-
Bug
-
Resolution: Unresolved
-
Undefined
-
None
-
4.19
-
None
-
False
-
-
None
-
Important
-
None
-
None
-
None
-
Rejected
-
NI&D Sprint 285
-
1
-
None
-
None
-
None
-
None
-
None
-
None
-
None
Description of problem:
Describe your problem. Include specific actions and error messages.
HAProxy 2.6+ strictly does not allow HTTP methods GET, HEAD and DELETE with payload. See code here: https://github.com/haproxy/haproxy/blob/master/src/mux_h1.c#L2024
I can verify this on OpenShift with CURL:
printf '{"test":"x"}\n' > /tmp/payload.jsoncurl -vk --http1.0 \
-X GET \
-H 'Content-Type: application/json' \
--data-binary @/tmp/payload.json \
https://<route-redact>/fail
and ingress response:
> GET /blub/fail HTTP/1.0
> Host: <route-redact>
> User-Agent: curl/8.11.1
> Accept: */*
> Content-Type: application/json
> Content-Length: 13
>
* upload completely sent off: 13 bytes
* TLSv1.3 (IN), TLS handshake, Newsession Ticket (4):
* TLSv1.3 (IN), TLS handshake, Newsession Ticket (4):
< HTTP/1.1 413 Payload Too Large
< Content-length: 106
< Cache-Control: no-cache
< Content-Type: text/html
<
<html><body><h1>413 Payload Too Large</h1>
The request entity exceeds the maximum allowed.
</body></html>
My concern is that I am not seeing these HTTP 413 responses in ingress access logs.
IngressController:apiVersion: operator.openshift.io/v1
kind: IngressController
metadata:
labels:
app.kubernetes.io/instance: ingress-operator-oso
name: default
namespace: openshift-ingress-operator
spec:
clientTLS:
clientCA:
name: ""
clientCertificatePolicy: ""
defaultCertificate:
name: apps.<redact>
httpEmptyRequestsPolicy: Respond
httpErrorCodePages:
name: ""
idleConnectionTerminationPolicy: Deferred
logging:
access:
destination:
syslog:
address: <redact>
facility: local1
maxLength: 1024
port: 514
type: Syslog
logEmptyRequests: Log
nodePlacement:
nodeSelector:
matchLabels:
node-role.kubernetes.io/infra: ""
tolerations:
- effect: NoSchedule
key: infra
value: reserved
- effect: NoExecute
key: infra
value: reserved
routeSelector:
matchExpressions:
- key: type
operator: NotIn
values:
- sharded-s1
tuningOptions:
reloadInterval: 0s
unsupportedConfigOverrides: null
Why do I not see it in access logs?
Version-Release number of selected component (if applicable):
4.19
How reproducible:
every time, can replicate internally (details in subsequent updates)
Steps to Reproduce:
1.
2.
3.
Actual results:
413 not logged
Expected results:
413 logged
Additional info:
see case for more data/analytics: 04380751
