End users getting Error 400 or Error 502 using IFS Cloud

Yes - we have seen this.
The root cause is that something (other than IFS) add cookies to the requests sent to IFS.
You can record a HAR file in you browser when the error happens. There you can see which cookies are large. Try to figure out which ones comes from non IFS origin (e.g loadbalancer cookies and network identity cookies)
IFS cloud in later SU’s later than 6months or so should be able to handle 16k of headers older version only handle 8k which is enough to handle the IFS cookies.
We have also seen that if the Remote infrastucture is old, then the patch to increase from 8k to 16k don’t get fully applied. 
 
Helm version on management server need to be > ~3.10

look in the ingress config:
kubectl get ingress ifs-sticky-ingress -o yaml -n <namespace>

you should find:
    nginx.ingress.kubernetes.io/server-snippet: |
        large_client_header_buffers 4 16k;