mirror of
https://github.com/netbirdio/netbird.git
synced 2026-10-06 13:39:07 +02:00
[management,proxy] Agent network: per-account LLM gateway (policy, metering, multi-provider) (#6555)
* [agent-network] Shared proto, OpenAPI schema, and generated types * [agent-network] Management: store, manager, synthesizer, policy engine, provider catalog, HTTP/gRPC API Adds the account-scoped agent-network module: provider/policy/budget CRUD and store, the reverse-proxy service synthesizer, policy selection + limit enforcement, the provider catalog (incl. Vertex AI and AWS Bedrock entries), and the management HTTP + proxy gRPC surfaces. * [management] Fix agent-network proxy-peer fan-out on affected-peer recompute The affected-peers resolver loaded only persisted reverse-proxy services, but agent-network services are synthesized on demand and never persisted. As a result the embedded proxy peer was never folded into the affected set when a client's group changed, so the proxy received no network-map update for a newly authorised client and rejected its handshake until a full resync (restart). loadProxyServices now merges the synthesized agent-network services (injected via a registration hook to avoid an import cycle), so proxy peers learn newly authorised clients immediately. * [proxy] Reverse-proxy middleware framework, chain, and request plumbing The per-target middleware chain (slots, dispatcher, mutation gate, metadata merger), body capture, access-log terminal sink, and the proxy wiring that builds + runs chains for synthesized agent-network services. * [proxy] LLM parsers, pricing, and builtin middlewares (OpenAI, Anthropic, Vertex AI, AWS Bedrock) Request/response parsers and SSE/event-stream metering, the embedded pricing table, and the builtin middleware set: request parser, router, policy limit-check/record, cost meter, guardrail, identity inject, response parser. Includes the path-routed providers — Google Vertex AI (keyfile:: service-account OAuth minting) and AWS Bedrock (bearer auth, invoke/converse/streaming, optional /bedrock prefix) — plus the Models allowlist and unmeterable-publisher deny. * [proxy] IPv6 in-place apply and TCP accept-loop hardening on netstack listeners * [agent-network] End-to-end test suite, module docs, and deployment preset * [agent-network] Fix codespell typos and exclude false positives - labelgen word pool: vermillion -> vermilion, racoon -> raccoon. - codespell ignore list: add flate (Go compress/flate package), recordin (a test-local identifier), and unparseable (a valid alternative spelling used consistently across identifiers + a metadata-value constant). * [management] Set LastSeen on injected proxy peer in realstack test (MySQL strict-mode) The injected embedded proxy peer had a PeerStatus with a zero LastSeen, which serializes to '0000-00-00' and is rejected by MySQL in strict mode (SQLite tolerates it). Set LastSeen to a valid time so SaveAccount succeeds on both engines. * [agent-network] Remove e2e shell-script suite from this branch The end-to-end shell scripts under scripts/e2e/ are maintained in a separate testing suite and are not part of this change set. * [agent-network] Polish module docs: remove internal review scaffolding, fix links, verify diagrams Strip PR-review framing, commit references, absolute paths, and stale internal references from the agent-network module docs; fix broken relative links; verify all diagrams against the current architecture. Remove the internal AI-reviewer prompt file. * [management] Refine session expiration handling to support 3-state encoding for SSO deadlines * [agent-network] Relocate agentnetwork package to internals/modules Move management/server/agentnetwork (and its catalog/, labelgen/, types/ subpackages) to management/internals/modules/agentnetwork, alongside the reverse-proxy module, and rewrite all importers. Pure relocation: package names, the synthesizer + affectedpeers registration hook, and store access (shared store.Store) are unchanged, so no import cycle is introduced (affectedpeers still depends only on the agentnetwork/types leaf). * [agent-network] Co-locate HTTP handlers in the module (RegisterEndpoints) Move the agent-network HTTP handlers from server/http/handlers/agentnetwork into the module at internals/modules/agentnetwork/handlers (package handlers) and rename the entrypoint AddEndpoints -> RegisterEndpoints, matching the reverse-proxy module convention. Wiring in http/handler.go updated accordingly.
This commit is contained in:
Maycon Santos
@@ -0,0 +1,126 @@
|
||||
package server
|
||||
|
||||
import (
|
||||
"context"
|
||||
"testing"
|
||||
|
||||
"github.com/stretchr/testify/assert"
|
||||
"github.com/stretchr/testify/require"
|
||||
|
||||
"github.com/netbirdio/netbird/management/internals/modules/agentnetwork"
|
||||
agenttypes "github.com/netbirdio/netbird/management/internals/modules/agentnetwork/types"
|
||||
"github.com/netbirdio/netbird/management/server/permissions"
|
||||
"github.com/netbirdio/netbird/management/server/store"
|
||||
)
|
||||
|
||||
// TestAgentNetwork_BudgetRuleCRUD_RealManager is the GC-1 no-mock guard for the
|
||||
// account budget-rule manager surface: real DefaultAccountManager, real store,
|
||||
// real permissions. It exercises create/get/list/update/delete through the
|
||||
// permission-gated manager (not the store directly) and asserts the reused
|
||||
// PolicyLimits cap shape and targets survive each step.
|
||||
func TestAgentNetwork_BudgetRuleCRUD_RealManager(t *testing.T) {
|
||||
am, _, err := createManager(t)
|
||||
require.NoError(t, err, "createManager must succeed")
|
||||
ctx := context.Background()
|
||||
|
||||
const (
|
||||
accountID = "agent-net-budget-acct"
|
||||
adminUserID = "agent-net-budget-admin"
|
||||
)
|
||||
account := newAccountWithId(ctx, accountID, adminUserID, "agent-net.test", "", "", false)
|
||||
require.NoError(t, am.Store.SaveAccount(ctx, account), "SaveAccount must succeed")
|
||||
|
||||
mgr := agentnetwork.NewManager(am.Store, permissions.NewManager(am.Store), am, nil)
|
||||
|
||||
created, err := mgr.CreateBudgetRule(ctx, adminUserID, &agenttypes.AccountBudgetRule{
|
||||
AccountID: accountID,
|
||||
Name: "eng-monthly",
|
||||
Enabled: true,
|
||||
TargetGroups: []string{"grp-eng"},
|
||||
TargetUsers: []string{"user-alice"},
|
||||
Limits: agenttypes.PolicyLimits{
|
||||
TokenLimit: agenttypes.PolicyTokenLimit{Enabled: true, GroupCap: 100_000, UserCap: 10_000, WindowSeconds: 2_592_000},
|
||||
BudgetLimit: agenttypes.PolicyBudgetLimit{Enabled: true, GroupCapUsd: 500, WindowSeconds: 2_592_000},
|
||||
},
|
||||
})
|
||||
require.NoError(t, err, "CreateBudgetRule must succeed")
|
||||
require.NotEmpty(t, created.ID, "create must mint an ID")
|
||||
|
||||
got, err := mgr.GetBudgetRule(ctx, accountID, adminUserID, created.ID)
|
||||
require.NoError(t, err, "GetBudgetRule must succeed")
|
||||
assert.Equal(t, "eng-monthly", got.Name, "name round-trips through the manager")
|
||||
assert.Equal(t, []string{"grp-eng"}, got.TargetGroups, "target groups round-trip")
|
||||
assert.Equal(t, int64(100_000), got.Limits.TokenLimit.GroupCap, "token group cap round-trips")
|
||||
|
||||
list, err := mgr.GetAllBudgetRules(ctx, accountID, adminUserID)
|
||||
require.NoError(t, err, "GetAllBudgetRules must succeed")
|
||||
require.Len(t, list, 1, "exactly the one created rule must be listed")
|
||||
|
||||
created.Limits.TokenLimit.GroupCap = 200_000
|
||||
updated, err := mgr.UpdateBudgetRule(ctx, adminUserID, created)
|
||||
require.NoError(t, err, "UpdateBudgetRule must succeed")
|
||||
assert.Equal(t, int64(200_000), updated.Limits.TokenLimit.GroupCap, "updated cap must persist")
|
||||
|
||||
require.NoError(t, mgr.DeleteBudgetRule(ctx, accountID, adminUserID, created.ID), "DeleteBudgetRule must succeed")
|
||||
_, err = mgr.GetBudgetRule(ctx, accountID, adminUserID, created.ID)
|
||||
assert.Error(t, err, "get after delete must fail")
|
||||
}
|
||||
|
||||
// TestAgentNetwork_UpdateSettings_PreservesImmutableAndTogglesCollection is the
|
||||
// GC-1 guard for UpdateSettings: it must apply the collection toggles while
|
||||
// preserving the immutable Cluster/Subdomain pinned at bootstrap.
|
||||
func TestAgentNetwork_UpdateSettings_PreservesImmutableAndTogglesCollection(t *testing.T) {
|
||||
am, _, err := createManager(t)
|
||||
require.NoError(t, err, "createManager must succeed")
|
||||
ctx := context.Background()
|
||||
|
||||
const (
|
||||
accountID = "agent-net-settings-acct"
|
||||
adminUserID = "agent-net-settings-admin"
|
||||
clusterAddr = "eu.proxy.netbird.io"
|
||||
)
|
||||
account := newAccountWithId(ctx, accountID, adminUserID, "agent-net.test", "", "", false)
|
||||
require.NoError(t, am.Store.SaveAccount(ctx, account), "SaveAccount must succeed")
|
||||
|
||||
mgr := agentnetwork.NewManager(am.Store, permissions.NewManager(am.Store), am, nil)
|
||||
|
||||
// Creating a provider bootstraps the settings row (cluster + subdomain).
|
||||
_, err = mgr.CreateProvider(ctx, adminUserID, &agenttypes.Provider{
|
||||
AccountID: accountID,
|
||||
ProviderID: "openai_api",
|
||||
Name: "openai",
|
||||
UpstreamURL: "https://api.openai.com",
|
||||
APIKey: "sk-test",
|
||||
Enabled: true,
|
||||
Models: []agenttypes.ProviderModel{{ID: "gpt-5.4"}},
|
||||
}, clusterAddr)
|
||||
require.NoError(t, err, "CreateProvider must bootstrap settings")
|
||||
|
||||
before, err := mgr.GetSettings(ctx, accountID, adminUserID)
|
||||
require.NoError(t, err, "GetSettings must succeed after bootstrap")
|
||||
require.Equal(t, clusterAddr, before.Cluster, "cluster pinned at bootstrap")
|
||||
require.NotEmpty(t, before.Subdomain, "subdomain pinned at bootstrap")
|
||||
assert.False(t, before.EnablePromptCollection, "prompt collection defaults off")
|
||||
|
||||
// Attempt to flip toggles AND smuggle a different cluster/subdomain — the
|
||||
// immutable fields must be ignored.
|
||||
updated, err := mgr.UpdateSettings(ctx, adminUserID, &agenttypes.Settings{
|
||||
AccountID: accountID,
|
||||
Cluster: "attacker.cluster",
|
||||
Subdomain: "evil",
|
||||
EnableLogCollection: true,
|
||||
EnablePromptCollection: true,
|
||||
RedactPii: true,
|
||||
})
|
||||
require.NoError(t, err, "UpdateSettings must succeed")
|
||||
assert.Equal(t, before.Cluster, updated.Cluster, "cluster is immutable and must be preserved")
|
||||
assert.Equal(t, before.Subdomain, updated.Subdomain, "subdomain is immutable and must be preserved")
|
||||
assert.True(t, updated.EnableLogCollection, "log collection toggle must apply")
|
||||
assert.True(t, updated.EnablePromptCollection, "prompt collection toggle must apply")
|
||||
assert.True(t, updated.RedactPii, "redact toggle must apply")
|
||||
|
||||
reloaded, err := am.Store.GetAgentNetworkSettings(ctx, store.LockingStrengthNone, accountID)
|
||||
require.NoError(t, err)
|
||||
assert.Equal(t, before.Cluster, reloaded.Cluster, "persisted cluster unchanged")
|
||||
assert.True(t, reloaded.EnablePromptCollection, "persisted prompt collection toggled on")
|
||||
}
|
||||
Reference in New Issue
Block a user