-
Feature Request
-
Resolution: Unresolved
-
Normal
-
None
-
None
-
None
-
None
-
Product / Portfolio Work
-
None
-
False
-
-
None
-
None
-
None
-
-
None
-
None
-
None
-
None
-
None
1. Proposed title of this feature request
Support for haproxy.router.openshift.io/backend-config-snippet annotation in OpenShift Router (HAProxy) for custom backend directives
2. What is the nature and description of the request?
Request native support in OpenShift Router (HAProxy) for the annotation "haproxy.router.openshift.io/backend-config-snippet" on Route resources. This annotation would allow users to inject custom HAProxy configuration directives directly into the generated backend section for a Route. This functionality is based on the widely-used community standard haproxy-ingress.github.io/backend-config-snippet, but adapted for OpenShift’s router implementation. A corresponding pull request has already been submitted by customer's partner Steadybit to the OpenShift Router upstream repository:https://github.com/openshift/router/pull/657 Importantly, this behavior could be made configurable (e.g., via an environment variable or router configuration option), so that cluster administrators can choose to enable or disable support for this annotation based on their security or policy requirements.
3. Why does the customer need this? (List the business requirements here)
They have partnered with Steadybit to conduct chaos engineering at the ingress layer. Their approach relies on the injection of HAProxy directives via annotations to simulate scenarios such as: - Blocking traffic to a specific backend (e.g., service outage simulation) - Injecting latency to simulate degraded network conditions Without support for this annotation, they are unable to perform these tests on their OpenShift clusters, which: - Blocks their ability to validate the resilience of services at the ingress layer. - Forces to consider non-standard workarounds (e.g., custom ingress controllers or sidecars), which increase operational complexity and drift from supported Red Hat configurations. - Undermines reliability testing efforts, which are critical for ensuring their applications meet SLOs and SLAs in production environments. This feature is essential to their service reliability engineering (SRE) and QA teams to maintain high availability and performance expectations.
4. List any affected packages or components.
OCP router/IngressController