mirror of
https://github.com/netbirdio/netbird.git
synced 2026-09-12 17:59:06 +02:00
Nothing recorded which run of the connection was current, so three defects followed from the same gap. A ConnectClient is single-use, and the daemon builds a fresh one per outer-retry turn (server.go connect). Each turn overwrote s.connectClient and nothing stopped the one it replaced — the outgoing run loop had returned, which is what brought control back to the retry, but that was assumed rather than enforced, and any teardown its error path left half-done got no second chance. cleanupConnection read s.connectClient, cancelled, then stopped that engine. Nothing established the client it read was still current by the time it stopped it, so a teardown could target a client a newer turn had already replaced and leave the live one running untracked. Down's wait on clientGiveUpChan and Up's refusal to start a second loop kept the window narrow, but by arrangement rather than by construction. Third, the engine was stopped twice concurrently: actCancel woke the run loop, which stops the engine on its way out, while cleanupConnection stopped the same engine directly. The TODO there said ConnectClient.Stop was the right call and that its unbounded wait was what ruled it out. RunSupervisor records the generation of the current run. Publish refuses a client from a superseded run and stops the client it displaces, so no ConnectClient is dropped without being stopped. Stop invalidates whatever run is in flight, stops the published client and waits for the run to exit. ConnectClient.StopWithContext bounds that wait, which removes the TODO's obstacle: cleanupConnection now hands the run loop sole ownership of engine shutdown and passes Down's 5s budget down. Stop() keeps its signature and its unbounded wait, so callers outside this change are untouched. embed.Client.Stop had built the same bound by hand with a goroutine and a select purely to watch its caller's context; it passes the context down instead. clientGiveUpChan and connectClient are gone — the supervisor answers both. The MDM restart path drops its hand-rolled 10s channel wait for the same Stop, which additionally stops the client the previous run left behind. Its deliberate choice to leave clientRunning set is unchanged. Down now waits inside cleanupConnection, under s.mutex, where it previously waited after releasing it. That is what pins the client being stopped to the one current when the call started; the cost is that Down can hold the mutex for up to its 5s budget. Found while fixing the iOS wifi-to-cellular black-hole (#7329), which was the same class of defect in the mobile SDKs. No bug report backs the daemon findings — they are read off the code, and the narrow windows above may be why they have not been observed.
235 lines
5.8 KiB
Go
235 lines
5.8 KiB
Go
package server
|
|
|
|
import (
|
|
"context"
|
|
"fmt"
|
|
"net/netip"
|
|
"slices"
|
|
"sort"
|
|
"strings"
|
|
|
|
"google.golang.org/grpc/codes"
|
|
gstatus "google.golang.org/grpc/status"
|
|
|
|
"github.com/netbirdio/netbird/client/proto"
|
|
"github.com/netbirdio/netbird/route"
|
|
"github.com/netbirdio/netbird/shared/management/domain"
|
|
)
|
|
|
|
type selectRoute struct {
|
|
NetID route.NetID
|
|
Network netip.Prefix
|
|
Domains domain.List
|
|
Selected bool
|
|
extraNetworks []netip.Prefix
|
|
}
|
|
|
|
// ListNetworks returns a list of all available networks.
|
|
func (s *Server) ListNetworks(context.Context, *proto.ListNetworksRequest) (*proto.ListNetworksResponse, error) {
|
|
s.mutex.Lock()
|
|
defer s.mutex.Unlock()
|
|
|
|
if s.checkNetworksDisabled() {
|
|
return nil, gstatus.Errorf(codes.Unavailable, errNetworksDisabled)
|
|
}
|
|
|
|
if s.runs.Current() == nil {
|
|
return nil, fmt.Errorf("not connected")
|
|
}
|
|
|
|
engine := s.runs.Current().Engine()
|
|
if engine == nil {
|
|
return nil, fmt.Errorf("not connected")
|
|
}
|
|
|
|
routeMgr := engine.GetRouteManager()
|
|
if routeMgr == nil {
|
|
return nil, fmt.Errorf("no route manager")
|
|
}
|
|
|
|
routesMap := routeMgr.GetClientRoutesWithNetID()
|
|
routeSelector := routeMgr.GetRouteSelector()
|
|
|
|
v6ExitMerged := route.V6ExitMergeSet(routesMap)
|
|
|
|
var routes []*selectRoute
|
|
for id, rt := range routesMap {
|
|
if len(rt) == 0 {
|
|
continue
|
|
}
|
|
// Skip v6 exit nodes that are merged into their v4 counterpart.
|
|
if _, ok := v6ExitMerged[id]; ok {
|
|
continue
|
|
}
|
|
|
|
r := &selectRoute{
|
|
NetID: id,
|
|
Network: rt[0].Network,
|
|
Domains: rt[0].Domains,
|
|
Selected: routeSelector.IsSelected(id),
|
|
}
|
|
|
|
// Merge paired v6 exit node prefix into this entry.
|
|
v6ID := route.NetID(string(id) + route.V6ExitSuffix)
|
|
if _, ok := v6ExitMerged[v6ID]; ok && len(routesMap[v6ID]) > 0 {
|
|
r.extraNetworks = []netip.Prefix{routesMap[v6ID][0].Network}
|
|
}
|
|
|
|
routes = append(routes, r)
|
|
}
|
|
|
|
sort.Slice(routes, func(i, j int) bool {
|
|
iPrefix := routes[i].Network.Bits()
|
|
jPrefix := routes[j].Network.Bits()
|
|
|
|
if iPrefix == jPrefix {
|
|
iAddr := routes[i].Network.Addr()
|
|
jAddr := routes[j].Network.Addr()
|
|
if iAddr == jAddr {
|
|
return routes[i].NetID < routes[j].NetID
|
|
}
|
|
return iAddr.String() < jAddr.String()
|
|
}
|
|
return iPrefix < jPrefix
|
|
})
|
|
|
|
resolvedDomains := s.statusRecorder.GetResolvedDomainsStates()
|
|
var pbRoutes []*proto.Network
|
|
for _, route := range routes {
|
|
rangeStr := route.Network.String()
|
|
for _, extra := range route.extraNetworks {
|
|
rangeStr += ", " + extra.String()
|
|
}
|
|
pbRoute := &proto.Network{
|
|
ID: string(route.NetID),
|
|
Range: rangeStr,
|
|
Domains: route.Domains.ToSafeStringList(),
|
|
ResolvedIPs: map[string]*proto.IPList{},
|
|
Selected: route.Selected,
|
|
}
|
|
|
|
// Group resolved IPs by their parent domain
|
|
domainMap := map[domain.Domain][]string{}
|
|
|
|
for resolvedDomain, info := range resolvedDomains {
|
|
// Check if this resolved domain's parent is in our route's domains
|
|
if slices.Contains(route.Domains, info.ParentDomain) {
|
|
ips := make([]string, 0, len(info.Prefixes))
|
|
for _, prefix := range info.Prefixes {
|
|
ips = append(ips, prefix.Addr().String())
|
|
}
|
|
domainMap[resolvedDomain] = ips
|
|
}
|
|
}
|
|
|
|
// Convert to proto format
|
|
for domain, ips := range domainMap {
|
|
pbRoute.ResolvedIPs[domain.SafeString()] = &proto.IPList{
|
|
Ips: ips,
|
|
}
|
|
}
|
|
|
|
pbRoutes = append(pbRoutes, pbRoute)
|
|
}
|
|
|
|
return &proto.ListNetworksResponse{
|
|
Routes: pbRoutes,
|
|
}, nil
|
|
}
|
|
|
|
// SelectNetworks selects specific networks based on the client request.
|
|
func (s *Server) SelectNetworks(_ context.Context, req *proto.SelectNetworksRequest) (*proto.SelectNetworksResponse, error) {
|
|
s.mutex.Lock()
|
|
defer s.mutex.Unlock()
|
|
|
|
if s.checkNetworksDisabled() {
|
|
return nil, gstatus.Errorf(codes.Unavailable, errNetworksDisabled)
|
|
}
|
|
|
|
if s.runs.Current() == nil {
|
|
return nil, fmt.Errorf("not connected")
|
|
}
|
|
|
|
engine := s.runs.Current().Engine()
|
|
if engine == nil {
|
|
return nil, fmt.Errorf("not connected")
|
|
}
|
|
|
|
routeManager := engine.GetRouteManager()
|
|
if routeManager == nil {
|
|
return nil, fmt.Errorf("no route manager")
|
|
}
|
|
|
|
if req.GetAll() {
|
|
routeManager.SelectAllRoutes()
|
|
} else if err := routeManager.SelectRoutes(toNetIDs(req.GetNetworkIDs()), req.GetAppend()); err != nil {
|
|
return nil, err
|
|
}
|
|
|
|
s.statusRecorder.PublishEvent(
|
|
proto.SystemEvent_INFO,
|
|
proto.SystemEvent_SYSTEM,
|
|
"Network selection changed",
|
|
"",
|
|
map[string]string{
|
|
"networks": strings.Join(req.GetNetworkIDs(), ", "),
|
|
"append": fmt.Sprint(req.GetAppend()),
|
|
"all": fmt.Sprint(req.GetAll()),
|
|
},
|
|
)
|
|
|
|
return &proto.SelectNetworksResponse{}, nil
|
|
}
|
|
|
|
// DeselectNetworks deselects specific networks based on the client request.
|
|
func (s *Server) DeselectNetworks(_ context.Context, req *proto.SelectNetworksRequest) (*proto.SelectNetworksResponse, error) {
|
|
s.mutex.Lock()
|
|
defer s.mutex.Unlock()
|
|
|
|
if s.checkNetworksDisabled() {
|
|
return nil, gstatus.Errorf(codes.Unavailable, errNetworksDisabled)
|
|
}
|
|
|
|
if s.runs.Current() == nil {
|
|
return nil, fmt.Errorf("not connected")
|
|
}
|
|
|
|
engine := s.runs.Current().Engine()
|
|
if engine == nil {
|
|
return nil, fmt.Errorf("not connected")
|
|
}
|
|
|
|
routeManager := engine.GetRouteManager()
|
|
if routeManager == nil {
|
|
return nil, fmt.Errorf("no route manager")
|
|
}
|
|
|
|
if req.GetAll() {
|
|
routeManager.DeselectAllRoutes()
|
|
} else if err := routeManager.DeselectRoutes(toNetIDs(req.GetNetworkIDs())); err != nil {
|
|
return nil, err
|
|
}
|
|
|
|
s.statusRecorder.PublishEvent(
|
|
proto.SystemEvent_INFO,
|
|
proto.SystemEvent_SYSTEM,
|
|
"Network deselection changed",
|
|
"",
|
|
map[string]string{
|
|
"networks": strings.Join(req.GetNetworkIDs(), ", "),
|
|
"append": fmt.Sprint(req.GetAppend()),
|
|
"all": fmt.Sprint(req.GetAll()),
|
|
},
|
|
)
|
|
|
|
return &proto.SelectNetworksResponse{}, nil
|
|
}
|
|
|
|
func toNetIDs(routes []string) []route.NetID {
|
|
var netIDs []route.NetID
|
|
for _, rt := range routes {
|
|
netIDs = append(netIDs, route.NetID(rt))
|
|
}
|
|
return netIDs
|
|
}
|