NE-2488: Add OpenShift router tools to NetworkEdge toolset

[alebedev87]

This PR introduces new tools for inspecting OpenShift router pods through the Kubernetes MCP Server.

Added tools:

• get_router_config: View the router's configuration
• get_router_info: Get router runtime information and statistics
• get_router_sessions: View all active sessions in the router

Documentation: Added NETEDGE.md covering the NetworkEdge toolset, including both router and CoreDNS tools.

Tests: Added evaluation tasks for the added tools.

[openshift-ci]

Skipping CI for Draft Pull Request.
If you want CI signal for your change, please convert it to an actual PR.
You can still manually trigger a test run with /test all

[swghosh]

#51 (comment)

Should we consider having this as a part of the "openshift" toolgroup?

[alebedev87]

Should we consider having this as a part of the "openshift" toolgroup?

@swghosh : That crossed my mind too. In my case "router" doesn't make much sense in Kubernetes context. However I didn't want to complicate things at this stage either.

[bentito]

Wherever this ends up we'll likely end up using it as part of our NIDS MCP tooling. https://issues.redhat.com/browse/NE-2278 inspect_route likely can just call one of these or be replaced by one or more of these. Also we might want to support all of these such that we have richer router data. cc @rikatz @Thealisyed

[matzew]

can just call one of these or be replaced by one or more of these

@bentito @alebedev87 any updates on this? or still on discussion.

[bentito]

@bentito @alebedev87 any updates on this? or still on discussion.
@alebedev87 When you can, please rebase and move to the ./pkg/toolsets/netedge dir

[openshift-ci]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: alebedev87
Once this PR has been reviewed and has the lgtm label, please assign cali0707 for approval. For more information see the Code Review Process.

The full list of commands accepted by this bot can be found here.

Details

Needs approval from an approver in each of these files:

• OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[openshift-ci-robot]

@alebedev87: This pull request references NE-2488 which is a valid jira issue.

Details

In response to this:

This PR introduces new tools for inspecting OpenShift router pods through the Kubernetes MCP Server.

Added tools:

• get_router_config: View the router's configuration
• get_router_info: Get router runtime information and statistics
• get_router_sessions: View all active sessions in the router

Documentation: Added NETEDGE.md covering the NetworkEdge toolset, including both router and CoreDNS tools.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the openshift-eng/jira-lifecycle-plugin repository.

[openshift-ci-robot]

@alebedev87: This pull request references NE-2488 which is a valid jira issue.

Details

In response to this:

This PR introduces new tools for inspecting OpenShift router pods through the Kubernetes MCP Server.

Added tools:

• get_router_config: View the router's configuration
• get_router_info: Get router runtime information and statistics
• get_router_sessions: View all active sessions in the router

Documentation: Added NETEDGE.md covering the NetworkEdge toolset, including both router and CoreDNS tools.

Tests: Added evaluation tasks for the added tools.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the openshift-eng/jira-lifecycle-plugin repository.

[openshift-ci]

@alebedev87: all tests passed!

Full PR test history. Your PR dashboard.

Details

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. I understand the commands that are listed here.

[alebedev87]

@bentito : I used false here before but then I copied true as in get_coredns_config tool. I'm not quite sure whether it's a good decision for router configs/sessions/info. Can you please advice?

[bentito]

I've reviewed this PR in the context of the NIDS MCP strategy and found a few critical areas for improvement, particularly regarding offline analysis.

Review Summary: Router Tools vs Offline Strategy

Context:

• Comparing against PR #115 (CoreDNS) and PR #131 (Prometheus).
• Goal: Align with the Offline Data Access Strategy to support must-gather workflows.

1. Critical Refactoring Required: Offline Compatibility

The current implementation of get_router_config relies on PodsExec (effectively oc exec), which will not work with offline must-gather archives.
Refactoring Recommendation:
Adopt the Client Abstraction Pattern demonstrated in ovn-kubernetes-mcp.

• Abstract the Data Source: Use a helper that switches behavior based on the context (live cluster vs. file dump).
• Implementation for get_router_config:
  • Live Mode: Continue using exec cat /var/lib/haproxy/conf/haproxy.config.
  • Offline Mode: Read the file directly from the filesystem. I confirmed that standard must-gather archives do contain this file at paths like .../ingress_controllers/default/router-*/haproxy.config.

2. Scope Clarification: "Live Only" Tools

The tools get_router_info and get_router_sessions rely on active HAProxy socket commands (show info, show sess).

• Finding: These outputs are not present in standard must-gather dumps.
• Action: These must be explicitly documented and marked as Live Only to set correct expectations for the agent and users, as they will fail in offline triage scenarios.

3. Consistency: Client Usage

PR #98 uses kubernetes.NewCore(params).PodsExec. For consistency with PR #115, request that the author switch to the controller-runtime client pattern where applicable (e.g., for listing pods), as it simplifies testing and mocking.