                             AVAILABLE   PROGRESSING   SINCE   STATUS
version   4.7.0-0.nightly-2021-03-13-033953   True        False         140m    Cluster version is 4.7.0-0.nightly-2021-03-13-033953

# 201 NodePort services

$ oc get svc -A | grep -c NodePort
201

sh-4.4# mb -i conf.json -d 120
Time: 120.10s
Sent: 278.84MiB, 2.32MiB/s
Recv: 2.37GiB, 20.22MiB/s
Hits: 1872002, 15587.13/s
sh-4.4# mb -i conf.json -d 120
Time: 120.09s
Sent: 259.38MiB, 2.16MiB/s
Recv: 2.21GiB, 18.81MiB/s
Hits: 1741354, 14500.04/s


# 1 NodePort service

$ oc get svc -A | grep -c NodePort
1

sh-4.4# mb -i conf.json -d 120
Time: 120.04s
Sent: 279.15MiB, 2.33MiB/s
Recv: 2.37GiB, 20.25MiB/s
Hits: 1874053, 15611.81/s

sh-4.4# mb -i conf.json -d 120
Time: 120.10s
Sent: 286.46MiB, 2.39MiB/s
Recv: 2.44GiB, 20.77MiB/s
Hits: 1923150, 16013.16/s

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory (OpenShift Container Platform 4.7.3 bug fix update), and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2021:0821