Files
netbird/management/internals/modules/agentnetwork/reconcile.go
Maycon Santos b416063bcc [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.
2026-06-27 13:41:00 +02:00

132 lines
4.2 KiB
Go

package agentnetwork
import (
"context"
log "github.com/sirupsen/logrus"
rpservice "github.com/netbirdio/netbird/management/internals/modules/reverseproxy/service"
"github.com/netbirdio/netbird/management/server/types"
"github.com/netbirdio/netbird/shared/management/proto"
)
// reconcile recomputes the synthesised reverse-proxy services for an
// account, diffs them against the previously-synthesised set in the
// in-memory cache, and emits Create / Update / Delete proxy mappings
// to the affected clusters. Also triggers a peer-side network-map
// recompute via accountManager.UpdateAccountPeers so the
// private-service ACL injection picks up the new state immediately.
//
// Reconcile failures are logged and swallowed — the underlying CRUD
// has already completed, and the next mutation (or proxy reconnect)
// will re-converge the cluster's view.
func (m *managerImpl) reconcile(ctx context.Context, accountID string) {
if accountID == "" {
return
}
defer func() {
if m.accountManager != nil {
m.accountManager.UpdateAccountPeers(ctx, accountID, types.UpdateReason{
Resource: types.UpdateResourceService,
Operation: types.UpdateOperationUpdate,
})
}
}()
if m.proxyController == nil {
return
}
services, err := SynthesizeServices(ctx, m.store, accountID)
if err != nil {
log.WithContext(ctx).WithError(err).Warnf("agent-network reconcile: synthesise services for account %s", accountID)
return
}
oidcCfg := m.proxyController.GetOIDCValidationConfig()
current := make(map[string]*proto.ProxyMapping, len(services))
for _, svc := range services {
if svc == nil || svc.ID == "" {
continue
}
current[svc.ID] = svc.ToProtoMapping(rpservice.Update, "", oidcCfg)
}
m.reconcileMu.Lock()
previous := m.reconcileCache[accountID]
if previous == nil {
previous = make(map[string]*proto.ProxyMapping)
}
creates, updates, deletes := diffMappings(previous, current)
if len(current) == 0 {
delete(m.reconcileCache, accountID)
} else {
m.reconcileCache[accountID] = current
}
m.reconcileMu.Unlock()
for _, mapping := range creates {
mapping.Type = proto.ProxyMappingUpdateType_UPDATE_TYPE_CREATED
m.proxyController.SendServiceUpdateToCluster(ctx, accountID, mapping, clusterFromMapping(mapping))
}
for _, mapping := range updates {
mapping.Type = proto.ProxyMappingUpdateType_UPDATE_TYPE_MODIFIED
m.proxyController.SendServiceUpdateToCluster(ctx, accountID, mapping, clusterFromMapping(mapping))
}
for _, mapping := range deletes {
mapping.Type = proto.ProxyMappingUpdateType_UPDATE_TYPE_REMOVED
m.proxyController.SendServiceUpdateToCluster(ctx, accountID, mapping, clusterFromMapping(mapping))
}
}
// diffMappings classifies the previous→current transition for a
// single account into Create / Update / Delete sets.
//
// Cluster moves (current.cluster != previous.cluster) are surfaced as
// a Delete on the old cluster + Create on the new — handled by
// emitting both a delete (on previous mapping) and a create (on the
// current mapping) for that service ID.
func diffMappings(previous, current map[string]*proto.ProxyMapping) (creates, updates, deletes []*proto.ProxyMapping) {
for id, cur := range current {
prev, existed := previous[id]
switch {
case !existed:
creates = append(creates, cur)
case prev.GetDomain() == "" || cur.GetAccountId() == prev.GetAccountId() && currentClusterChanged(prev, cur):
deletes = append(deletes, prev)
creates = append(creates, cur)
default:
updates = append(updates, cur)
}
}
for id, prev := range previous {
if _, stillThere := current[id]; !stillThere {
deletes = append(deletes, prev)
}
}
return creates, updates, deletes
}
func currentClusterChanged(prev, cur *proto.ProxyMapping) bool {
return clusterFromMapping(prev) != clusterFromMapping(cur)
}
// clusterFromMapping returns the cluster the mapping should be sent
// to. ProxyMapping doesn't carry the cluster directly, so we rely on
// the synthesised service's domain (`<slug>.<cluster>`) and split on
// the first '.'.
func clusterFromMapping(m *proto.ProxyMapping) string {
if m == nil {
return ""
}
domain := m.GetDomain()
for i := 0; i < len(domain); i++ {
if domain[i] == '.' {
return domain[i+1:]
}
}
return ""
}